The effect of ageism on law enforcement response to elder mistreatment

This study focused on how law enforcement officers' incorrect knowledge and ageist beliefs about older adults affected their response to scenarios of elder mistreatment. Regression analyses was conducted using data from a survey of 200 law enforcement officers from the intermountain west area. The Utah Commission on Aging created the survey in response to low usage of Utah's Vulnerable Adult Statute. The survey incorporated the Palmore's Facts on Aging Quiz 1 (FAQ1) and two scenarios of elder mistreatment. This analysis assumed that accurate knowledge of aging and of older adults, as measured by the FAQ1, was an adequate proxy for ageism. The factors of gender, years in law enforcement, education, and rank were included as controls. The examination found that officers' correct knowledge as measured by their FAQ1 Knowledge scores was positively correlated with the scenario response scores (an index reflecting the number of protective measures respondents described). The regression furthermore indicated a negative correlation to the scenarios when officers held incorrect, but positive views of older persons. Training specific to the Utah Vulnerable Adult Statute was also examined to determine if such training would increase appropriate responses by law enforcement officers to the scenarios. It was found that statute specific training did increase officers' use of the Vulnerable Adult Statute, but did not increase the overall effective response of officers toward vulnerable older persons.

THE EFFECT OF AGEISM ON LAW ENFORCEMENT RESPONSE TO ELDER MISTREATMENT by Richard M. Whiting A dissertation submitted to the faculty of The University of Utah in partial fulfillment of the requirements for the degree of Doctor of Philosophy College of Social Work The University of Utah December 2014 Copyright © Richard M. Whiting 2014 All Rights Reserved The University of Utah Graduate School STATEMENT OF DISSERTATION APPROVAL The dissertation of Richard M. Whiting has been approved by the following supervisory committee members: Marilyn Luptak , Chair 09/29/2014 Date Approved Patrick T. Panos , Member 09/29/2014 Date Approved Caren Jean Frost , Member 09/29/2014 Date Approved Lawrence Henry Liese , Member 09/29/2014 Date Approved Scott D.Wright , Member 09/29/2014 Date Approved and by , Chair/Dean of Jannah H. Mather the Department/College/School of and by David B. Kieda, Dean of The Graduate School. Social Work ABSTRACT This study focused on how law enforcement officers' incorrect knowledge and ageist beliefs about older adults affected their response to scenarios of elder mistreatment. Regression analyses was conducted using data from a survey of 200 law enforcement officers from the intermountain west area. The Utah Commission on Aging created the survey in response to low usage of Utah's Vulnerable Adult Statute. The survey incorporated the Palmore's Facts on Aging Quiz 1 (FAQ1) and two scenarios of elder mistreatment. This analysis assumed that accurate knowledge of aging and of older adults, as measured by the FAQ1, was an adequate proxy for ageism. The factors of gender, years in law enforcement, education, and rank were included as controls. The examination found that officers' correct knowledge as measured by their FAQ1 Knowledge scores was positively correlated with the scenario response scores (an index reflecting the number of protective measures respondents described). The regression furthermore indicated a negative correlation to the scenarios when officers held incorrect, but positive views of older persons. Training specific to the Utah Vulnerable Adult Statute was also examined to determine if such training would increase appropriate responses by law enforcement officers to the scenarios. It was found that statute specific training did increase officers' use of the Vulnerable Adult Statute, but did not increase the overall effective response of officers toward vulnerable older persons. To my wife Beverly, my children, my family, and my students. CONTENTS ABSTRACT ....................................................................................................................... iii LIST OF TABLES ........................................................................................................... viii ACKNOWLEDGEMENTS .............................................................................................. ix Chapter 1 INTRODUCTION .........................................................................................................1 Research that Addresses Elder Mistreatment and Law Enforcement ............................4 UCOA Survey...... ..........................................................................................................6 Hypotheses .....................................................................................................................7 Primary Hypothesis ..................................................................................................7 Secondary Hypotheses .............................................................................................7 Benefits of the Analysis of Agency Data .......................................................................8 Organization of the Dissertation ....................................................................................9 2 LITERATURE REVIEW ............................................................................................11 Definition, Scope, and Prevalence ...............................................................................11 Elder Mistreatment.................................................................................................13 Difficulties of Definition........................................................................................14 Underreporting .......................................................................................................16 Ageism .........................................................................................................................17 Explanations of Ageist Attitudes and Behavior ...........................................................21 Cognitive Theory ...................................................................................................23 Sociological and Social Learning Theories ...........................................................23 Terror Management Theory ...................................................................................25 The History of Elder Mistreatment Policy ...................................................................26 Statutes ...................................................................................................................27 Criminalization ......................................................................................................28 Failure of Funding..................................................................................................30 The Police as Gatekeepers ...........................................................................................31 Police Response .....................................................................................................33 The Victim's Attitude as an Obstacle ....................................................................35 Fear of retaliation .............................................................................................36 Conflicted attitudes ..........................................................................................37 Fear of loss of rights ........................................................................................38 Police Training .......................................................................................................40 The Elder Justice Act .............................................................................................42 Police Attitudes ......................................................................................................44 Unclear laws.....................................................................................................45 Changes in reporting ........................................................................................47 How well the reporting system is working ......................................................48 Ageism and Law Enforcement.....................................................................................49 Discrimination........................................................................................................49 Effect of the Media ................................................................................................51 Summary ......................................................................................................................52 3 METHODOLOGY ......................................................................................................56 UCOA's Methodological Choices ...............................................................................57 The UCOA Survey Sample Recruitment ...............................................................58 The UCOA Survey Development Procedures........................................................58 The UCOA Survey section one-participant experience training and familiarity with vulnerable adults ....................................................................58 The UCOA Survey section two-vulnerable adult/mistreatment scenarios ......60 The UCOA Survey section three-Facts on Aging Quiz 1(FAQ1) instrument.61 The UCOA Survey section four-potential moderating variables ....................64 Ethical Considerations Regarding the UCOA Survey ...........................................65 Limitations of the UCOA Survey Affecting the Data Analysis ............................66 Justification and Current Research Objectives Given the Identified Limitations .............................................................................................................66 A Study of the UCOA Dataset Using Linear Regression to Determine the Effect of Ageism on Law Enforcement Response to Elder Mistreatment..............................69 Use of the FAQ1 Knowledge Score.......................................................................70 Statistical Methods, Data Analysis, and Interpretation ................................................71 Examination of Statistical Assumptions ................................................................71 Sample representativeness ...............................................................................71 Level of measurement ......................................................................................71 Protecting against bias .....................................................................................71 Correcting for multicollinearity .......................................................................72 General Study Procedures ......................................................................................73 Dependent Variable ...............................................................................................74 Independent Variables ...........................................................................................77 Control Variables ...................................................................................................79 Data Analysis .........................................................................................................80 Summary of Methodology ...........................................................................................81 Objectives ..............................................................................................................81 Design ....................................................................................................................81 vi 4 FINDINGS ...................................................................................................................83 Sample..........................................................................................................................83 Variables ......................................................................................................................85 Dependent Variable: Scenario Response ...............................................................85 Focal Variables ......................................................................................................86 Regression Models .................................................................................................92 Scenario Response Subscales ..............................................................................101 5 SUMMARY ...............................................................................................................106 Discussion ..................................................................................................................107 Strengths and Limitations ....................................................................................116 Conclusion and Recommendations ......................................................................120 Implications for Social Work Practice .................................................................122 Appendices A DEFINITIONS ...........................................................................................................127 B UCOA SURVEY .......................................................................................................129 C VULNERABLE OLDER ADULT SCENARIOS .....................................................141 D PALMORE FACTS ON AGING QUIZ ....................................................................143 E SCENARIO SCORING RUBIC ................................................................................150 F REPORTING REQUIREMENTS FOR LAW ENFORCEMENT ............................152 G STRATEGY FOR RESPONSES TO THE TRAINING ITEM CONCERNING RANK, "HIGHEST POSITION SERVED" .................................154 H RESULTS FOR LINEAR (OLS) REGRESSION USED FOR PREDICTING POLICE OFFICER RESPONSE TO SCENARIOS OF ELDER MISTREATMENT-BOTH FOCAL AND CONTROL VARIABLES ....................156 I CORRELATION TABLE FOR THE PREDICTOR VARIABLES .........................158 REFERENCES ................................................................................................................160 vii LIST OF TABLES 3.1 Elder mistreatment scenario scoring elements .......................................................76 3.2 FAQ1 Scores ..........................................................................................................78 4.1 Control continuous variable response frequencies ................................................84 4.2 Categorical control variable response frequencies ................................................84 4.3 Descriptive statistics for the elder mistreatment response .....................................85 4.4 Elder mistreatment scenarios 1 and 2 combined scores.........................................86 4.5 Descriptive statistics the FAQ1 scores ..................................................................87 4.6 Training item response frequency..........................................................................90 4.7 Results for nested linear (OLS) regression models predicting police officer response to scenarios of elder mistreatment ..........................................................95 4.8 Individual item frequencies totals from elder mistreatment scenarios 1 and 2 .....................................................................................................................102 4.9 Results for nested OLS regression models predicting police officer use of the VA statute in their response to scenarios of elder mistreatment ............................... 104 ACKNOWLEDGEMENTS I would like thank Dr. Marilyn Luptak, dissertation chair, for her support and encouragement through my dissertation process and to Dr. Hank Liese, Dr. Patrick Panos, Dr. Caren Frost, and Dr. Scott Wright for serving on my committee, for all of the time spent encouraging, consulting, and reviewing. Further, thanks to Maureen Henry and the Utah Commission on Aging along with Sheriff Jim Winder and the Utah Sheriffs' Association for their support in this project and in allowing me to use their data and to each of the officers who took part in the UCOA Survey. Special thanks to colleagues Dr. Tiffany Jensen, Dr. Mike Able, Dr. Michael Tatum, and to Ashton Holbrook and all the faculty and students who have been so encouraging and understanding while I completed this project. Finally thanks to my greatest support, my wife for all of her ideas, time and patience, editing, and believing in me. Also to my children and all of my family and especially my parents-my father for a life time of preparation and my mother for her example of service. CHAPTER 1 INTRODUCTION In 2008, the Utah Commission on Aging (UCOA), with the Utah Association of Sheriffs, conducted a survey of law enforcement officers to explore the causes of underreporting and lack of prosecution in cases of elder mistreatment. One part of the survey focused on officers' knowledge of and beliefs about older adults. This portion of the survey questioned whether law enforcement personnel mirror society's pervasive ageist attitudes and behaviors. An ageist attitude (or ageism) refers to biased attitudes or stereotypical beliefs held about a group of people based on age (Wilkinson & Ferraro, 2004). Ageism includes the negative treatment of those over 65 (Cuddy, Norton, & Fiske, 2005). As a secondary analysis of data, this dissertation explores the results of the UCOA 2008 survey to see whether law enforcement officers' responses to incidents of potential elder mistreatment reflect an ageist attitude toward older adults and affect the number of cases reported to adult protective services (APS). It further questions how ageism among officers influences the number of cases that result in arrest, prosecution, and conviction in the criminal justice system. This project seeks to identify correlations between ageist attitudes or behavior by law enforcement personnel and effective response to elder mistreatment. Recent estimates show that 11.4% of elders 60 and older experienced some form 2 of mistreatment (excluding financial) in the previous year (Acierno et al., 2010), which is equal to more than one out of every nine elders experiencing mistreatment, an annual rate that is almost seven times greater than that of child abuse (Sedlak et al., 2010). Census Bureau estimates (Federal Interagency, 2010) suggest there are 39.6 million people age 65 and older in the United States, making the number of mistreated elders more than 4.7 million each year. The Census Bureau estimates the number of people 65 and older will be about 72.1 million by 2030 (U. S. Census Bureau, 2012). That would constitute more than 19% of the population. Several studies show the number of reported cases of elder mistreatment has risen dramatically. The estimates range between 150% from 1986 and 1996 (Tatara & Kuzmeskus, 1997) to 85% from 1993 and 2004 (Teaster et al., 2006). Given the expected increase in the older adult population, the number of mistreated elders may double over the next 20 years without a more effective response by society. When the problem of elder mistreatment first emerged in the late 1970s, policy makers created oversight agencies required to intervene for mistreated elders. After initial efforts to intercede proved ineffective, lawmakers changed statutes to make elder mistreatment a criminal offense. Despite these attempts, there was no increase in the number of arrests or prosecutions (Quinn & Heisler, 2004; Ramsey-Klawsnik, Teaster, Mendiondo, Marcum, & Abner, 2008). National survey estimates found that in 2004, 565,747 cases of elder mistreatment were reported to adult protective service (APS) agencies, which represented less than a quarter of actual cases (Teaster et al., 2006). This disturbing trend has persisted despite all efforts to reverse the tide. Lack of change suggests those who make policy must take a more studied and serious look at the underlying reasons for the lack of intervention or nonresponse. Stiegel (2000) noted that 3 while not all cases needed to go to trial, the overwhelming majority never reached the judicial system. Connolly (2010) has observed that this pattern of nonresponse has continued into the present; most cases never come to the attention of authorities and, if they do, prosecution remains as unlikely as ever. Some studies explored the problem of underreporting but ignored the almost nonexistent number of prosecutions. The limited available data, however, support the view that this problem is serious. Dubble (2006) stated that in one study (Teaster, Roberto, Duke, & Kim, 2000), law enforcement investigated only 14 of 42 substantiated cases of sexual abuse for individuals age 60 and above. Two resulted in prosecution. In a more recent report (Schuller, 2006), of 639 reported elder mistreatment cases, police and APS referred only five to authorities for investigation that led to criminal prosecution. The report of the Utah Commission on Aging (UCOA) further supports the claim that elder mistreatment is perpetuated by both underreporting and nonresponse. Maureen Henry (executive director of the commission at the time of the UCOA study) pointed out the UCOA Safety committee identified the most serious cases of elder mistreatment reported in Salt Lake County between 2007 and 2008 (personal communication, August, 2008 & November, 2012). Of the 263 cases, none resulted in formal prosecution as violations of the elder mistreatment criminal statute. After examination, UCOA formed an alliance with the Utah Association of Sheriffs to explore how law enforcement contributed to underreporting and low prosecution rates. Their hope was to identify underlying causes that perpetuate this confusing response to the problem of elder mistreatment in order to reverse the trend. 4 Research that Addresses Elder Mistreatment and Law Enforcement The present study examined data gathered by UCOA to explore underreporting of elder mistreatment in Utah. The study examined the intersection between research on elder mistreatment and law enforcement in order to understand the role law enforcement plays in perpetuating underreporting. The findings mirror what is happening throughout the country with elder mistreatment. Existing research has shown that law enforcement, as principal gatekeepers in cases of elder mistreatment, have often used their discretion to handle cases at the scene without referring them on for further investigation. Daniels, Baumhover, Formby, and Clark-Daniels (1999) found that police officers reported less than half of the cases of suspected elder mistreatment to APS. When policies gave first responders discretion to decide how to respond to elder mistreatment cases, findings showed officers were unlikely to refer for further investigation. Those decisions have contributed significantly to the problem of underreporting and to low levels of prosecution. The number of elder mistreatment cases law enforcement officers report will only increase when policy makers understand the mechanisms that prevent officers from referring cases for further investigation. While a limited number of studies address law enforcement response to elder mistreatment, their conclusions suggest cause for concern (Daniels et al., 1999; Davies, 2010; Payne, 2002; Plotkin, 1988; Sever & Youdin, 2006). Researchers often cite the lack of collaboration between agencies as a major barrier to improving response to elder mistreatment (Blakely & Dolon, 2000; Dauenhauer, Mayer, & Mason, 2007; Plotkin, 1996). This is true of the research that addresses the interaction between law enforcement and other responding agencies. The literature indicated the 5 underlying assumptions of law enforcement and other social services agencies varied significantly (Bergeron, 2001; Blakely & Dolon, 2000; Paveza & VandeWeerd, 2006; Payne & Berg, 2003; Plotkin, 1996). APS, for example, was more likely to see the problem as one of caregiver stress (Antezberger, 2000). Law enforcement, on the other hand, identified the problem as a criminal matter (Heisler, 2000; Payne & Berg, 2003). This contrast explains in part why these two agencies endorse significantly different responses to elder mistreatment. Another aspect of this problem is the lack of familiarity by law enforcement personnel in working with the older adult population. Several studies have suggested that this reflects a lack of training (Heisler, 2013; Heisler & Stiegel, 2002; Payne & Berg, 2001; Plotkin, 1996; Stiegel, 2006). Further, the lack of cooperation, lack of familiarity with the population, and lack of training suggests there is also a lack of political will to make the changes needed to improve how this problem is handled (Payne, 2005). The struggle is most evident in the lack of uniform legislation or funding necessary to create the needed resources. Researchers have long contended that familiarity with the laws, accompanied by training on the realities of responding to the needs of older people, will decrease frustration and lead to an increased application of the laws (Bonnie & Wallace, 2003; Stiegel, 2006). To this end, Chermak (1993) argued that lack of training specific to the issue of elder mistreatment is the primary problem leading to underreporting by law enforcement personnel and that developing training programs is the most effective means for bringing about change. Ironically, the argument that training will lead to an overall improvement in the 6 response to elder mistreatment has been around for 30 years. At the same time, the number of suspected and reported cases has increased dramatically with little or no increase in the rate of prosecutions. Few direct studies have looked at whether training leads to an increase in proper responses to older adults by law enforcement (Stiegel, 2006). Still fewer studies have looked at the factors that influence how law enforcement officers respond to possible cases of elder mistreatment (Davies, 2010). Moreover, no studies have examined the effect of ageism on how law enforcement personnel respond to elder mistreatment. There is a limited understanding of how and why officers decide whether to report elder mistreatment to APS, to treat the referral as a criminal case, or to do nothing. Many unknowns remain concerning the response of law enforcement officers towards elder mistreatment. UCOA Survey In examining the factors believed to affect the underreporting and lack of prosecution of elder mistreatment, UCOA and the Utah Association of Sheriffs sought to understand whether ageism played a significant role in how law officers respond to elder mistreatment. The agencies hoped that a better understanding of these factors would lead to more effective response by law enforcement. The survey also examined whether officers who report receiving specific elder mistreatment training responded more appropriately to scenarios involving elder mistreatment than officers who do not report such training. As part of the evaluation, the UCOA survey looked at variables thought to influence ageism as well as variables that may affect law enforcement efficacy in general, including self-reported training, experience working with older adults, familiarity with 7 legal statutes pertaining to elder mistreatment, officers' age, number of years in law enforcement, level of education, and rank on the police force. Specifically, the UCOA survey was designed to answer the two questions. First does ageism as knowledge of and bias toward older adults play a significant role in how law enforcement officers respond to elder mistreatment? And secondly, if ageism is a factor in law enforcement response to elder mistreatment, are there other factors that could help to explain the relationship between ageism and law enforcement response to elder mistreatment? Hypotheses The hypotheses tested in this study are as follows: Primary Hypothesis Hypothesis #1: An increase in law enforcement officers' knowledge of older adults, as measured by the Facts on Aging Quiz 1 (FAQ1) Knowledge score, will predict a corresponding increase in response scores on elder mistreatment scenarios. Secondary Hypotheses Hypothesis #2: An increase in officer bias as indicated by FAQ1 Bias Score will lead to a decrease in officer's response scores to the elder mistreatment scenarios. Hypothesis #3: Law enforcement officers who reported having received statute specific training will have higher scores on elder mistreatment scenarios than those who report no training. The purpose of this current research project was to assist UCOA and the Utah Association of Sheriffs further in evaluating the data from the survey and to test the 8 original proposition that ageism does affect underreporting of elder abuse. Further, this project explored the data to determine which, if any, of the other identified factors were related to the independent variable of ageism and played a significant role in how law officers respond to elder mistreatment. Linear or ordinary least squares (OLS) regressions were used to analyze the dependent variable, police officer response to scenarios of elder mistreatment. The use of linear regressions made it possible to determine which factors were significant and then to quantify how much effect each predictor variable had on the outcome variable, law officers' response to scenarios of elder mistreatment. The data were further entered into modeling equations to determine whether a combination of variables increased the value of the statistical model in predicting changes in the dependent variable. Benefits of the Analysis of Agency Data Local law enforcement agencies and oversight bodies such as UCOA can use the findings from this study to inform policies that direct the response to elder mistreatment. Benefits include developing new strategies for improving officer response to elder mistreatment and identifying those officers most likely to be effective in working with victims of elder mistreatment. Also, administrators will have information that can lead to more effective training of officers who work with cases of elder mistreatment. The findings of the study are useful for informing future research and developing more effective instruments to examine how police, as well as other professionals, respond to elder mistreatment. Findings also provide a greater understanding of the importance and magnitude of the role ageism plays in influencing officers' response to elder mistreatment as well as identifying other factors that may affect that relationship. 9 Additionally, the study provides an indication whether the FAQ1 as a measure of knowledge of aging and of older adults and as an indirect measure of age bias is useful in discriminating between officers who respond appropriately to elder mistreatment scenarios and those who do not. By examining one potential explanation for the lack of reporting and prosecution of elder mistreatment, this study helped to unravel the troubling reality of elder mistreatment in American society. The current study also provides preliminary indicators of whether self-reported past training relative to elder mistreatment correlates with responses to FAQ1 scores. By assessing the relationship between officers who reported having received elder abuse training and their scores on the FAQ1, a model was developed to explain how these and other variables interacted to contribute to effective police responses to potential elder mistreatment. The model will assist agency administrators and local and state policy makers in determining whether to continue existing training strategies, adjust training programs, or explore other options. Ultimately, this can lead to more effective use of limited funding to produce real change. Organization of the Dissertation Chapter 2 reviews the literature relevant to elder mistreatment and law enforcement response to that phenomenon; it also reviews the importance of ageism as it relates to this topic and the theoretical bases needed to understand it. Chapter 3 describes the methods of this study, including the research questions, datasets, and plans for analysis. Chapter 4 presents the results of the data analysis, the characteristics of the sample, and the major findings. In order to conduct regression analyses of the dependent variable, open-ended responses to scenarios of potential elder mistreatment were 10 analyzed and coded to correspond with objective indicators of appropriate response by law enforcement officers. The final chapter discusses the results and summarizes the findings regarding law enforcement response to elder mistreatment and the extent to which ageism affects that response, as well as those factors found to moderate that effect. The chapter also describes the limitations of the study and discusses implications for the field of social work and policy makers. Finally, the chapter includes recommendations for future research that addresses the intersection between elder mistreatment, law enforcement, and social work practice. CHAPTER 2 LITERATURE REVIEW Society's response to elder abuse over the course of the past 40 years has been slow and incremental. This slow response may reflect an effort to avoid a dramatic disruption to society's existing institutions while searching for remedies to address the needs of elders. Yet, a careful examination of the response to elder mistreatment reveals that the approach has not been effective in reducing the rate or incidence of elder abuse. The following sections provide a basic understanding of the definition of elder mistreatment and the extent of the problem, as well as a discussion of ageism and relevant theories, a history of policies directed at the problem, and research concerning law enforcement response, including the interaction between law enforcement and ageism. Definition, Scope, and Prevalence To begin a review of elder mistreatment, it is necessary to understand the differences and similarities with the more frequently used term elder abuse. Broad guidelines intended to assist states in defining elder abuse were created in the 1987 amendments to the Older Americans Act (OAA; Older Americans Act, 1987). The specific section addressing the issue of elder abuse states that abuse "means the willful (A) infliction of injury, unreasonable confinement, intimidation, or cruel punishment with 12 resulting physical harm or pain or mental anguish; or (B) deprivation by a caretaker of goods or services which are necessary to avoid physical harm, mental anguish, or mental illness" (p. 52). The term elder, meaning "an older individual," is then applied to this definition to give us elder abuse (p. 52). The act defines Caretaker as "an individual who has the responsibility for the care of an older individual, either voluntarily, by contract, receipt of patient for care, as a result of family relationship, or by order of a court of competent jurisdiction" (p. 53). The term exploitation is defined in the amendment as "the illegal or improper act or process of a caretaker using the resources of an older individual for monetary or personal benefit, profit, or gain" (p. 53). Neglect is characterized as "the failure to provide for oneself the goods or services which are necessary to avoid physical harm, mental anguish, or mental illness or the failure of a caretaker to provide such goods or services " (p. 53). Finally, physical harm refers to that which leads to "bodily pain, injury, impairment, or disease" (p. 53-see also Appendix A). It is evident that the term elder abuse includes a range of behaviors. The definition of elder abuse provided by the National Center for Elder Abuse (NCEA) includes the following types: physical abuse, sexual abuse, emotional or psychological abuse, neglect, abandonment, financial or material exploitation, and self-neglect (National Center, 2011b). The issue of elder abuse divides into three primary areas of focus: institutional, self-neglect, and abuse (National Center, 2011a). According to Stiegel and Klem (2008), three distinct policy areas aid in defining the intersection between law enforcement and elder abuse: those laws that are directed toward adult protective services (APS), those pertaining to institutional abuse, and those intended to address the issues directed to long-term care ombudsman programs (LTCOP). 13 Legislation and statutory definitions have been created to regulate and otherwise respond to the differing concerns in each of these areas. Next, it is important to distinguish between domestic and institutional abuse. Domestic forms of elder abuse occur in homes or private residences and are primarily the domain of adult protective services (APS). Law enforcement most often encounters concerns in this area for which laws have been specifically created. Institutional elder abuse refers to those instances when abuse occurs in medical or long-term care facilities. Oversight for the latter is generally provided by the Long-Term Care Ombudsman Program (LTCOP), Medicare fraud units, and institutional oversight entities such as the Joint Commission on the Accreditation of Healthcare Organizations (JCAHO). While not entirely excluded, community law enforcement agencies such as police or sheriff's departments are much less likely to be called on to respond to cases of institutional abuse. Another area, self-neglect, seldom involves law enforcement and response to it is appropriately left primarily to social welfare agencies, including APS. While these different areas of focus may overlap with one another and call for significantly different responses, the UCOA survey and the current analysis focused on law enforcement response to elder abuse that occurred in domestic settings. Elder Mistreatment In their effort to clarify some of the definitional problems, Bonnie and Wallace (2003) made a distinction from the more inclusive term of elder abuse, instead choosing the term elder mistreatment to specify acts of abuse directed at vulnerable elders in domestic settings. They suggested that elder mistreatment consists of the following: (a) intentional actions that cause harm or create a serious risk of harm (whether or 14 not harm is intended) to a vulnerable elder by a caregiver or other person who stands in a trust relationship to the elder, or (b) failure by a caregiver to satisfy the elder's basic needs or to protect the elder from harm. The term…‘mistreatment' is meant to exclude cases of so-called self-neglect-failure of an older person to satisfy his or her own basic needs and to protect himself or herself from harm- and also excludes cases involving victimization of elders by strangers. (p. 1) As discussed previously, this study primarily focused on domestic APS situations and drew on the narrower definition suggested by Bonnie and Wallace of elder mistreatment. However, law enforcement personnel may be called upon to respond to situations in which a crime has been committed in a nursing home or other institution that falls outside this definition. Thus, the term elder mistreatment will be used when discussing domestic circumstances rather than institutional forms of elder abuse. The term elder abuse will be used when the discussion at hand extends beyond the narrow use of mistreatment. It is useful to note that while the term elder abuse was more prevalent in earlier research and discussion, the term elder mistreatment will be substituted when to do so would be more accurate and consistent. Difficulties of Definition Decisions as to what should be included or excluded in the definition of elder abuse and mistreatment directly affect the efforts to establish the scope and prevalence of the problem as well. With respect to the present study, this issue is particularly relevant for law enforcement officers and the criminal justice system. Law enforcement officers must identify whether an act is abusive and has affected a vulnerable elder. They must then determine whether the abusive behavior was carried out by someone in a caretaking role in a domestic setting, all of which must be assessed before applying statutes intended to address the problem. All states have enacted legislation to address the problem of elder mistreatment, 15 but statutes differ significantly from state to state (Jogerst, Daly, & Ingram, 2001; Roby & Sullivan, 2000). One area in which the statutes are unclear has to do with age. In some states, as well as under federal law, people are considered to be elders if they are 60 years of age or older. In other states, they must be 65 years of age or older to be recognized as elder. Stein (1991, 2006) observed that this lack of an agreed-upon definition for whether an individual is considered an elder represents just one of the many challenges that emerged almost as soon as researchers began to explore the problem of elder mistreatment. Another problem that prevents researchers from effectively determining the scope of the problem is whether to include estimates of financial exploitation and neglect in these numbers. For example, Wasik (2000) estimated that each year 5 million older Americans are victims of financial exploitation and that only 4% of those cases are reported. Bonnie and Wallace (2003) have argued that in order for research on elder mistreatment to move forward and for necessary funding to be made available, it is imperative to resolve the problems of scope and definition. This is clearly an important issue and one that will continue to plague researchers as well as policymakers seeking to advance the cause of mistreated elders. This issue is important to the discussion at hand only insofar as it serves to explain the difficulty encountered by those who are required by statute to respond to the problem. While the issues of scope and prevalence, including incidence, have eluded researchers, national concern has grown regarding the mistreatment of elders, especially in light of data that suggest that persons 65 and over constitute the fastest growing part of the population (Vincent & Velkoff, 2010; Treas, 1995). As noted in Chapter 1, 16 significant increase in reported mistreatment corresponds with an increase in the elder population. While the increase in the annual number of reported cases cannot be attributed solely to an increase in the older adult population, it does confirm that as the older adult population has increased the incidence of elder mistreatment has also increased. Though there continues to be disagreement as to the extent of elder mistreatment, the urgency of the problem cannot be ignored. One glaring concern is the increase in the mortality rate for elders experiencing neglect or abuse (Dong et al., 2009; Lachs, Williams, O'Brien, Pillemer, & Charlson, 1998). Data gathered by the Federal Bureau of Investigation (FBI) show that violence directed at older adults has included over 1,000 homicides in a given year and an additional 120,000 assaults (Chu & Kraus 2004) highlighting the need for a more effective response by society. Underreporting Bonnie and Wallace (2003) observed that estimates of the incidence of elder abuse ranges between 2% and 10% of the total older adult population, which speaks to the issues of definition and the difficulty of obtaining realistic numbers. At the same time, most experts agree that the majority of cases of mistreatment go unreported (Acierno et al., 2010; Bonnie & Wallace, 2003; Tatara, 1993). The widely held view that reported cases represent the tip of the iceberg is supported by a variety of studies (Daniels et al., 1999; Pillemer & Finkelhor, 1989; Tatara, Kuzmeskus, Duckhorn, & Bivens, 1998), which have found that the majority of cases of mistreatment never come to the attention of authorities. Tatara et al. (1998) observed that 84% of cases of elder abuse were not reported to any agency designated to respond. The number of mistreated elders 17 seems to lay somewhere between reports and estimates-a range between .05% and 10.0% of the total elder population according to Bonnie and Wallace (2003). One reason offered for this underreporting is that elders are targets for different types of crimes than are younger persons. According to Siegel (2010) and Payne (2002), elders are more likely to be victims of telemarketing fraud, occupational fraud, patient abuse, and medical crimes. Because FBI's Uniform Crime Report (UCR) does not track many of the crimes perpetrated against older adults, it is difficult to establish crime rates. Payne added that while elders may experience a significantly lower rate of crime, these statistics can be misleading because they are less likely to be victims of typical violence or street crimes. It is also interesting to note that while crime rates have fallen dramatically since the mid-90s in almost every category, according to the UCR and the National Crime Victimization Survey (NCVS), rates of crimes against older adults have not declined (Payne, 2002). One explanation for the lack of change in the rate of crimes committed against older adults may be that American society is not concerned with this population due to the influence of ageism (Harbison, 1999). Ageism Ageism was first defined by Butler (1989) when he stated, Ageism can be seen as a systemic stereotyping of and discrimination against people because they are old, just as racism and sexism accomplish this with skin color and gender. Old people are categorized as senile, rigid in thought and manner, old-fashioned in morality and skill…. Ageism allows the younger generation to see older people as different from themselves; thus they subtly cease to identify with their elders as human beings. (p. 139) Palmore (1999) added to our understanding of what it means to be "old." To him, old age in America came to mean that one could no longer contribute to the society. 18 Palmore defined nine negative characteristics that he believed to be part of stereotypical view of older adults: (a) sick and disabled, (b) impotent, (c) ugly, (d) cognitively impaired and unable to learn, (e) mentally ill, (f) useless, (g) isolated, (h) impoverished, and (i) miserable because of the conditions listed above. Terms that were synonymous with being older such as retired have become another way of saying "incompetent, useless, and worn out" (Falk & Falk, 1997). These negative stereotypes that occur based primarily on age are an important part of ageism. That age bias in American society is widespread is well accepted (Harbison, 1999; Kite & Wagner, 2002). One study found that ageism is more common than all other forms of discrimination (Ray, Sharp, & Abrams, 2006). An example of ageist attitudes can be found in the study by Blakely and Dolon (1998) in which the young adult respondents did not think they would ever encounter elder mistreatment and described elders as "depressing to be around" and "creepy." These authors mentioned that the young adult respondents in their study did not enjoy being around older adults and only did so out of a sense of obligation. Blakely and Dolon concluded that sympathy toward older adults may be just another manifestation of ageism. In either circumstance, feeling a sense of obligation toward older adults or believing they are incompetent creates a separation between this population and others (Block & Sinnott, 1979). As a physician working among older adults in Washington, DC, Robert Butler observed the mistreatment of elders and later stated that all mistreatment of elders is a result of ageism (Falk & Falk, 1997). He went on to say, Ageism, like all prejudices, is not based on reason, but on emotion. Therefore, like all prejudices, ageism will have it both ways. On the one hand, ageists seek to eliminate the old from the ranks of workers as soon as possible on the grounds that the jobs held by the old should be given to the young. At the same time, the 19 ageists complain that they need to support the old who are no longer working. (p. 24) The media have also contributed to this stereotype. Leedahl and Ferraro (2000) observed that media influence has encouraged ageism, which in turn has led to a lack of societal concern for problems associated with older adults, including mistreatment. Compounding the feelings of alienation and isolation from the community experienced by older people, stereotyping effectively has allowed society to ignore many of the problems associated with being an older adult, such as unacceptable living conditions, institutionalization, and illness (Ragan & Bowen, 2001). It is not surprising then, that ageism and crime are more prevalent in urbanized societies where the multitudes of elders go unnoticed and unprotected (Block & Sinnott, 1979). This indifference toward older adults undoubtedly affects law enforcement personnel as well. Age bias has not been confined to negative stereotypes. The media have also stereotyped elders in positive yet unrealistic ways. Falk and Falk (1997) pointed out that positive ageism portrays older adults as physically fit, wealthy, carefree, and enjoying their "golden years." However, statistically elders are most likely not physically fit, not without cares, and not wealthy. This type of positive stereotyping has created the same dynamic discussed above in which society is disconnected from a set of its members. For law enforcement officers, specifically, this means they must make decisions based on inaccurate and age-biased information that contributes to inappropriate and ineffective responses to elders' needs. These competing stereotypes effect interactions between professionals and older adults. According to Reed, Cook, Cook, Inglis, and Clarke (2006), when law enforcement officers encounter older adults they have a difficult time knowing whether they are responding to a person who is poor, old, incompetent, and in 20 need of their help and protection or to an autonomous, affluent, capable adult who requires no assistance. Faulkner (1982) observed that elder mistreatment laws, which are intended to protect, are by their nature, ageist because they assume elders need our help based on age alone. Many elder mistreatment laws were created by borrowing from the existing child abuse laws and simply inserting the word "elder" for the word "child" (Anetzberger, 2000; Wolfe, 2003). Some would argue that this approach has infantilized older adults, perpetuated negative stereotypes, and increased system mistrust, thus having compromised confidence in the structure elders must rely on for aid and protection (Payne, 2002). On the other hand, Reed et al. (2004) observed that tensions have arisen when society has attempted to mandate reporting. As an example, despite the fact that the laws mandate that physicians report elder mistreatment, they have resisted doing so. Gibbs and Mosqueda (2007) found that only 2% of elder mistreatment reports were filed by family physicians. Faulkner (1982), after reviewing the literature, stated that mandated reporting by physicians does little to protect either society or older adult victims and instead promotes an ageist attitude toward elders. In Faulkner's view, mandatory reporting demands that physicians deny patients' rights to autonomy and selfdetermination and compromises the doctor-patient relationship. Faulkner concluded that this compromised relationship may lead to a situation in which older adults choose not to seek medical treatment that may lead to their loss of freedom and personal control as well as the fear that they may not be viewed as capable. This is just one instance of the complicated intersection between laws, the needs of elders exposed to mistreatment, and 21 reporters. This intersection is further complicated by how bias seems to affect the criminal justice system's response to elder mistreatment. Several researchers (Payne, 2002; Payne, Berg, & Toussaint, 2001) have observed that law enforcement crime reporting systems (such as the Uniform Crime Report and the National Incidence Based Reporting system) have not monitored elder mistreatment. Poor reporting results in poor tracking of mistreatment and other crimes against elders. These authors also observe that very little funding has been made available for researching crimes that affect primarily elders such as financial exploitation and neglect. Nerenberg (2002) acknowledged that until society eliminates negative stereotyping and ageist attitudes, there will not be necessary support for effectively addressing the policies and statutes that govern law enforcement response to elder mistreatment. This same author contended that educating the public about vulnerability and other factors that affect older adults is the best way to influence public policy for good. Johnson (2002) stressed that breaking down the misperceptions and stereotypical views of older adults is critical to changing policy and effectively addressing the needs of this vulnerable population. Butler (2008) added that if society is unwilling to examine the effects of age bias, it will not be able to generate policies that can adequately address the effects of that bias. Explanations of Ageist Attitudes and Behavior There is no unifying theory that clearly explains the lack of response by society generally, and law enforcement specifically, toward elder mistreatment (Walsh, Olson, Ploeg, Lohfeld, & MacMillan, 2010). Butler (2008) considered elder mistreatment a 22 manifestation of ageism, which also lacks a model or theory to illuminate the behavior associated with it (Walsh et al., 2010). Nelson (2002) suggested that this endures because ageism remains socially acceptable and has become so institutionalized that even the elders against whom it discriminates do not acknowledge it. Without a vocal victim, nothing appears wrong, and ageism persists, unacknowledged for the most part (Nelson, 2002). This view is supported by the fact that while elder mistreatment has been attributed to ageism, it has not been a focus of research. When victims do not acknowledge oppression against them and instead accept it and refuse to take efforts to rectify the situation, the phenomenon can be considered internalized oppression (Walsh et al., 2010). Mullaly (2002) observed the marginalization that occurs through prejudice, stereotyping, and oppression makes elder mistreatment even more difficult for elders to reject. Penhale (1993) pointed out that the lack of theory regarding elder mistreatment compounds this phenomenon and further ensures that elder mistreatment will not receive the systematic evaluation necessary to create effective policy. Although no specific theory directly addresses ageism, several theories help to better understand the phenomenon. First, cognitive theories help researchers to understand how stereotypes are created and how they function. Second, sociological and social learning theories help to explain how individuals incorporate attitudes already present in society into personal schemas and emerging stereotypes. Finally, conflict and terror management theories have been used to explain the seemingly contradictory nature of bias and discrimination toward a population to which each will eventually belong. 23 Cognitive Theory Cognitive theories explain the underlying processes of the psyche and draw on the early work of Allport (1954), who addressed the profound influence prejudice has on behavior. In regard to prejudice based on age, Cuddy et al. (2005) stated that children begin to categorize others based on age almost immediately. According to Cuddy et al. as children interact with members of a category, they begin to associate certain attributes and characteristics with that category. A number of elements contribute to this process, including categorical types and attributions held by significant others, personal experiences and interactions with members of a category, and so on. Eventually, these composites of experience form the basis of expectations for a member of a certain category such as "young" or "old," "male," or "White" and become stereotypes. Theorists describe this cognitive process as both universal and necessary to make sense of the world and to keep it ordered. Therefore, attitudes are formed very early, including those held about older adults. These stereotypical views can incorporate all of the biases held within society. Allport theorized that stereotyping leads to biased treatment toward members of a category and strongly predicts behavior. Regardless of whether stereotypes reflect positively or negatively on the group, they can contribute to potentially harmful attitudes and behaviors toward individuals because they are not based on reality (Palmore, 1988). Finally, cognitive theories suggest that stereotypes formed at a young age are very resistant to change. Sociological and Social Learning Theories Sociological and social learning theories emphasize the importance of the socialization process in acquiring attitudes and beliefs (Doerner & Lab, 2012). 24 According to these authors, theories vary across disciplines including social work, sociology, psychology, and child development, but all tend to agree that early socialization contributes to the formation of beliefs and attitudes that are consistent with modeled stereotypes. Particularly important in this process are the attitudes and beliefs held by intimate others, family, peers, as well as those in authority (Bandura, 1971). While children may categorize older adults based on their own interaction with them, socialization theories emphasize the influences of persons other than those belonging to the category. Although a child may hold a positive view of a grandparent early in life, this view eventually evolves to a more negative view of older adults in general. Seefeldt, Jantz, Calper, and Seroc (1977) observed that children's attitudes toward older adults become progressively more inaccurate, negative, and rigid over time-a change they attributed to this socialization process. Older adults become stereotyped as incompetent and incapable, and at once, warm and fuzzy (Nelson, 2009). These culturally held biases, reflected in the attitudes and behaviors of significant others, are readily transmitted to individual members of society. Internalized biases determine personal attitudes, beliefs, and behavior directed toward any group, but in this case, the older adult population (Montepare & Liebowitz, 2002). Nelson suggested that because these attitudes have become institutionalized, they are difficult to challenge and seldom are. Sorgman and Sorenson (1984) noted that the influence of media and literature to create and perpetuate these negative stereotypes is particularly true for older adults because children have such limited contact with older adults today. The media are ubiquitous and powerful mechanisms for social learning that have had a profound effect 25 on how elders are viewed in society and the creation of older adult stereotypes. Once again, stereotypical beliefs and attitudes held by society become a significant challenge that law enforcement must contend with as they respond to the needs of mistreated vulnerable older adults. Considering the formation and prevalence of these biases exposes the complexity of getting the criminal justice system to respond to the needs of vulnerable older adults. Terror Management Theory Terror management theory proposes that when individuals are reminded of their own mortality, they act in ways that attempt to disconfirm or avoid that reality (Nelson, 2005). Death is the final and natural consequence of aging. As such, society focuses a great deal of effort avoiding aging or any of its reminders (Martens, Greenberg, Schimel, & Landau, 2004). This theory suggests that ageism is not just discrimination but an act intended to deny the inevitable. Denial is consistent with lack of interest in ageism; if it can be denied, then it does not exist. Reducing the level of denial regarding ageism requires clear understanding of the scope of the problem and the mechanisms that maintain it. According to Nelson (2002), individuals use three primary areas to categorize or stereotype others: race, gender, and age. Nelson observed that while the corresponding prejudice and discriminatory practices of racism and sexism have received a great deal of attention, ageism has received much less. Thus far the available theories help to explain the mechanisms that both create and perpetuate ageism at an individual or small group level, but do less to explain how ageism functions on a societal level. At the broadest level, ageism has become an 26 integral part of our culture and society, and any effort that looks to change or eliminate ageism can be viewed as threatening to the existing social order. This may be why efforts to address the problem of elder mistreatment have only taken small incremental steps. In keeping with the conflict perspective, society must determine how to best allocate limited resources. Discriminatory behavior, in the form of ageism toward older adults, becomes a means to avoid change and to justify withholding resources from those viewed as "other" in our society (Turner, 1998). Consistent with this view, older adults have increasingly become a scapegoat for the problems associated with a lack of funding for all social programs (Lieberman, 2013; Street & Cossman, 2006). Understanding the effect of ageism in society can elucidate society's policies toward elder mistreatment and the limitations it imposes on law enforcement efforts to respond (Harbison, 1999). The History of Elder Mistreatment Policy To understand why the response to elder mistreatment has not significantly reduced its incidence, it is instructive to review the history of social policy directed at reducing elder abuse. Steinmetz (1988) observed that attention began to focus on this problem as experts began testifying before the U.S. Congress in the 1970s. Quinn and Heisler (2004) noted that these Senate investigations were conducted to better understand elder abuse and what was occurring at a national level. One of the findings of this early exploration was that most elder abuse was perpetrated by family members and could be realistically seen as a domestic problem (Wolf, 2003). With the federal government conducting hearings surrounding the mistreatment of elders, states began to respond in anticipation of federal mandates and expected funding (Anonymous, 2002a). 27 Statutes By the 1980s, three principal policies were introduced in an effort to bring this emergent problem under control (Quinn & Heisler, 2004). The first was to create statutes that defined elder mistreatment, accompanied by appropriate social sanctions against such behavior. The second was the creation of oversight bodies responsible for receiving reports of elder abuse, investigating those reports, and intervening in identified cases of mistreatment. By the mid-1980s most states had adult protective services (APS) agencies or their equivalent, which functioned to provide services to victims of elder mistreatment (Anonymous, 2002a; Faulkner, 1982). A third response was to use mandatory reporting laws to compel those professionals who come in regular contact with older adults to report instances of mistreatment (Crystal, 1987). Legislators anticipated that once these policies were implemented the problem of elder mistreatment would come under control (Quinn & Heisler, 2004). Payne and Berg (2003) noted that by implementing state reporting laws with enhanced penalties, government policy makers were also able to take a popular stance against elder abuse and mistreatment and make an aggressive show of their "tough on crime" policies regarding elder mistreatment. Payne and Berg observed this could be done without incurring any real financial cost and without regard for whether there was empirical evidence to suggest that these policies would be effective. As one feature of the new legislation, states created penalties associated with the violation of the new mandatory reporting laws. Part of the original intent of these early responses was to make sure that this type of domestic problem remained within the purview of social service agencies and kept out of the criminal justice system (Payne & 28 Berg, 2003). This was consistent with the contention that elder mistreatment, which is often perpetrated by caregivers, was caused by the stress created as they cared for their elders (Clarke, 2007). Perpetrators in the caregiver-stress model were essentially absolved of criminal intent and rather seen as individuals in need of services to reduce the stress associated with caring for an older person. Interventions to reduce caregiver stress were thereby provided by social service agencies, including removing the adult from the setting where the abuse occurred. This was similar to society's response to child abuse and domestic violence, which were viewed as primarily civil matters and not as criminal behaviors (Quinn & Heisler, 2004). Criminalization According to Payne, Berg, and Toussaint (2001), by the late 1980s and early 1990s social policy makers were beginning to recognize that elder mistreatment was a criminal rather than strictly a civil issue. Heisler (2000) noted that this change may have reflected a growing awareness by a variety of professionals working with elder mistreatment that, at least in the most serious cases, only the criminal justice system might be capable of stopping the problem and protecting the victim while holding the offender accountable. At the same time, research showed that the caregiver-stress model was not supported by the data and that instead a model of caregiver dependency was emerging. This view held that those who mistreated elders shared the same motivations as other criminals and were more likely to respond to law enforcement policies aimed at deterrence as opposed to policies directed at reducing stress and offering support to caregivers (Brownell, 1998; Quinn & Heisler, 2004). While there is not a specific moment in which the tide turned, congressional testimony by leading proponents and 29 researchers in the field in the mid-90s increased policymakers' understanding of this complex problem, changed the direction of public thinking, and influenced elder mistreatment policies (Wolf, 1996). Once elder mistreatment began to be viewed as criminal rather than strictly a civil matter (Beard & Payne, 2005), legislation reflecting this change followed. The decision to criminalize elder abuse was done despite a lack of consensus and limited research to support such a change. After implementation, there was little research to illuminate how law enforcement personnel could intervene in cases of elder mistreatment (Payne, Berg, & Toussaint, 2001). In addition to providing federal jurisdiction to certain crimes against elders, changes in the laws also increased the attached penalties. The changes, taking place nationally, were soon followed by changes at the state level (Daniels et al., 1999; Payne & Berg, 2003). For example, South Carolina changed elder neglect from a misdemeanor to a felony for which a prison sentence of 30 years was the maximum sentence (Heisler, 2000). Further, policy began to reflect the view that mistreatment of older adults was not just a personal act against older adults, but was also an antisocial act against the entire community (Heisler, 2000). Because policy makers believed that law enforcement could simply put these new laws into practice (Anonymous, 2002a; Payne, Berg, & Toussaint, 2001), little thought was given as to how best to implement such change. Payne et al. stated that policies were put in place without proper concern for their successful implementation because state lawmakers were preoccupied by the need to comply with anticipated federal mandates in order to receive the accompanying federal funding. 30 Failure of Funding In a cruel twist of fate, states' preemptive efforts to address the issue of mistreated elders were not followed by federal funding as expected (Anonymous, 2002a). Without clear federal guidelines in place or the necessary funds, laws were hastily constructed with little consistency from state to state. As discussed above, there was also concern that these laws had been created by substituting the term elder into existing child abuse laws (Faulkner, 1982). Beard and Payne (2005) added that differences between various states' laws included how elders were defined, what constituted elder abuse or mistreatment, whether to classify it as criminal or civil, what reporting standards to use, how it should be investigated, and what the sanctions or penalties should be. In 2002, Congress introduced, and failed to pass, the Elder Justice Act, which was intended to provide an important impetus for local government efforts and to fund private research (Anonymous, 2002a). Funding was to be provided for distributing information, employee training, ensuring secure shelters, and increasing prosecution. This funding proposal continued to work its way through Congress 8 years after first being introduced and was signed into law as part of health care reform in March of 2010. The law authorized the federal government to spend $757 million, a huge step for the federal government in responding to issues that concern elders, but it has yet to be funded by Congress (Willard, 2011). In spite of federal and state efforts to enhance laws addressing elder mistreatment, there continues to be a less than satisfactory response. As discussed in the first chapter, even the most serious cases in Utah were not prosecuted under the statutes addressing elder mistreatment. Utah has followed the national trend in which elder mistreatment is 31 unlikely to be prosecuted and legal interventions are considered to be a secondary solution to the problem, as noted by Dolon and Blakely (1989) and Byers and Hendricks (1993). Byers and Hendricks observed that the solution to this problem eluded lawmakers from the time it was identified more than 15 years earlier, and little has changed since they first made the observation. Today, the questions remain, why is elder mistreatment not prosecuted and what obstacles contribute to this lack of utilization of existing elder mistreatment laws? The Police as Gatekeepers Even prior to criminalization of elder mistreatment, it was evident that law enforcement personnel would be a logical point of entry for cases of elder mistreatment (Hodge, 1998). Other potential access points to the service delivery system included physicians, adult protective services, and other social service agencies working with elders (Daly, Jogerst, Brinig, & Dawson, 2003). Given that a significant portion of elder mistreatment concerns were thought to be physical or medical in nature, it was believed that medical doctors would be best qualified to identify mistreatment (Gibbs & Mosqueda, 2007). In a review of the research concerning the relationship between physicians and elder mistreatment, Gibbs and Mosqueda considered whether primary care physicians or emergency room doctors did in fact report elder mistreatment. As mentioned above, Rosenblatt, Cho, and Durance (1996) observed that reporting by family physicians made up only 2% of the total. They also noted that less than 25% of physicians believed that they should address the issue of elder mistreatment with their patients, and the majority reported they do not address this issue with their patients. It is not surprising then that so few cases are reported by physicians. While elder 32 mistreatment can be a medical issue, physicians are not effective gatekeepers for this problem. On the other hand, law enforcement officers are often the first to be called on when problems arise for older adults. They are frequently referred to as the "gatekeepers" to the criminal justice systems and play a primary role in accessing the social services systems (Payne & Berg, 2003). Often police are the first point of contact for those who report incidents of elder mistreatment (Daniels et al., 1999). In a review of practices of law enforcement, Schack and Frank (1978) found when problems arise unexpectedly, law enforcement personnel often represent the only agency to which elders can turn. Shack and Frank observed that police officers are society's safeguard for this otherwise isolated and disenfranchised population. Arcuri (1981) commented that police are in a unique position to help older adults because of their high visibility and ease of access. As early as 1988, Plotkin observed that in a third of all jurisdictions law enforcement was the designated reporting agency for issues of elder mistreatment. Citing a 1981 survey conducted by the Association for the Advancement of Retired Persons (AARP), Plotkin further noted that of the 1,001 people surveyed, nearly half indicated they would first contact the police when confronted with an instance of elder mistreatment. Plotkin also observed that this is particularly common when situations involving elder mistreatment have escalated to the point that those involved feel unsafe. In a subsequent study that examined police records, Brownell (1994) found that in a significant number of cases the victims reported the mistreatment to police. In short, the public perception that law enforcement is the appropriate point of contact when elder mistreatment has occurred is supported by the 33 data. Police Response Given that police are a logical and chosen point of access in responding to the problem of elder mistreatment, it is important to understand what their responses have been. According to Plotkin (1988), prior to the late 1980s, police did not aggressively respond to crimes against elders. Plotkin found that nearly a third of the police chiefs studied were not aware of laws related to elder mistreatment despite more than half having acknowledged that they had encountered various forms of elder mistreatment within their jurisdictions. Plotkin further observed that the majority of police chiefs of major cities had come in contact with elder mistreatment, including physical mistreatment, neglect, and financial exploitation. Half of these police chiefs reported that mistreatment was not typically being reported within their jurisdictions. So while unaware of laws and being less than aggressive in responding to cases of elder mistreatment, law enforcement agencies were acknowledging the problem and had come into contact with cases of elder mistreatment by the mid-1980s. One telling point revealed by the survey Plotkin (1988) conducted was that police chiefs reported they had not received adequate training regarding appropriate responses to elder mistreatment and had little knowledge as to how to proceed in such cases. Given that the social policy direction at the time was that adult protective service (APS) agencies would bear the primary responsibility for responding to cases of elder mistreatment, the lack of involvement and training of law enforcement personnel was not surprising. Complicating this matter has been the unstated police policy of "de minimus," in 34 which police officers attempt to resolve situations independently and thereby avoid further involvement with the criminal justice system (Parnas, 1973). This approach may benefit law enforcement personnel by diverting cases of elder mistreatment away from the criminal justice system, avoiding unwanted paperwork (Parnas, 1973). Law enforcement officers may also wish to convey respect to older adults and their right to autonomy and self-determination. While it is reasonable to assume that there is a perceived benefit to the system by having fewer cases to deal with and avoiding those cases traditionally thought of as civil, motivations are not yet clear. Daniels et al. (1999) provided evidence for this concern with their finding that only half of the officers who observed elder mistreatment and neglect indicated that they actually reported it, with even fewer indicating they reported financial exploitation. Daniels et al. explained that when police do not have adequate definitions of elder mistreatment, they are not clear on their obligation to report these cases. Lacking the resources necessary to intervene on behalf of older adults and society, officers often assume a broad discretion in such cases and again choose not to enter them into the criminal justice system for further investigation. Heisler (2000) observed that this problem persisted some 20 years after it was initially reported in the literature and suggested that the traditional responses by law enforcement assume that these victims do not want to become involved in the criminal justice system and at best are seeking only a solution for the immediate crisis. A study done by Brownell (1994), which obtained data on elder mistreatment using law enforcement reports in Manhattan, showed that a majority of victims indicated a willingness to prosecute at the time of the report. Brownell further mentioned that the 35 greater the severity of mistreatment, the more likely victims were to be willing to prosecute. Brownell's study is limited by the fact that, of course, unreported cases were not included. This study offers support for the position suggested by Heisler, that law enforcement officers may erroneously assume that elders do not wish to report cases of elder mistreatment. Heisler concluded that once a case of elder mistreatment has been identified, law enforcement personnel have difficulty moving the case through the criminal justice system toward prosecution and conviction and may not be motivated to do so. In short, while older adults may or may not be reluctant to report, law enforcement appeared to be either reluctant or unmotivated to make the report or enter it into the criminal justice system. While little research has considered the role of the police and other criminal justice agents in responding to elder mistreatment, the research that has been done suggests that the criminal justice system can be a useful method of disposing of elder mistreatment cases (Heisler, 2000; Payne, 2003). For this reason, it is particularly important to understand why police are choosing not to enter cases into the criminal justice system. The Victim's Attitude as an Obstacle While it is known that some elders who are mistreated do report their mistreatment, research has also shown significant obstacles to reporting by older victims (Brownell, 1994). One of the identified concerns is that law enforcement officers have a direct effect on the continued underreporting that exists among victims of elder mistreatment. Because of law enforcement officers' gatekeeping function, researchers have examined the role the older adult victim's attitude and beliefs about law 36 enforcement play in underreporting. Fear of retaliation. In the past, one factor believed to have been significant to both nonreporting and failing to press charges was fear on the part of older adult victims. This may explain findings that police chiefs believed mistreatment was not being reported to their organizations (Plotkin, 1988). Another concern continues that even when efforts have been made to invoke the full extent of the law, cases without credible witnesses or clear and convincing evidence are not likely to be prosecuted (Nerenberg, 2002). This can then create the opportunity for freed offenders to retaliate against those who have reported or testified concerning the mistreatment with the knowledge that the system was unable to effectively sanction them in their previous mistreatment of the elder. Nerenberg concluded that outreach efforts to identify victims of elder mistreatment were followed by an increase in violence. Elders who report mistreatment have a legitimate fear of retaliation and are justified when they are reluctant to report. The end result has been to further reduce the deterrent effect of the legal system. Blakely and Dolon (1989) went further, finding that fewer than 20% of APS workers surveyed perceived interventions to be effective in restoring safety and stability to the victims of elder mistreatment. Klein, Tobin, Salomon, and Dubois (2008) found that a small group of elder Rhode Island women whose offenders had been arrested and prosecuted were typical of the mixed findings on domestic violence. In this study, the women were as likely to be revictimized by their offenders regardless of prosecution. One reason cited for the limited deterrent effect was the relative lack of severity of punishment, with misdemeanor penalties often being meted out for felony offenses. These same researchers noted prosecutors often do not handle cases aggressively in order 37 to avoid requiring the victim to testify, which can then lead to reduced charges and briefer sentences. Again, the net effect conveys to the offender that their actions did not warrant severe penalties and may contribute to the belief that offenders will not have to answer for their actions. It is also important to recognize that older adult victims of mistreatment suffer far greater consequences than those who are younger (Arcuri, 1981; Brandl & Horan, 2002; Hamel, 1979; Jogerst, Daly, Brinig, Dawson, Schmuch, & Ingram, 2003; Payne, 2002). Hamel stated that the effect for all categories of mistreatment is more severe for older adult victims regardless of whether the neglect or mistreatment is physical, psychological, or financial. Hamel reminded us that this vulnerable and even fragile population does not easily recover from the effects of abusive crimes. They are less likely to rebound from financial losses, material exploitation, physical injuries, or emotional traumas associated with mistreatment. Lachs, Williams, O'Brien, Hurst, and Horowitz (1996) showed that elder mistreatment victims have shorter survival rates than other older adults. Given that elders suffer more, are less likely to recover, and are more likely to experience injury than other victims, it is understandable that this population has a greater fear of mistreatment and its consequences. Conflicted attitudes. In spite of these concerns, elders tend to have extremely favorable attitudes toward the criminal justice system. However, this changes once elders begin interacting with that system. A survey conducted by Arcuri (1981) found that the more contact respondents had with the police the more negative their attitudes were toward them. In other words, older adults tend to hold positive attitudes toward the police until they have occasion to interact with them. Almost 20 years later, Blakely and 38 Dolon (2000) also observed that those elders who had the most frequent contact with police often gave the most negative assessments of law enforcement. A related finding discussed by Sykes (1977) and Arcuri (1981) was that older adults were more disrespectful and hostile than younger populations when interacting with law enforcement. Research has suggested that this unexpected occurrence may be related to the negative expectations held by police toward older adults (Chermak, 1993). This author commented that police often believe that elder victims of crime are poor witnesses, particularly if there is evidence of dementia or other cognitive disabilities, and that these concerns may be warranted as elders often are overwhelmed by the adjudication process. Dunlap, Golding, Hodell, and Marsil (2007) confirmed that older witnesses are less likely to be perceived as credible. Dubble (2006) also noted that when older adult victims refused to press charges, the system was then required to determine whether it was in the best interest to protect the vulnerable older adult or allow the right of self-determination. According to Dubble, in cases of elder mistreatment, police must choose between respecting the victim's right to self-determination or overriding that right in favor of protection. Police must also consider that the elder's options for receiving care may be limited (Plotkin, 1996). Given these limited options, which further complicate how officers should respond to cases of elder mistreatment, it is not surprising that the choice to take the least intrusive path becomes more appealing. Fear of loss of rights. With respect to interventions, by its nature the criminal justice system encroaches on the right of older adults to determine how to manage their lives (Heisler & Quinn, 1995). This can include decisions about whether to prosecute or 39 whether to remain in a relationship that may or may not be perceived as abusive; decisions that can possibly affect where one will live-choices that some believe rightfully belong to older adults adult who may or may not be fully competent to make such choices. It also includes alternatives that may put the elder at greater risk for revictimization. The potential to have such control taken away can create a measure of stress and invoke a fearful and avoidant response. It can be reasonably postulated that a positive relationship with law enforcement based on positive experiences may improve the outcomes when complaints of elder mistreatment are made. From this perspective, it may be argued that in order to lower elder mistreatment within a society, the society must ensure that elders have a more positive experience when interacting with the criminal justice system as well as with all agencies designed to help. Current interactions with law enforcement and the criminal justice system may be doing more to distance mistreated elders from the system than to engage them. As already mentioned, research has also shown that the current elder mistreatment laws and policies have not effectively remedied the situation confronting older adult victims of mistreatment. One view has been that law enforcement, APS, and other agencies should intervene only when to do so would lead to an outcome desired by all (Crystal, 1987). For many researchers and professionals (Payne, 2001), the obvious solution is to provide police officers with the training on how to effectively intervene in the situations encountered when working with elders who are victims of mistreatment. Payne found that most cases were still not recognized as mistreatment nor was the role of adult protective services understood by law enforcement. One question that emerges is 40 whether or not ageist attitudes factor into law enforcement response toward older adults and if ageism contributes to elders' dissatisfaction with law enforcement interventions. Police Training As discussed above, studies have shown that even though laws have been created and police are aware of the problems, they are not familiar with the laws. Plotkin (1988) found that a third of all police departments surveyed were unaware of the laws affecting issues related to elder mistreatment as well as the number of cases within their jurisdiction. As law enforcement became more involved in addressing the emerging issue of elder mistreatment, many of those involved in the field supported the need for developing training programs to bring this branch of the service delivery system up to speed. Byers and Hendricks (1993) argued that such training programs would be the most effective approach to overcoming the obstacles encountered among law enforcement personnel. Blakely and Dolon (2000) found that more than 80% of the police departments and sheriffs who responded indicated that adult protective services workers had never provided any training on elder mistreatment or related topics to personnel in their organizations. From this statistic, it may be inferred that law enforcement personnel have an interest in training. According to Arcuri (1981), the need for law enforcement training was evident long before elder mistreatment was criminalized. The author observed that, prior to laws criminalizing elder mistreatment, law enforcement chose not to focus on elder mistreatment as a unique problem and rather viewed it as part of a constellation of problems associated with older adults. He argued further that at the time, police did not 41 understand the unique needs of older adults or the importance of redefining the role of law enforcement to that of a service orientation. It is important to note that in addition to anecdotal support there is some limited evidence that training has been effective in improving law enforcement response to elder mistreatment. Davies (2010) examined both law enforcement understanding of and response to elder mistreatment. In this study, participants responded to questions related to case scenarios that assessed both their knowledge of elder mistreatment statutes as well as how they would respond to various cases. One of the questions asked of participants was how much training they had received (in hours) and whether they believed that they were adequately trained. There was a significant positive correlation between selfreported training hours and how appropriately they responded to the elder mistreatment scenarios. Unfortunately, the small sample size, n = 56, and the use of purposive snowball sampling limited the generalizability and power of the study. Although researchers have consistently recommended training as the most effective mechanism to improve law enforcement response to elder mistreatment, adequate funding for training or other strategies for dealing with elder mistreatment has not been forthcoming. The problem of mandates without funding was well established by the mid-80s, and funding for elder mistreatment prevention has lagged behind other categories of mistreatment (Anonymous, 2002a; Payne, 2002). If authorities are aware of the problem and the laws and statutes necessary to address such problems exist, the lack of funding and training may be significant enough to effectively undermine all efforts directed at the prevention of elder mistreatment. It is imperative then that any funding provided toward the prevention of elder mistreatment be 42 used in ways that have been researched and proven effective. One obstacle to the successful prevention of elder mistreatment is the lack of cooperation between agencies. Dubble (2006), reviewing the progress of interventions aimed at improving elder mistreatment response, concluded that only when there is collaboration between APS and law enforcement will there be effective justice and safety for older adults. More than 20 years earlier, Quinn and Tomita (1986) expressed the need for a multidisciplinary approach in order for professionals to effectively deal with the victims of elder mistreatment. Moreover, Dubble noted that in one major eastern city, prior to such cooperative efforts, no substantiated cases of elder mistreatment were successfully prosecuted. Subsequent to the introduction of collaborative efforts between law enforcement and APS units, prosecution of cases increased. The Elder Justice Act At the national level, ongoing efforts to pass the Elder Justice Act led to its passage in March 2010 (Elder Justice Coalition, 2011). A major thrust of this act is to provide additional federal dollars to enhance states' efforts to prevent, detect, prosecute, intervene, and treat all forms of elder mistreatment (O'Shaughnessy, 2010). While the Elder Justice Act authorized over $150 million in federal funding, to date no funds have been appropriated. Thus, there is no difference today in terms of any added protections for elders who have been abused or mistreated (Elder Justice Coalition, 2011; Jirik & Sanders, 2014). One feature of the Elder Justice Act (Dubble, 2006) is the requirement that law enforcement collaborate with health and social service agencies. Many in the field have held that efforts to require or encourage collaboration among the various community 43 agencies that respond to elder abuse remains the most effective, and possibly the only, way to prevent and successfully intervene in cases of elder mistreatment (National Committee, 2005). Consistent with this, some states have attempted to require criminal justice professionals to work with adult protective services in cases of elder mistreatment (Schneider, Mosqueda, Falk, & Huba, 2010; Tatara, 1990). It would seem that an increase in interagency cooperation would correspond with an increase in effectiveness as well as satisfaction. However, a closer look at the changes in perceived efficacy identified by Blakely and Dolon (2000) revealed that even when funding and resources were provided and mechanisms for collaboration established, APS workers continued to have low expectations regarding law enforcement response to elder mistreatment. While 80% of APS workers believed that law enforcement would provide protection for them, the majority held onto the belief that law enforcement would be unlikely to intervene in ways that directly helped older adult victims. Blakely and Dolon (2000) identified several areas in which cooperation was lacking, including arresting perpetrators, completing investigations, communicating information, removing perpetrators from the home, and reporting of cases of mistreatment. Furthermore, respondents to their survey indicated bureaucratic obstacles from the criminal justice system hindered them from accomplishing their purposes. These authors further noted that while police perceived their interventions to be effective, they also reported they were dissatisfied with the outcome when they were unable to take a case through to prosecution and conviction. This may be due to the divergent philosophies discussed previously related to the cause and subsequent interventions for elder mistreatment. As has been noted, adult 44 protective service workers tend to respond in a manner consistent with the caregiverstress model when intervening with elders and their caregivers. On the other hand, the emerging criminal justice perspective has leaned more toward the caregiver-dependency model, which focuses on the characteristics of the offender. This model views elder mistreatment as a result of criminal misconduct no different from any other type of crime and justifies responses consistent with the goals of deterrence (Dubble, 2006). It becomes evident to those who are concerned with effectively intervening in elder mistreatment and ultimately preventing its occurrence that attitudes held by law enforcement officers remain at odds with those of other social service providers such as APS. These differences remain an obstacle to effective intervention and deserve further attention. Police Attitudes A number of factors can affect how law enforcement responds to elder mistreatment. One issue that has received limited attention in the literature is how law enforcement personnel's attitudes toward older adults influence their response when intervening in cases of elder mistreatment. Several studies (Bayley & Bittner, 1996; Daniels et al., 1999; Goldstein, H., 1996; Goldstein, J., 1996; Reiss, 1996) have shown the importance of officers' attitudes that determine the handling of elder mistreatment cases and which services victims may ultimately receive. As previously mentioned, law enforcement officers have become an important resource for older adults victimized by crime. However, the fact that these officers are mandatory reporters and are still reluctant to report elder mistreatment is cause for concern. Daniels et al. (1999) suggested that, where statutes are vague and front line 45 responders have a great deal of discretion in whether or not to report these cases, their attitudes toward reporting have a significant bearing on whether such cases enter into the system. To better understand this phenomenon Daniels et al. (1999) identified three different aspects of law enforcement officers' attitudes that in turn affect how they respond to situations involving elder mistreatment: attitudes toward the law, attitudes toward the act of reporting, and attitudes toward how well the reporting system is working. Unclear laws. Beginning with a review of how law enforcement perceives the criminalization of elder mistreatment, Payne (2002) observed that unless law enforcement personnel agree with the laws and are willing to enforce them, statutes and policies in and of themselves mean very little. When Payne and Berg (2003) surveyed police officers, they found respondents believed that elder mistreatment should be a priority in law enforcement. If police attitudes toward laws against elder mistreatment are favorable, what then leads to such low levels of reporting? One consideration is that although police support the laws against elder mistreatment, the language of the statutes is often written in a style consistent with human services and not readily understood or interpreted by law enforcement personnel (Wright, 1996). This can be quite frustrating for law enforcement and is compounded by the lack of definitional clarity already noted. Further, statutes that do not clearly define the subtleties and nuances of elder mistreatment have been interpreted differently by intervening agencies and thereby create conflict when these agencies attempt to work together (Dubble, 2006; Payne 2003). Heisler (2000) stated that the direction of elder 46 mistreatment cases continues to go toward civil rather than criminal response, where the solutions are redress rather than criminal sanctions. Again, social policies that made elder mistreatment a civil matter rather than a criminal one effectively reduced law enforcement personnel to the status of mandatory reporters and prevented them from responding in the manner they deemed most appropriate. When law enforcement officers have been asked how they would respond to elder mistreatment, it has been shown that they support incarceration more than other groups (criminal justice, social service, nursing home professionals) for offenses against older adults (Johnson, Sigler, & Crowley, 1994; Payne, Berg, & James, 2001). Consistent with this finding, police officers regularly suggest that they are frustrated by the justice system when they witness offenders who are arrested being punished too lightly or released with no punishment at all (Payne & Berg, 2003). For this reason, if police officers do support elder mistreatment laws, they are likely to experience dissatisfaction if there is not enough support to intervene in the situation once the police officer has made an arrest. Blakely and Dolon (1989) documented feelings of frustration among law enforcement officers as they attempted to collaborate with APS workers in cases of elder mistreatment. While law enforcement has expressed frustration with the way in which society has responded to elder mistreatment, studies have shown that the response by law enforcement does not take into account the unique needs and laws regarding elder mistreatment. Klein et al. (2008) reported that the response by law enforcement to crime was the same regardless of the age of the victim. In other words, their response patterns showed no unique treatment of elder mistreatment cases regardless of the different 47 guidelines provided by the various statutes. According to Payne, Berg, and Toussaint (2001), a survey of police chiefs revealed that the majority reported that their organizations used traditional techniques and methods in response to cases of elder mistreatment. Payne et al. found that even when statutes and policies changed, the only difference in law enforcement response was that they were more likely to coordinate with other community social service agencies. The end result was that even though elder mistreatment laws were in place, police often tended to resort to past practices and to not acknowledge changes in the elder mistreatment laws. When their responses did vary, it was to coordinate with agencies that then did not respond in a manner familiar to law enforcement officers, which only contributed to their frustration. As noted above, the lack of a response that incorporates the values and assumptions of law enforcement officers leads to dissatisfaction with how society responds to mistreatment of older adults. Changes in reporting. As already discussed, police officers underreport cases of elder mistreatment. Several researchers have considered the lack of adequate training (Blakely & Dolon, 2000; Plotkin, 1996; Sever & Youdin, 2006; Stiegel, 2006) with respect to the laws governing elder mistreatment. Payne, Berg, and James (2001) suggested that prioritizing criminal behaviors as independent criminal acts rather than mistreatments of older adults changes the way one views mistreatment. Police officers may see robbery as a more serious offense and a slap as less serious, but others thinking in terms of elder mistreatment may consider a slap an act of mistreatment (Payne, Berg, & James, 2001). Some researchers feel that police should report both criminal and noncriminal forms of mistreatment (Jogerst, Daly, Dawson, Brinig, & Schmuch, 2004; 48 Plotkin, 1988). Nationwide, appropriate elder mistreatment statutes, which require reporting patient mistreatment, neglect, and financial exploitation, are lacking (Hodge, 1998). Also, because of the lack of clarity as to the autonomy of elders and how mistreatment should be approached, there are still questions related to elders' right to self-determination and due process surrounding mandatory reporting (Heisler & Quinn, 1995). How well the reporting system is working. Payne, Berg, and James (2001) found that a significant number of police chiefs who responded to their survey acknowledged that they did not expect to successfully follow through with cases of elder mistreatment. These law enforcement personnel expressed concern over difficulties interviewing older adults with disabilities and other impairments such as dementia. Older adult victims for this and other reasons were not seen as good witnesses. These authors observed that for some police chiefs there was a sense that such cases could not be successfully prosecuted so there was little point following through all of the steps necessary to report elder mistreatment and move the case through the system. This is consistent with an earlier study (Forst & Hernon, 1985) that observed prosecutors were concerned about the possibility that victims would not press charges. Chermak (1993) observed that prosecutors were also concerned with how credible the witnesses would be due to mental impairments and tendencies to become confused as well as whether older adults would be overwhelmed by the court process. This was supported by Dunlap et al. (2007) when they found that older witnesses were less likely to be perceived as credible. Nerenberg (2002) noted that prosecutors are often reluctant to work with "witnesses" who have cognitive or communication impairments. This leads 49 to a situation in which police are reluctant to make arrests and file charges knowing that the case will most likely not be prosecuted. Although the three aspects identified by Daniels et al. (1999) that affect law enforcement attitudes have been researched, one important aspect that has been overlooked is that of possible bias based on age, otherwise known as ageism. Ageism and Law Enforcement Discrimination Because ageism is a form of discrimination that may affect law enforcement response to elder mistreatment, it is beneficial to examine what role discrimination in general plays in police response. The body of research concerning profiling with respect to racism may be instructive in understanding what role ageism plays in law enforcement response to elder mistreatment. Smith and Alpert (2007) examined the interaction between stereotyping and racial profiling and concluded that there is less support for the idea that police are intentionally profiling racial minorities. They hypothesized that the mechanisms that create and maintain stereotypes can lead to unconscious behavior toward minorities. They found that while police stop and search minorities more than nonminorities, there is no observable bias on the part of law enforcement officers in their stated perceptions or behavior outside of law enforcement. This finding supports Nelson's assertion that stereotypes are so automatic and universal that the process is simply done unconsciously (Nelson, 2009). This is compounded because research does indicate that minorities do in fact commit more crimes (Smith & Alpert, 2007). Law enforcement personnel, with their knowledge of increased crime among minorities, easily fall back into their stereotypical views when dealing with perceived crime. And while 50 there is no perceived bias outside of their criminal purview, inside their law enforcement role, officers are harsher toward minorities. It has been shown that prejudice against other populations affects the discretionary outcomes of police decisions (Smith & Alpert, 2007). Sherman (1996) and Daniels et al. (1999) suggested that the primary influence on police discretion appears to be the attitude of police officers. Similarly, earlier research examined the attitudes of law enforcement officers toward older adults, finding that law enforcement personnel had mixed feeling with respect to dealing with cases involving this population. Sykes (1977) found that law enforcement officers have more negative feelings toward older adults. As part of his research, he found that older adults are twice as likely to seek noncriminal-related services proportionate to the overall population. According to Sykes, this may have been caused in part by what may be perceived as frivolous cases for which older adults sought assistance. This study also observed that older adults are far more likely to be disrespectful or hostile than are younger persons (also cited in Blakely & Dolon, 2000). A study conducted by Goldsmith (1976) found that more than half of police departments surveyed indicated that older persons called repeatedly to register the same complaint and made fictitious reports and that law enforcement personnel believed that a significant portion of these reports were motivated by loneliness. In their discussion of this study, Blakely and Dolon suggested that law enforcement officers during the 1970s may have viewed older adults as "unpleasant nuisances" that distracted officers from focusing on what they perceived as their real purpose-responding to crime. Surveys conducted in Alabama suggested that police knowledge and attitudes about elder mistreatment strongly influenced the officer's ability to identify and report 51 mistreatment (Daniels et al., 1999). Many police officers do not understand the unique social, economic, and physical problems that affect older adults (Arcuri, 1981; Nerenberg, 2002). When those intervening with elders do not have an accurate perception of elders and their unique circumstances, bias in the form of ageism is more likely to play a significant role. This has been verified by Daniels et al. (1999), who acknowledged law enforcement officers' responses have changed based on the amount of exposure they have had to older adult population. According to Daniels et al., officers who had more frequent contact with older adults were actually more likely to become disillusioned with the criminal justice system's response to elder mistreatment and less likely to rely on the system to provide an effective response to cases of mistreatment once they were entered into the system. Effect of the Media As previously mentioned, media plays a role in formulating society's view of older adults, sometimes positively (wealthy, golden years) and sometimes negatively (feeble, cognitively impaired). Stereotyping is difficult to eliminate when it seems to be supported by data, such as older adults are more likely to be cognitively impaired, which has then resulted in fewer efforts to prosecute allegations of mistreatment because it is assumed elders will be ineffective witnesses. When law enforcement officers' experiences working with older adults confirm stereotypical beliefs, the beliefs persist. Because elders do possess attributes consistent with stereotypes, such as cognitive impairment, which may cause confusion and physical limitations such as difficulty hearing and seeing, ageist behavior seems justified. Part of the response to this information is that some police and prosecutors have created special units specifically 52 trained to deal with older adults. Such units have more consistent-even exclusive- contact with older adult victims and can then make more effective assessment of their needs (Quinn & Heisler, 2004). While this recommendation may help to improve the situation, it is a far cry from solving the underlying issue of ageism. Summary The findings from research, government, and agency sources indicate that elder mistreatment is a significant problem. The exact numbers are not known, but what is known is that the percentage of older adult population is increasing dramatically and the cost of elder mistreatment is substantial. Society's initial response to this mistreatment was to create mandatory reporting laws. Adult protective service agencies were created to receive, investigate, and resolve reports. These agencies relied more on the caregiverstress model to understand the problem and focused on providing support for caregivers and problem resolution, including redress in the civil court system. The number of reports of elder mistreatment nationwide increased 85% from 1993 to 2004 (Teaster et al., 2006). The caregiver-stress model has not explained the data and has not led to a successful resolution of the problem. Continuing research indicates that those who perpetrate elder mistreatment, like all other criminals, are motivated by greed, power, and control (Brownell, 1998; Quinn & Heisler, 2004). Legislation responding to changing perceptions of elder mistreatment led to criminalizing the mistreatment of elders, increasing penalties for nonreporting, and eventual enhancement of penalties. Throughout this legislative process, there has been a lack of funding to support elder mistreatment prevention. Laws have been put into place without provision for implementation. 53 In spite of all of the policies established and efforts made, the system has not been responsive to the needs of older adult victims. As a consequence, victims of elder mistreatment have not had favorable experiences. Some elders who report mistreatment want to prosecute. Police are the gatekeepers most often turned to by older adults, but the police, at least half of the time, do not report mistreatment when they encounter it and often do not recognize mistreatment when it occurs (Daniels, Baumhover, Formby, & Clark-Daniels, 1999). Legal statutes have traditionally defined the role of APS workers and have not been written for the criminal justice system. Historically, law enforcement did not refer to APS and APS avoided law enforcement. One explanation for this is that law enforcement and APS workers do not agree on what is the best way to respond to elder mistreatment. When the law enforcement role is not clearly defined, police officers fall back on old traditional crime prevention strategies and disregard elder mistreatment laws. From the vantage point of ageism, the history of, and response to, elder mistreatment can then be seen in a new light. By providing an understanding that older adults is a label that is socially constructed and represents a group that society wishes to avoid and with which other groups in society compete for scarce resources, ageism then readily explains why there is limited to no funding for elder mistreatment. Society readily accepts the stereotype of the greedy geezer who just wants all he can get without concern for anyone else. If older adults are stereotyped as obsolete with nothing to contribute to society, it makes little sense to use scarce resources to better their existence unless those resources help to further segregate older adults from the rest of the population. From the viewpoint of terror management theory, therefore, it can be 54 expected that APS workers as well as law enforcement would choose to interact with caregivers or perpetrators and avoid contact with older adults and the reminder of their own mortality. Both of these individuals may be seen as more worthy and less overwhelming to help than the stereotypical elder. This explanation may clarify the philosophical differences between these police and other social service agencies. By choosing to work with perpetrators or caregivers, responders are able to avoid focusing on the realities of older adults and their own aging and instead interact with individuals they either identify with (caregivers) or persons they clearly see as different from themselves (criminals). Ageism offers an explanation as to why laws were created without understanding the elder and instead substituting "elder" for "child" in the existing child mistreatment laws. Ageism explains why prosecutors do not prosecute, why crimes against older adults are not recognized as valid crimes, and why crimes against more socially acceptable victims are more likely to be successfully resolved. In choosing to avoid elder mistreatment, prosecutors and law enforcement are able to avoid interacting with people who are perceived as much more difficult to work with and who remind them of their own inevitable aging process and death. In light of ageism, it is rational for police not to want to fill out reports and make the effort to move a complaint through a system that is not going to respond because of its overall bias. It is no surprise that elders are unhappy with the system once they discover that it is not responsive to their needs. Ageism prevents it from being responsive. To be an elder is to be an "other" outside of society's mainstream and incompatible with its objectives. The conflict perspective emphasizes a fundamental change is needed to address 55 the challenges of integrating and valuing older adult members, as opposed to keeping them in the position of "other." With this in mind, it is worth looking at ageism within law enforcement as one aspect of this broad ubiquitous problem. Given the literature suggesting that ageism and all of its permutations may affect how law enforcement responds to incidences of suspected elder mistreatment, it was reasonable for UCOA to survey law enforcement officers with regard to ageism. The first question to be answered was whether ageism is significantly correlated with appropriate responses to cases of elder mistreatment. Some of the factors also addressed by the UCOA survey and which were thought to moderate this effect include age of officer, training, years on the force, level of contact with older adults, and level of education. By analyzing the data collected, it was hoped significant relationships would be identified between ageism and law enforcement response to elder mistreatment that would establish the value of assessing ageism among law enforcement personnel as well as other social agencies. As Johnson (2002) stated, Education is needed to bring the problems associated with elder mistreatment, neglect, and exploitation to the public eye; to diminish myths of aging; and to enable legislators and policy makers to make appropriate decisions regarding needed legal and administrative action to address health, legal, and social problems resulting from elder mistreatment. (pp. 185-186) Education cannot occur without first understanding the problem. Payne, Berg, and Toussaint (2001) understood that it was important to continue to observe the police response to elder mistreatment to assess whether policies and practices are effective in overcoming elder mistreatment within our society. CHAPTER 3 METHODOLOGY My study is an analysis of data from a survey conducted by the Utah Commission on Aging (UCOA). Consequently, there are inherent limitations due to the methodological choices made by the UCOA Safety Committee when they designed and implemented the survey. Nevertheless, the use of the UCOA database was justified because of the unique information it contains and the critical insight it can provide to the criminal justice system and adult protective services regarding the response of law enforcement officers to elder mistreatment. Specifically, the data collected by UCOA, in conjunction with the Utah Association of Sheriffs, are unique due to the comprehensive access that was given to a population not typically accessible, sheriff officers in Salt Lake County in the spring of 2008. The UCOA survey elicited responses to questions that might further understanding regarding the underreporting, lack of arrest, and prosecution of cases of elder mistreatment. The survey focused on how, as first responders, law enforcement officers' responses to elder mistreatment may be affected by limited knowledge of and biased attitudes toward older adults. One specific consideration for the UCOA study was to learn whether knowledge of older adults and the aging process affected officers' responses to situations of mistreatment of older adults (personal communication, UCOA executive committee, August, 2008). Surprisingly, no data analysis was conducted upon 57 this data other than simple descriptive statistics. As a result, the UCOA data were made available for this dissertation with the hope that review could provide additional analyses and insight into the issues related to law enforcement response to elder mistreatment. This chapter briefly describes the methodology used by the UCOA in conducting the survey of the Utah Sheriff's Association, including research objectives, design, sample recruitment, data collection instrument and procedures, and ethical considerations. Next, strengths and limitations of the methods chosen by the UCOA researchers are reviewed, which allows discussion of the inherent limitations introduced to this study. Finally, the chapter focuses on the methodological strategy of this study used for conducting the analysis of the UCOA data. UCOA's Methodological Choices UCOA used a survey approach to collect data that explored factors thought to influence law enforcement officers' response to cases of mistreatment of vulnerable adults. The questions included in the UCOA survey examined officers' knowledge of and experience with elders and vulnerable adults. More specifically, the questions asked about officers' knowledge of the legal definitions and familiarity with statutes associated with this population. Survey participants were then asked to respond to four case scenarios involving vulnerable adults, two of which met the criteria for elder mistreatment (personal communication, Utah Commission on Aging, Executive Committee, August 2008). Additionally, an embedded scale [Facts on Aging Questionnaire (FAQ1)] measured officers' knowledge of older adults and the aging process. The FAQ1 is also thought to provide a rough measure of officers' bias towards older adults (Palmore, 1998). These factors were targeted by the specific methodological 58 choices made by the UCOA survey designers. The UCOA Survey Sample Recruitment In spring 2008, Salt Lake County sheriff officers were invited to complete the UCOA study. The questionnaires were given to ranking officers who in turn distributed them to those officers under their command. Two-hundred surveys were completed by law enforcement officers. The response for the sample was roughly 200 out of a possible 275 officers, or a 73% return rate. However, it is unclear under what circumstances each ranking officer distributed or administered the surveys, whether there were any incentives or penalties for participation and completion, and whether there were any follow-up efforts made to ensure that all officers had an opportunity to participate. Consequently, in my data analysis, I was cautious in interpreting the findings, in that this may have introduced bias. Nevertheless, the substantial response rate made response bias less likely, and the information provided a reasonable basis for a provisional understanding of how sheriff officers in Utah respond to cases of elder mistreatment. The UCOA Survey Development Procedures The UCOA Survey section one-participant experience training and familiarity with vulnerable adults. The data collection instrument was developed by the members of the UCOA's Public Safety Work Group including Tanta Lisa Clayton (the chair) and Wendy M. Thomas (then a graduate student in the University of Utah's Gerontology Interdisciplinary Program) (personal communication, August 2008; December 2011, respectively-refer to Appendix B). The first part of the instrument consisted of six questions regarding training experience, attitudes toward that training, and familiarity 59 with the definition of vulnerable adult and the Utah statute relating to mistreatment of vulnerable adults. The questions specifically addressed 1) officers' training regarding the vulnerable adult population, 2) when the training took place, 3) officer familiarity with legal definitions of vulnerable adults, 4) familiarity with Utah Statute 76-5-111 (abuse, neglect, or exploitation of a vulnerable adult), and 5) application of 76-5-111. The sixth question further established the officer's understanding of the definition of a vulnerable adult. It should be noted that while there were limitations concerning the validity of this aspect of the survey, there was evidence of construct and face validity because questions were created and selected by a committee comprised of law enforcement personnel, APS workers, and persons with legal backgrounds and expertise in the field of elder mistreatment. While the questions were useful and offered answers to the immediate questions posed by the developers, the fact that they had not been normed or validated with law enforcement personnel or any other population imposed further limitations that will need to be verified by future research. Nevertheless, there were no obvious problems with the validity of the questions. Data analysis therefore provides insight into how ageism affects officers' response to cases of elder mistreatment while recognizing that the data from the survey remain open to challenges by future researchers regarding validity. Another concern is that the content of the initial section conveyed information that directly affected how participants would respond to subsequent questions. For example, the statutes pertaining to vulnerable adults were identified in this first section. Because of this, officers may have ascertained the proper response to each of the scenarios regarding vulnerable adults found in the next section regardless of their 60 previous training. This may have led to a systematic bias reflecting social desirability or self-preservation. Understanding that the questions in the first section may have biased officers' responses to the scenarios, any interpretation of the data required the recognition of this limitation and how it may have skewed officers' responses. In other words, the data may reflect how an officer would respond immediately after being given a cue and not necessarily how they would respond in the field without such a cue, thereby opening the findings to challenges regarding validity. Therefore, although this analysis provides an important understanding of the effect of ageism on police response to elder mistreatment, future researchers may wish to challenge the assumption of validity of the responses, despite high face and construct validity of the questions. The UCOA Survey section two-vulnerable adult/mistreatment scenarios. In this section, officers were presented with case scenarios that described different types and levels of mistreatment. The four case scenarios, two of which described cases of elder mistreatment, measured law enforcement officers' response to vulnerable adult mistreatment. The case scenarios were developed by the members of the UCOA and were based on actual situations familiar to members of the UCOA Public Safety Work Group (UCOA Public Safety Work Group, personal communication, August, 2008). Because the scenarios represented actual cases, the committee considered them to be valid measures for assessing how law enforcement officers respond to situations that contained elements readily identifiable as mistreatment of vulnerable adults including older adults. (See case scenarios in Appendix C.) Once again, the case scenarios have a measure of face and expert validity, but have not been validated through other means. Moreover, no reliability data were 61 available as this was the first time they had been used. The survey piloted the use of the four scenarios, and subsequent analysis will hopefully help to establish the utility, validity, and reliability of the scenarios and whether they can be used effectively in future studies. Moreover, no criteria were established a priori for evaluating the responses to the scenarios, making it necessary to create a systematic method for analyzing this portion of the data, which were qualitative in nature. While these limitations affected the ability to generalize findings, the UCOA dataset provided a unique opportunity to gain an in-depth understanding as to what factors influenced officers' responses to scenarios of elder mistreatment based on responses to open-ended questions. The UCOA Survey section three-Facts on Aging Quiz 1 (FAQ1) instrument. The third section of the instrument used an imbedded measure, the entire multiple choice version of Facts on Aging Quiz 1(FAQ1) created by Palmore (1980-see Appendix D) and refined by Harris, Changas, and Palmore (1996). The primary purpose of the original FAQ1 as stated by Palmore was to measure the accuracy of knowledge held by respondents concerning older adults and the aging process. The original Facts on Aging Quiz (FAQ1) is available in a true/false format and a multiple choice format, the latter consisting of 25 items. The primary score is obtained by simply summing the number of correctly answered questions, with a range from 0-25, reported as a percentage. Palmore (1998) went further by suggesting that the FAQ1 could be used as an indirect measure of bias toward older adults because, as part of its unique design, the FAQ1 assigns either a positive or negative score to each response depending on whether the belief reflected is pro-aged or anti-aged bias toward older adults. Thus a positive overall score reflects either an inaccurate or unrealistically pro-aged belief about older 62 adults, whereas a negative score would suggest an inaccurate and anti-aged response that negatively characterizes older adults. For example, one item begins, "Most workers over 65:" with choices including, " a) work less effectively than younger workers; b) work as effectively as younger workers; c) work more effectively than younger workers; or d) are preferred by most employers." (See Appendix A, item 28.) Choosing "c" or "d" would be inaccurate and reflect an unrealistically positive view of, or bias toward, older persons, and either choice would receive a score of +1. Choosing response "a" is also inaccurate but reflects a negative view or bias toward older adults and would be scored as -1. The choice of "b" is accurate and receives a 0 score. These biases reflect misconceptions or commonly held stereotypes about older adults and aging (Palmore, 1998). For incorrect responses on the FAQ1, 10 items have the possibility of only a negative Bias score, and five yield only positive scores. For eight of the questions, both a positive and a negative score are possible (two items have no positive or negative valence-see Appendix D). The Bias score was obtained by subtracting the percentage of incorrect negative or anti-aged bias responses from the percentage of positive or proaged bias responses. This yielded a percentage that could be either positive or negative between -1 and 1. In this way, officers' negative biased responses would be offset by their positive responses. Because this scoring procedure allowed pro-aged beliefs to cancel out anti-aged beliefs, persons who hold a significant number of inaccurate positive views may still have a relatively low overall Bias score if they also hold an equally high number of negative views toward older adults. Palmore (1999) attempted to account for this weakness by considering only those scores which fell below -.20 or above .20 as significant. Using 63 this adjustment resulted in a significant loss of information for the Bias score, and so the net bias percentage was chosen as the measure representing bias instead of the cutoff score. The use of the Bias scores was better understood by examining the literature regarding the FAQ1. Palmore (1998) originally designed the FAQ1 to be used in his courses on aging and gerontology at Duke University in order to generate interest and discussion surrounding the topics of aging and older adults. It has subsequently been used in a wide variety of studies (Palmore 1998). According to Palmore, another use of the FAQ1 is as an edumetric that can measure the change in knowledge following an educational experience designed to increase knowledge on aging and is the only established instrument for this purpose. The instrument focuses on various aspects of the aging process and general knowledge regarding older adults, as well as common misconceptions. Research using the FAQ1 has consistently found that scores increase as education increases (Palmore, 1998). Palmore (1998) found that individuals who have taken a course on gerontology consistently score higher on the FAQ1 than those who have not, supporting the use of this instrument as an edumetric. A study by Miller and Dodder (1980) supported Palmore's assertion for the use of the FAQ1 as an edumetric. Norris, Tindale, and Matthews (1987) disagreed, finding that the FAQ1 had only moderate validity and reliability as an edumetric and was therefore not suited for use with research but rather only in educational settings. Regardless of these findings, the FAQ1 and its subsequent variations (FAQ2) have been used in more than 150 research studies to examine various aspects of knowledge of older adults as well bias towards them (Palmore, 1998). 64 As previously discussed, Palmore (1998) suggested that the FAQ1 may be used to detect age bias. Aushermen, White, and Chenier (1991) observed that Knowledge scores on the FAQ1 correlated well with another measure of bias toward aging and older adults. However, not all researchers agree. Holtzman and Beck (1979) argued that the FAQ1 does not measure attitudes toward older adults. Despite these differing views, the FAQ1 continues to be widely used in research, primarily as a measure of change of knowledge following an educational experience and as a secondary measure of age bias. Part of the reason for this is that other instruments that have attempted to measure knowledge of older adults require much more time to administer and have not been demonstrated to be either more valid or reliable (Palmore, 1998). While the controversy continues as to whether or not the FAQ1 can be used as a reliable and valid measure to accurately assess knowledge of or bias towards older adults, the instrument remains the most widely accepted and frequently used tool for assessing either purpose. This study assumed that the FAQ1 Knowledge score is a valid and reliable measure of whether respondents' knowledge of older adults is accurate and, therefore, is a useful proxy measure for ageism as a predictor of how law enforcement officers will respond to scenarios of elder mistreatment. The UCOA Survey section four-potential moderating variables. The last section of the instrument explored other variables that are thought to explain some of the variance in the dependent variable, response to scenarios of elder mistreatment, including age, gender, number of years in law enforcement, highest rank, and level of education. 65 Ethical Considerations Regarding the UCOA Survey Because the UCOA survey was designed and distributed as an internal study, the original researchers did not deem it essential to seek informed consent (UCOA Safety Committee, personal communication, August, 2008). According to the commission members, the surveys were distributed and collected in a manner generally consistent with the culture of a community law enforcement agency for the purpose of fact finding and information gathering and were treated as part of the regular course of organizational functioning. Senior officers responsible for survey distribution had differing strategies for circulating surveys, which may have led to some systematic bias. Understanding that officers responsible for dissemination of the surveys may have known or been able to ascertain which officers returned surveys, particularly in smaller agencies, may have ethical implications for the original study. These factors may have influenced how officers responded and may have biased the data, limiting confidence in the validity and requiring additional caution in the interpretation of any findings. The relatively high response rate as well as overall numbers should have reduced this concern, and future researchers will have to decide whether and how such biases can be overcome. While some concerns existed regarding the manner in which the surveys were distributed and collected, there were no ethical concerns in the present study. The dataset given to this researcher consisted of the raw surveys, which were completely deidentified from the original participants prior to their being released to the principle investigator. This further mitigates the concern about the confidentiality of the participants with respect to the present study. Participants from the original UCOA study were not contacted during the present study; thus there was no additional risk to anonymity. Also, recognizing the UCOA sought out this data analysis and understood that the results 66 would be made available to them prior to any publication further mitigates concern over ethical issues. The University of Utah Institutional Review Board granted approval January 9, 2012 to conduct the data analysis. Limitations of the UCOA Survey Affecting the Data Analysis Perhaps the most profound limitation of the survey, other than those already mentioned, is the amount of missing data. This required the use of statistical techniques to overcome this limitation. Another concern, noted above, is that the order of the questions may have biased responses to the case scenarios. The survey information contained in the section Participant Experience Training and Familiarity with Vulnerable Adults, provided information that could and perhaps should have been inferred as directly pertaining to the case scenarios in the subsequent section. Nevertheless, because so much research has recommended training as a way to improve police response to elder mistreatment, the measure of self-reported training was included. The opportunity to access self-reported training and then evaluate how this might affect a response to a scenario was unprecedented. Justification and Current Research Objectives Given the Identified Limitations To date, only one study (Davies, 2010) has directly examined the relationship between law enforcement officers' knowledge of elder abuse statutes and how they respond to cases of elder mistreatment. Davies explored the relationship between training specific to elder abuse (which included both domestic as well as institutional abuse), understanding of elder abuse and how to respond to it from a law enforcement perspective, and responding through scenarios. Results served to assess how effectively 67 or appropriately officers applied their knowledge and understanding to case scenarios of elder abuse. The principle findings were that self-reported training was a significant predictor of knowledge or understanding of elder abuse and how well law enforcement officers responded to elder abuse scenarios (Davies, 2010). Furthermore, as the amount of elder abuse specific training reported by respondents increased, so did the likelihood of an appropriate response. While Davies' study was encouraging, there were several significant limitations. It was conducted with a purposive, snowball sample of only 56 usable responses. The study was also limited in that it looked narrowly at the correlation between training specific to elder abuse and appropriate response by law enforcement. In addition, the scenarios were designed to be consistent with California legal statutes and do not necessarily translate well to other populations or settings. Another study (Cupello, 2010) drew from the same UCOA dataset used in this study but did not make use of the police officer scenario responses. What is more, no study to date has examined the relationship between knowledge of aging and older persons and how this might influence police officer response to scenarios of elder mistreatment. While the responses to the survey may not be generalizable beyond the specific population, they provided an opportunity to explore this issue in a cost-efficient manner. Such analyses not only provide direction for future studies, but also help determine whether ageism exhibited by law enforcement officers contributes significantly to the underreporting and lack of prosecution of cases of elder mistreatment. The Cupello study examined the relationship between the demographic variables and officers' knowledge about aging, as measured by the FAQ1, and found no significant 68 correlations between the demographic variables such as age, years in law enforcement, level of education, or rank, and the scores on the FAQ1 (Cupello, 2010). Further, Cupello found that police officers' knowledge of older adults was consistent with the broader society. But, contrary to Palmore's assertion that education improves FAQ1 scores (Palmore, 1988), Cupello found that increased levels of education had little effect on police officers' knowledge or understanding of older adults. The present study went beyond Cupello by examining the interaction between the FAQ1 and self-reported training with officers' response to the case scenarios. The most important reason for using the data provided by the UCOA survey is that it is the only known effort that directly or indirectly addresses issues related to ageism and law enforcement response to elder mistreatment. Despite flaws, the data provide an inexpensive and efficient means for exploring the questions raised in the literature regarding ageism and the influence it may have on police response to elder mistreatment. In summary, the UCOA study provides a unique opportunity to examine the effect of ageism and elder abuse specific training on law officers' responses to elder mistreatment. While Davies (2010) examined the link between training and law enforcement response to elder mistreatment, she did not tackle the broader issue of ageism. The Cupello (2010) study explored how several demographic variables affected officers' knowledge of older adults as measured by the FAQ1 but stopped short of an examination of how these factors may affect officers' responses to elder mistreatment. This study was the first to explore the relationship between ageism and law enforcement response to elder mistreatment and how this contributes to society's overall ineffective 69 response to the growing problem of elder mistreatment. By including the effect of elder abuse specific training, this study was also able to explore the interaction between training, age bias, and officer responses to elder mistreatment. Thus this research extended beyond these two studies by examining both ageism and knowledge of older adults and explored how these interact with efforts to train officers to respond effectively to elder mistreatment. A Study of the UCOA Dataset Using Linear Regression to Determine the Effect of Ageism on Law Enforcement Response to Elder Mistreatment The primary objective of this study was to determine whether ageism, as measured by the FAQ1 Knowledge score, affected police officers' response to cases of elder mistreatment. Specifically, the objective was to determine whether the dependent variable, officers' response to the elder mistreatment scenarios, was predicted by the explanatory variable, Knowledge, as a proxy for ageism, through linear regression analysis. Given the limitations of the FAQ as a measure of bias, the FAQ1 Bias score (Bias) was included in the regression analysis along with the Knowledge score (Knowledge) to assess whether it would increase the ability of the FAQ1 in discriminating between those who respond appropriately to the elder mistreatment scenarios and those who do not. The secondary objective of the regression analysis was to explore whether or not training specific to the elder abuse statutes would override inaccurate beliefs, as measured by the FAQ1, that otherwise would lead to ageist behavior. Training was expected to have a moderating effect on officer response. Training has consistently been 70 mentioned as a mechanism for changing how police respond to older adults. Thus, the regression model explored the correlation between ageism and elder mistreatment and the effect of training as a potential moderating variable. Use of the FAQ1 Knowledge Score The use of the Knowledge score, in conjunction with police officers' stated responses to elder mistreatment, allowed for the analysis of whether inaccurate beliefs of older adults led to inappropriate responses to older adults as well. Because this is precisely the knowledge Palmore sought to measure with the FAQ1, understanding the correlation between the FAQ1 and police response was an appropriate use of the measure and a rough measure of ageism. The basic concept of ageism includes inaccurate and biased beliefs, or otherwise inaccurate knowledge, which may lead to discriminatory behavior towards older adults. The present study thus accepted the original use of the FAQ1 as a measure of knowledge of older adults and relied on the original scoring method used by Palmore (1998). For purposes of this study, it was assumed that the FAQ1 measured the inaccurate beliefs and knowledge that police officers held about older adults. Therefore, the FAQ1 was used to test if there was a correlation between how officers responded to scenarios of elder mistreatment and their ageist beliefs and biased knowledge. In other words, the FAQ1 was used to test whether differences in knowledge and beliefs regarding older adults led to differences in response to cases of elder mistreatment by police officers. The FAQ1 Knowledge score was chosen as the primary focal variable because of greater, previously proven validity and increased sensitivity to variation found in officers' 71 responses to the elder mistreatment scenarios. Because the FAQ1 yields an additional measure of bias, the Bias score, and because the Bias score is potentially a more direct measure of bias, it was included in the regression to further test the hypothesis that age bias or ageism contributed to how officers responded to elder mistreatment. Because the Bias score has not been found to be either a reliable or valid measure of bias (Aushermen, White & Chenier, 1991; Holtzman & Beck, 1979; Palmore, 1998), it was included only as a tentative and secondary measure of bias in the regression models. Statistical Methods, Data Analysis, and Interpretation Examination of Statistical Assumptions Sample representativeness. Because all members of the population were surveyed and because there was a high response rate (200 out of 275, or 73%) it was assumed that the data from the survey were a good representation of the population of sheriff's officers to whom the study was generalized. Level of measurement. To use linear regression, the data from the dependent variable must be at least interval level. Because response scores had a potential range of 0-6 and there were two different scenarios addressing elder mistreatment, summing the two separate scores for each respondent resulted in a variable with a potential range from 0-12. Rating and scoring the qualitative scenario responses in this manner allowed for range of variation and met the assumptions for interval-level data. Protecting against bias. To ensure that bias was minimized, all variables considered important in the literature and obtained through the survey were included in the full regression model. After addressing cases with missing data, the initial 200 respondents were reduced to a usable n of 123 in the full model. While there is not 72 agreement on the number of observations required for each variable, the number of observations needed for regression analyses ranges from 10 to about 50 (from liberal to conservative) (Abu Bader, 2011; Harrell, 2001; Sweet & Grace-Martin, 2012). The sample size provided enough degrees of freedom within the lower end of the range to include all of the variables in in the full model. Nevertheless, the small sample size did warrant greater caution if there was only a small effect size. A nested model allowed for comparison between the full model and the subsequent restricted models in which variables were removed (Hoffman, 2010). A nested model thus displayed how the removal of variables affected the explanatory power of the model or the R-square value. In the most parsimonious model, only eight explanatory variables were retained. Correcting for multicollinearity. Because there was potential multicollinearity between several of the variables including training and bias, all relationships were tested for multicollinearity and when this was evidenced, selected variables were removed from the model as recommended (Abu-Bader, 201; Harrell, 2001). Variables were also removed when there was no theoretical support for inclusion to reduce the risk of bias. The different training variables presented a risk of multicollinearity because the training questions were similar in nature. When tests were conducted, significant correlation was found between several of the measures. For theoretical reasons, and for consistency with previous research, the variable that represented self-reported receipt of training (Training) was included in the model as a focal variable. When it was found to be have no significance and to have a high degree of correlation with other training measures, it was removed after the second set of variables in the nested model. However, because the UCOA survey qualified the type of training and how recent the training 73 occurred, these additional measures of training were included in the regression analysis to explore whether this additional information might prove useful in understanding how police responded to elder mistreatment scenarios. The last regressions in the nested model included only those training variables that had no problems with multicollinearity and showed significance (Abu-Bader, 2011). Likewise, the literature showed that both age and years of law enforcement experience are considered important in understanding how officers respond to elder mistreatment (Daniels et al., 1999; Davies, 2010; Dolon & Hendricks, 1989; Sever & Youdin, 2006). However, after conducting the full regression, age and years in law enforcement showed high levels of multicollinearity. Therefore the variable Age was removed because of the two variables, Age and Experience, Experience was more central to the questions of the study. Further, the variable, Experience, effectively served as a weak proxy for age and thus retained some of the original meaning. General Study Procedures There was no contact with participants from the original study. Once the review of the literature was completed and approvals obtained, the analysis of data was conducted. This was followed by compiling, reporting, summarizing, and interpreting the findings. The data points that were coded included demographic information such as age, gender, years in law enforcement, highest position served (rank), and highest level of education. The data representing the independent or explanatory variable, Knowledge, reflected the number of correct responses to the Palmore Facts on Aging Quiz (FAQ1), which has 25 multiple choice questions. The variable, Bias, was also derived from the FAQ1 by analyzing the positively or negatively biased responses to incorrectly answered 74 questions for each respondent as identified by Palmore (1998). Furthermore, quantitative and categorical data corresponding to questions concerning officers' training and familiarity with the Utah statutes that address abuse to vulnerable adults were used as measures of training. The data reflecting the dependent variable, Scenario Response, law enforcement officers' responses to scenarios of elder mistreatment, were derived from qualitative responses to two case studies involving mistreatment of elders. Once coded, the rated Scenario Response data constituting the dependent variable were analyzed using linear (OLS) regression analysis to determine how they were affected by the predictor variables, Knowledge (the FAQ1 scores for knowledge), and Bias (net-aged bias ), as well as Training (self-reported training). The strength of a linear (OLS) regression analysis is that it is the most direct procedure for evaluating relationships between interval level data (Hoffman, 2010). The procedure yielded an estimate of the variation accounted for by each of the explanatory variables on the outcome variable, thus determining the best predictors. Dependent Variable The outcome of greatest interest was how police officers responded to scenarios of elder mistreatment. In assessing the law enforcement officers' responses, the main goal was to recognize when officers applied elements of the vulnerable adult statute and explicitly recognized the vulnerability of victims due to their status as elders. Given that each scenario represented a case of elder mistreatment, an effective response would necessarily have acknowledged the vulnerable adult statute as it applied to elder mistreatment. Response appropriateness and completeness were determined based on 75 several factors consistent with the criteria identified in the literature as well as those identified by UCOA and statutory definitions. Response evaluation criteria also included those variations that emerged from the dataset during the process of evaluation consistent with analyzing qualitative data. Because of their subjective nature, the responses were given to multiple experts to evaluate the level of appropriateness of each response. Persons with expertise in law enforcement, social work, and academia evaluated the scenario responses. Each of the examiners reviewed the scoring strategies and looked over all of the data, identifying and clarifying inconsistencies and reaching agreement on each of the scores. Several major factors or elements were identified for evaluating officer responses based on input from UCOA, law enforcement experts, and social work professionals familiar with APS. The vulnerable adult statute requires law enforcement officers to notify APS when they become aware of a case of vulnerable adult mistreatment. Therefore, this element was included as an aspect of an effective or appropriate response included in the rating method. Next, because elder safety and well-being were important to UCOA, this factor was also included in the desired behaviors for an effective response. Further, responses were evaluated in terms of appropriateness of the referral made. Ultimately, the areas identified were whether 1) the case was viewed as civil or criminal and an investigation was recommended, 2) the officer actually investigated or only referred for later investigation, 3) a charge was filed or an arrest made, 4) the need to involve or work with APS was indicated, 5) concern for the safety and well-being of the elder was mentioned, and 6) the vulnerable adult statute was acknowledged. Response scores increased by one point for each of these elements that was explicitly or implicitly 76 acknowledged in a response (see Table 3.1). Because an officer could respond appropriately without explicitly acknowledging the fact that the case involved a vulnerable adult, points were added when responses included appropriate law enforcement actions not unique to elder mistreatment statutes. Evaluators understood that in both scenarios a crime had occurred warranting an investigation, charges, or an arrest. Points were added to the scenario response score for the following recommendation elements: immediate investigation should be conducted, need to arrest perpetrator, or charges were warranted. Therefore, one point was added to the scenario response score if the officer's response explicitly or implicitly recognized the seriousness of the situation. The lowest score, zero, indicated for both cases the officer believed that no referral was necessary, the matter was of a civil nature, or that no law enforcement involvement was required. Higher scores indicated that in each case the officers adhered to principles consistent with good law enforcement and also recognized the need for Table 3.1: Elder mistreatment scenario scoring elements Point value 0 Officer responses Viewed as civil matter / Case not opened 1 Opened case / Referred for investigation (Open/Refer) 1 Investigated (Investigate) 1 Charged or arrested (Charge/Arrest) 1 Offered protection / assistance (e.g., Help contacting family, community resources, etc.) (Help/Protect) 1 Referred to / Involved APS (Involve APS) 1 Recognized vulnerable adult statute applied (VA Statute ) 77 assistance and protection of a vulnerable adult independent of criminal sanctions against the offender. The rating of case scenarios relied on a simple rubric (see Table 3.1) to quantify responses to open-ended questions. In many cases, it could be assumed that if a respondent mentioned one discrete response element, such as charging or arrest, this would presume another element, such as investigation, even though not explicitly stated. Therefore, a response of "arrest" would automatically receive more response score points than someone who stated they would contact APS because of the additional elements implied. Because of the importance in acknowledging the value of acting to assist an older adult, even when the officer perceived that the case was not criminal and that no investigative action was needed, a point was added when the officer recognized the elder needed assistance. The rating method allowed for points to be added to the response score even when it was not viewed as a crime worthy of further investigation. Independent Variables There were three focal variables, Knowledge, Bias, and Training. Foremost was Knowledge, which simply reflects the overall number of correct responses to the 25 question FAQ1, all of which relate to knowledge of older adults and the aging process (see Table 3.2). The FAQ1 is relatively reliable but is considered to have low to moderate validity as a measure of bias regarding older adults (Palmore, 1998). The second variable, Bias, was derived from the FAQ1 and reflected the extent to which the respondent endorsed beliefs that indicated either negative or positive bias toward older adults or the aging processes. As noted, using the FAQ1 to generate scores that measure age bias beyond the Knowledge score is controversial and is not widely 78 Table 3.2: FAQ1 scores Independent Variables Knowledge Bias Range 0 to 25 Method to calculate Total number correct responses -1 to 1 (# of pro-age bias responses/12) minus (# of anti-age bias responses/18) = % accepted as a valid measure. Initial values for Bias were obtained from the FAQ1 scoring rubric (see Appendix D). The next focal variable addressed the issue of training. Several survey questions or measures attempted to uncover a meaningful indicator of the effect of training related to the elder mistreatment statute. The different measures specifically sought answers to whether training for and familiarity with laws that addressed the abuse, neglect, and exploitation of a vulnerable adult were correlated with an increase in scenario response scores. The first question addressing training asked the following, "Have you received training regarding the abuse or neglect of a vulnerable adult?" Response choices to this question reflected both whether respondents had received training and whether or not they were interested in training (see Appendix B). This question was coded into two separate dichotomous or dummy variables: the first, addressing receipt of training, and the second, interest in receiving training? When respondents reported having received training and when they expressed an interest in receiving training their responses were coded with a value of 1. The second training question asked how recently training had occurred. This resulted in a ranked variable with most recent training given the highest value. The next 79 question asked whether the respondent was familiar with the legal definition of a vulnerable adult (see Appendix B). A "yes" response was given the value of 1 and a "no" response was valued as 0. The sixth and last training question included in the regression model clarified which of the listed elements defined a vulnerable adult. This statement, "Vulnerable Adult," means an elder adult (65+) or an adult 18 years of age and older who has a mental and physical impairment. Noting the manner in which respondents chose their answers, it was decided to collapse the measure into a dichotomous variable. This resulted in a variable with an adequate cell size for each value. Control Variables There were five control variables: Experience (the number of years of experience), Supervisory Status (rank), Age, Gender, and College (level of education). The first item, Experience, was easily coded as a continuous variable ranging from 1-35. No problems were found with the data on this variable other than those cases that were missing data. The item concerning rank asked about the highest position served and posed some difficulties. Because of the open-ended nature, officers responded with 23 different titles that were reclassified based on whether or not they indicated a commanding or supervisory status. After some consultation, this was deemed to be the most feasible and meaningful distinction, with a relatively high degree of reliability (see Appendix G). This variable was termed Supervisory Status. The item that addressed the location where officers had served was also open-ended in nature and resulted in 88 distinct responses, reflecting location, rank, and area of service or expertise. The range and inconsistency of 80 the responses rendered this measure meaningless, and it was excluded from the regression model. The next variable was Age, a continuous variable, requiring only that any missing data be verified. Gender was another relatively straightforward control variable with males being coded as "0" and females as "1," making "female" the attribute of interest with males as the comparison group. The survey also addressed level of education identified as the variable, College, in the study. Because there was not adequate variation across all values (small cell sizes), the variable was coded as a dummy variable to reflect whether or not the respondent had completed college. Having completed a college degree was given the value of "1" and not completing was given the value of "0." Data Analysis A model was formulated to determine if each of the hypotheses was supported by the data. The control or moderating variables chosen by UCOA, as previously noted, represented theoretical considerations as well as input based on the extensive cumulative knowledge, expertise, and experience of the members of the commission. As such, each of the variables was utilized in the full regression analysis to assess the significance and B-coefficient value for each. Lastly, because the dependent variable could be reduced to individual elements, each of the six elements was examined to see if any significant correlation existed with the independent variables included in the first model. 81 Summary of Methodology Objectives The primary objective of this study was to determine whether ageism, as indicated by the variable, Knowledge, affects Scenario Response (police officers' responses to cases of elder mistreatment). This objective was achieved by analyzing the data from the survey of law enforcement officers conducted by UCOA. Additionally, Bias and variables that measured different aspects of elder mistreatment specific training were included to test the secondary hypotheses. The addition of the variable Bias could increase the predictive validity of the variable Knowledge, and the measures of training could explore whether training has some moderating effect. Design This study design consisted of an analysis of cross-sectional data using linear (OLS) regression to examine the effect of knowledge of older adults as measured by the Palmore Facts on Aging Quiz (Knowledge) and the response of law enforcement officers to two scenarios describing mistreatment of older adults. Additional focal variables, Bias and Training (self-reported training), were included in the model to explore the additive effect of Bias and Training on officer response. Likewise, the additional measures of training were included after testing for multicollinearity along with all control variables. The regression was conducted using all variables mentioned to ascertain the factors that account for the greatest variance and significance of police response to the scenarios based on the theoretical model set forth in the review of literature and included in the UCOA survey (see Figure 3.1). 82 Knowledge Response Scores to Elder Mistreatment Scenarios Bias Training Experience Supervisory Role Age College Gender Figure 3.1 Variable Diagram CHAPTER 4 FINDINGS Analyses of the results will begin with an examination of the descriptive tables for all variables of interest starting with the control variables, principle dependent variable, Scenario Response, and then moving through focal variables. After review of the descriptive data, the findings for the regression models will be discussed. Sample The descriptive data reveal that the participants in the initial survey consisted of 200 officers. The median age was 39 with the number of valid subjects (n) 171. The average years of law enforcement experience was 13.6 years (n = 183). Of those who reported their gender (182), only 19% were females. The frequencies for males (148) and females (34) were similar to other studies that surveyed law enforcement officers (Daniels et al., 1999; Davies, 2010; Dolon & Hendricks, 1989; Sever & Youdin, 2006). Of those who reported their level of education (178), 94% had attended at least 1 year of college. Thirty-three percent completed a college degree, and less than 6% held graduate degrees. Of those who reported their highest position or rank served (173), 20% indicated they were in a supervisory role. Race or ethnicity were not among the measures employed in the study, so no data are available. (See Tables 4.1 and 4.2). For each of the independent variables, the descriptive data indicated that there 84 Table 4.1: Continuous control variable response frequencies Continuous Variables Experience Range 1-35 Missing Age Frequency Mean 183 13.6 Percentage of Total 91.5 39 85.5 17 25-65 Missing 171 23 Table 4.2: Categorical control variable response frequencies Categorical Variables Supervisory Status Supervisor Not a supervisor Missing Total Gender Male Female Missing Total Highest level of education High School Graduate Partial College ( ≥ 1 year) or Specialized Training College Degree Graduate or Professional Training (Graduate degree) Missing Total Frequency 173 35 138 27 200 Percent 86.5 17.5 69.0 13.5 100.0 148 34 18 200 178 11 74.0 17.0 9.0 100.0 89.0 5.50 108 48 11 54.0 24.0 5.50 22 200 11.0 100.0 85 was considerable variance in the amount of missing data due to nonresponse. For example, the number of missing data for those completing the questions related to training varied from 0 missing data points to 26. Because of this significant difference in the amount of missing data and because cases varied on what data were missing, analyses included corresponding amplification in the number of missing responses. The valid n for which analysis could be conducted was significantly reduced in the full regression model. Variables Dependent Variable: Scenario Response Of the 200 respondents, 180 completed all elder mistreatment scenarios. Item scores represent officers' open-ended responses to the two scenarios of elder mistreatment included in the UCOA survey. A brief overview shows respondents who completed the scenarios, on average, included just over five out of the possible 12 discrete elements used to measure the effectiveness of officer response (M = 5.22). This reflects combined scores for the responses to the two scenarios. Six points were possible for each scenario (see Table 4.3). The most frequent response was 6 with 18.9% of officers scoring at that level. The next most common response was a score of 5 for 16.1% of officers. The highest score obtained by any of the officers for the scenarios was 10. Only one officer received Table 4.3: Descriptive statistics for the elder mistreatment response Dependent Variable Elder mistreatment scenario scale N Minimum Maximum Mean Std. Deviation .00 12.00 5.22 2.06 180 86 a score of 0, which meant the officer viewed both scenarios as requiring no intervention from law enforcement (see Table 4.4). A score of 7 or above signified the respondent recognized that the older person was vulnerable. While an officer may have acknowledged that the vulnerable adult statute applied with a lower score, a score of 7 or higher indicated that the officer necessarily indicated that at least one of the scenarios involved a vulnerable adult. Therefore, a score of 7 or above indicates that the officer acknowledged both criminality and risk to a vulnerable adult. There were 50 officers (27.8%) who reached this level, which approached the level of effectiveness desired by the UCOA. Focal Variables In the study, 161 law enforcement officers completed the FAQ1. When an officer did not answer all of the survey items for the FAQ1, the data for that respondent were Table 4.4: Elder mistreatment scenarios 1 and 2 combined scores Scores .00 1.00 2.00 3.00 4.00 Valid 5.00 6.00 7.00 8.00 9.00 10.00 11. 00 12.00 Total Missing 99.00 Total Frequency Percent 1 0.5 5 2.5 10 5.0 25 12.5 26 13.0 29 14.5 34 17.0 25 12.5 16 8.0 6 3.0 3 1.5 0 0.0 0 0.0 180 90.0 20 10.0 200 100.0 Valid Percent .6 2.8 5.6 13.9 14.4 16.1 18.9 13.9 8.9 3.3 1.7 0.0 0.0 100.0 Cumulative Percent .6 3.3 8.9 22.8 37.2 53.3 72.2 86.1 95.0 98.3 100.0 100.0 100.0 87 unusable, and the score was excluded from the regression analysis. The mean score for these officers (11/ 25 or 46%) shows that on average they answered fewer than half of the questions correctly on a measure intended to discover how familiar they were with elders and the process of aging. The range of correct responses fell between 6 and 19. This range is somewhat below the expected frequency estimated by Palmore (1998) but was not significant (see Table 4.5). Responses to the FAQ1 were used to calculate the values for both knowledge and bias. Within the UCOA survey, the concept of training was operationalized using six questions. The questions regarding training appeared at the beginning of the survey. With the exception of question 5, almost every respondent completed this section of the survey. Question 1 served as the most direct measure of whether or not an officer had received training; only 43.5% reported affirmatively. In other words, of the 200 officers, 113 officers indicated they had not received training. In a similar study by Sever and Youdin (2006), 89% reported receiving training. Whether or not they had received training, 72% of respondents said that they would be interested in training. Only 29% were certain they were familiar with the definition of a vulnerable adult. With respect to the Utah elder abuse statute, just over 26% said "yes," they were familiar, and of those, 15 (7.5% of the total) said they had Table 4.5: Descriptive statistics the FAQ1 scores Independent Variables Knowledge N 161 Minimum 6.00 Bias 161 -0.64 Maximum 19.00 0.58 Mean 11.47 Std. Deviation 2.49 -132 .230 88 actually applied the statute. For question 2, only 43.5 % of respondents indicated how recently their training had taken place. Of these, the majority (51/87 or 59%) indicated training had occurred within the past year; an additional 26% (n = 23) observed that training had taken place within the past 1-2 years. Only 13 indicated training had occurred more than 2 years prior to the survey (see Figure 4.1). Question 3 was unusual in that it allowed officers to choose whether they were familiar, unfamiliar, or uncertain with the definition of a vulnerable adult. Twenty-nine and a half percent responded that they were unfamiliar (n = 59) with almost the same number (29%, n = 58) indicating they were familiar. The greatest proportion (41.5%) Training Recency Number of Respondents 120 100 80 60 40 20 0 No Training Less than 6 6 months to 1 to 2 years 2 to 3 years More than 3 months 1 year years Time since training Figure 4.1 Training Recency 89 reported they were uncertain how familiar they were with the definition (see Table 4.6). Once again, with respect to the issue of familiarity, question 4 asked officers to respond to how familiar they were with the Utah elder abuse statute. The majority of officers responded that they were at least somewhat familiar with the statute (n = 143, or 71.5%). The concern with this question and the one that followed was the discrepancy between the numbers who reported having used the statute when arresting (question 4) and those who then reported in what setting (question 5). Question 5 asked officers to identify the location in which they had applied the vulnerable adult statute. The officers weighed the following options: private care, longterm care facility, a hospital, or other. In question 4, only 15 officers said they had applied the statute; however, in question 5, 24 reported doing so with 26 instances of no response or missing data. This is a significant difference, particularly with such a small number reporting having applied the statute in an arrest. This discrepancy calls into question whether officers understood the question and, if so, why there was such a difference in the responses. This discrepancy made it necessary to exclude both questions from the regression analysis due to unreliability, despite addressing an interesting issue identified in the literature. The final training item, which further explored officers' familiarity with the meaning of "vulnerable adult," had a high response rate but appeared to be confusing for respondents. The question provided the basic definition of a vulnerable adult and asked officers which of six different response choices were consistent with their understanding of the definition or the distinct components of what a vulnerable adult was unable to do. All of the elements included in the question represented aspects of the limitations of 90 Table 4.6: Training item response frequency Questions / Responses Have you received training regarding abuse, neglect or exploitation of a vulnerable adult? Yes, no further training needed on this topic Yes, and would like further training No, not interested in training No, but would like training Missing data Total If you answered "Yes" to the prior question, please approximate when the training took place. Not applicable Within last six months Six months to one year ago One to two years ago Two to three years ago Over three years ago Missing data Total Are you familiar with the legal definition of vulnerable adult? No Yes Uncertain Missing data Total Are you familiar with the Utah elder abuse statute? No Somewhat familiar Yes, but have not applied Yes, and have applied in an arrest Missing data Total If you have applied 76-5-111 in an arrest, in which location? Private residence Long term care facility Hospital Other I have not applied 76-5-111 Missing data Total Valid Percent Cumulative Percent Frequency Percent 24 63 32 81 0 200 12.0 31.5 16.0 40.5 0.0 100.0 12.0 43.5 59.5 100.0 113 24 27 23 10 3 0 200 56.5 12.0 13.5 11.5 5.0 1.5 0.0 100.0 56.6 68.5 82.0 93.5 98.5 100.0 100.0 59 58 83 0 200 29.5 29.0 41.5 0.0 100.0 29.5 58.5 100.0 100.0 56 90 38 15 1 200 28.0 45.0 19.0 7.5 0.5 100.0 28.0 73.0 92.0 99.5 100.0 18 4 2 0 150 26 200 9.0 2.0 1.0 0.0 75.0 13.0 100.0 10.3 2.3 1.1 0.0 87.4 100.0 91 Table 4.6: continued Questions / Responses Frequency Percent 97 48.5 b. Provide necessities such as food, shelter, clothing, or medical or other health care 134 67.0 c. Obtain services necessary for health, safety, or welfare 108 54.0 d. Carry out the activities of daily living (ADL) 105 52.5 99 49.5 120 60.0 7 3.5 200 100.0 87 7 6 6 3 84 7 200 106 87 7 200 43.5 3.5 3.0 3.0 1.5 42.0 3.5 100.0 53.0 43.5 3.5 100.0 "Vulnerable Adult" means an elder adult (65+) or an adult 18 years of age and older who has a mental and physical impairment which substantially affects the person's ability to: a. Provide personal protection e. Manage the adult's own resources f. Comprehend the nature and consequences of remaining in a situation of abuse, neglect, or exploitation Missing data Total Frequency of total number correct for question 6 1 Correct 2 Correct 3 Correct 4 Correct 5 Correct 6 Correct Missing data Total Number of correct responses (1-4) Number of correct responses (5-6) Missing data Total Valid Percent 56.6 100.0 Cumulative Percent 92 vulnerable adults. These elements included the following capacities: 1) Provide personal protection, 2) Provide necessities, 3) Obtain necessary services, 4) Carry out the activities of daily living (ADL), 5) Manage personal resources, and 6) Comprehend the nature and consequences of remaining in a situation of abuse, neglect, or exploitation. One-hundred and ninety-three officers responded, and 87 selected only one response while 84 selected all six, the correct response. Each of the remaining responses was so small as to necessitate collapsing all of the values for the variable (Abu-Bader, 2011) and recoding into a dummy variable. Respondents who included fewer than five items were the comparison group and coded as 0. Those who selected five or six were coded with a 1. The numbers, while not dramatically different, showed that 53% (106) selected fewer than five responses with 87 choosing five or six. The information could be collapsed with little loss in meaning of the variable under consideration (Sweet & Grace-Martin, 2012). Regression Models In choosing the variables for the full model (model 1), all variables deemed important in the review of the literature were included. Tests were conducted for multicollinearity. As mentioned above, Age and Experience were highly correlated. The full model included both variables because they were expected to account for some of the variances according to the literature. Additionally, as noted earlier, questions four and five in the UCOA survey presented problems of reliability and, therefore, were not included the full model. Questions 3 and 6 addressed a similar issue to questions 4 and 5 and could effectively be viewed as a proxy for being familiar with the Utah Statute 76-5111. When tested for multicollinearity, questions 3, 4, and 5 had high correlation values 93 as well. This confirmed the need to choose the best of these variables and the decision to include only the variable, Legal Definition, derived from question 3 in the full regression. In the variable subsets that followed, variables were removed from the full model using a backwards elimination strategy to further explore their relationships. Variables were eliminated from the model based on problems with correlation (as seen with Age and Experience), as well as low predictive value relative to the outcome variable. Those with theoretical importance remained in the model despite low predictive value or low levels of significance to minimize bias (Abu-Bader, 2011). In the second subset of variables, the issue of multicollinearity was addressed for two pairs of variables. As noted, Age and Experience were highly correlated. Age was removed rather than Experience because, for this study, officer's experience was more salient. Also, whether officers indicated they had received training was highly correlated with how recently they had been trained. An officer's self-reported training was selected for the model over recency of training because this was most comparable to the measures used by Davies (2010). In that study, self-reported training and the amount of training one received were highly predictive of how well officers responded to scenarios of elder mistreatment. When the regression for this set of variables was run, Training or self-reported training was found to be an insignificant predictor of officers' responses to the scenarios. Because of the importance of training (Jogerst, Daly, Dawson, Brinig, & Schmuch, 2004; Johnson, 2002; Nusbaum, Mistretta, & Wegner, 2007; Plotkin, 1996), the variable, Training Recency (which was initially found to be more significant) was included to retain some measure of training in the model. The analysis of this subset showed that 94 Training Recency or how recently officers were trained was positively the variable, Scenario Response, at the p < 0.1 level. In the fourth subset of variables, when Supervisory Status was removed, both remaining training variables, Training Interest and Training Recency, proved to be more significant predictors of police response. In the fourth nested model, Training Interest had a significance of p < .05 (.035), and Experience had a significance of p < .05 (.015). Supervisory Status was removed because while there was not multicollinearity in the strictest sense, the correlation between Supervisory Status and Experience was still .54- high enough to cause some concern. As mentioned above, this variable collapsed the response related to rank from 23 values to two, whether or not an officer's rank included a supervisory role or not. Because of this, there was some concern as to how much of the original meaning or value of the variable, Supervisory Status, was retained in the model. As the more valid and reliable measure, Experience also captured some variance accounted for by the rank variable, Supervisory Status, as evidenced by the shift from no significance to a relatively high level (p < .05) between the third and fourth set of variables. It is noteworthy that when Supervisory Status was removed, Gender lost significance. The role of being a commanding officer then interacted in some important way with Gender as well as Experience. In the last subset, because the variable Legal Definition did not show any significance when it was removed, there was no change in the R-squared value (.229) or the coefficients for the remaining variables (see Table 4.7). In regard to hypothesis 1, a quick glance of Table 4.7 reveals that Knowledge 95 Table 4.7: Results for the nested linear (OLS) regression models predicting police officer response to scenarios of elder mistreatment VARIABLES Knowledge Bias Experience Definition Elements College Training Interest Gender (Female) 0.160* -1.846* 0.028 0.766* -0.990** 0.599 1.005* Training Recency Legal Definition Supervisory Status Training Age Constant R-squared Observations 0.414† -0.2 1.261* -0.531 -0.0184 2.679* 0.279 123 Scenario Response 0.175* 0.170* 0.161* 0.160* -1.738* -1.697* -1.257† -1.261† 0.016 0.0201 0.0557* 0.0558† 0.761* 0.754* 0.603† 0.602† -0.903* -0.905* -0.865* -0.863* 0.558 0.565 0.826* 0.825* 0.930† 0.877† 0.648 0.647 -0.102 1.043† 0.379 1.964** 0.251 125 0.244† -0.162 1.076† 0.247† -0.0181 0.245† 1.946** 0.267 125 1.639* 0.229 131 1.631* 0.229 131 ** Significance at the p < .01 level. * Significance at the p < .05 level. † Significance at the p < 0.10 level. remained consistent across each of the five nested variable sets modeled with a minimum significance at the p < .05 level regardless of which variables were included or removed. The range of the slopes (.16 to .175) conveyed that for each correctly answered question on the FAQ1 (unit increase = 1) there was approximately a sixth of a point increase in the scenario response score. Roughly, every six additional correct items on the FAQ1 predicted that officers would include an additional appropriate element (protective behavior) in their responses to the elder mistreatment scenarios. The R-squared values (.23 to .27) and the high level of significance (p < .05) indicate the variables in the model account for between 23-27% of the variance in the scenario response scores. As officers' overall knowledge of older adults increased, there was a significant increase in the number of protective actions they included in their description of how to 96 respond to scenarios of elder mistreatment. Again, the increase in the total variation that is explained by the model, when all variables are included, represents a medium effect size for how appropriately officers responded to elder mistreatment scenarios (AbuBader, 2011). In regard to hypothesis 2, Bias was significant across all five models with a minimum significance level of p < 0.1 or higher. In the last subset of variables, Bias or the difference between the anti-aged bias percent and the pro-aged bias percent was found to be significant at the p < 0.1 level (.094). While not as significant as the Knowledge score, the Bias score was also a consistent predictor of the variance in the scenario response scores, with a B value of -1.697 in the third nested model. Respondents then consistently scored approximately .017 fewer points on the scenario for every percent increase on the Bias score. In other words, for every 10 percentage points officers gained in Bias score, their predicted scenario response score dropped by .17 points. While this is a relatively small increment, considered across the range of responses, the difference between the lowest Bias score, -.64, and the highest, .58, accounted for nearly two full points on the scenario response score. Palmore considered net-aged bias, represented by Bias score, to be a useful measure of bias when a more cumbersome and time consuming measure was deemed unnecessary (Palmore, 1998). With respect to hypothesis 3, whether training would override inaccurate knowledge and bias, the factor of training was analyzed in different forms. The second subset of the nested table explored whether or not an officer's self-reported training (Training) affected the outcome variable. As previously discussed, because of correlation between the variables representing whether an officer had received training 97 (Training) versus how recently the officer was trained (Training Recency), model 3 included only Training Recency. Training was found to be insignificant with how officers responded to the scenarios, even when Training Recency was removed. However, when Training Recency replaced Training in the model, Training Recency proved to be significant at the p < 0.1 (.065) level with a B coefficient of .244. For every step decrease in the length of time since the officer was trained, the scenario score increased by .244. Officers who were trained within the last 6 months versus those who were trained more than 2 years ago, gained a full point on their scenario response scores. The confidence level for Training Recency warrants that it be interpreted with caution and requires further assessment to establish the importance of this relationship. Additional patterns emerged as variables changed across nested models. First, Experience was not initially significant. There was an obvious theoretical overlap with age and even with rank, which was captured in the variable, Supervisory Status. In order to explore these relationships further, in subset four, the variable, Supervisory Status, was removed from the equation along with Age, leaving only Experience from this trio. In that subset, Experience had a B coefficient of .056 at the p < .05 (.014) level of significance. Each additional year of experience corresponded with a .06 increase in the scenario response score. While this accounts for another small amount of change, there was a strong positive correlation between Experience and how an officer responded to the elder mistreatment scenarios. Three additional measures associated with training were found in the UCOA survey and included in the regression analysis to further explore the effect of training on police response. These included whether or not officers stated they were familiar with 98 the legal definition of a vulnerable adult (Legal Definition), whether officers could correctly identify the components of that definition (Definition Elements), and whether or not officers expressed an interest in training (Training Interest). As noted above, when the variable, Supervisory Status, was removed Experience and Training Interest increased in significance while Gender decreased in significance. For this reason, the three indirect training variables are discussed with regard to models 4 and 5. The variable, Definition Elements, was significant across all five subsets of variables in the nested model, at a minimum of p < 0.1 level. In subset five, the B coefficient for Definition Elements was .602 with a p < 0.1 (.099) level of significance. This indicates that if officers correctly identified five or more of the six elements of the legal definition of a vulnerable adult, their scenario response score increased by 0.6 more than officers who identified fewer than five correct elements. The variable Legal Definition or whether an officer stated that they were familiar with the legal definition of a vulnerable adult was not significant in any of the first four models. Consequently Legal Definition was removed from the last subset to see if it interacted with the other training variables; it did not. Whether or not respondents indicated familiarity with the legal definition had no significant effect on how they responded to elder mistreatment scenarios. In other words, when officers said they had training or that they were familiar with what training was supposed to teach, it was not predictive. However, when respondents reported that training had occurred more recently and could respond in an informed manner to what they had been trained on, the predictive value of their responses became more significant. The last two variable subsets highlight whether or not an officer had an interest in 99 training, represented by the variable Training Interest. As noted, when Supervisory Status was removed, Training Interest showed significance. The fifth subset shows this variable has a B value of .825 with a p < .05 (.034) level of significance. This means that officers who expressed an interest in training were expected to have almost a one point increase in their scenario scores over officers who did not. Within this regression model, the first set of variables explains the greatest variance. However, upon closer inspection it can readily be seen that by removing the variables Age, and Training, the R-squared value moves from 0.279 to 0.267, a loss of only 1% without a great deal of risk of increased bias due to omission of variables. Knowing that Age and Experience demonstrated problems with multicollinearity as well as Training and Training Recency, this result is not surprising. The seven variables of significance in this model were Knowledge, Bias, Definition Elements, College, Gender, Training Recency, and Supervisory Status. The 0.267 R-squared value indicated that there are a number of variables other than those considered in the UCOA study that account for the variance in officers' response scores for elder mistreatment scenarios. It is important to note that this included College, the measure for education, which according to Palmore (1998) should have predicated a strong positive relationship with the FAQ1 score. However, completion of a college degree correlated with a 0.9 decrease of or loss of approximately one point on the scenario scores. This is opposite of what might have been predicted based on Palmore's observations and will be discussed in Chapter 5. Looking at the second subset, when respondents held a supervisory position, their predicted scenario scores increased by 1.043 points compared to nonsupervising officers. 100 This was significant to the p < 0.1 level (.06). This finding shows that whether or not officers had supervisory experience made a somewhat significant difference in how they responded to elder mistreatment. Those in a supervisory role may have had more training in general and more experience. As the variable, Supervisory Status, was removed from the model, Experience became more significant. Another interesting element of this model, Gender was initially a significant predictive of officer's scenario response scores. Female respondents were more likely to score one point (.930) higher than males at the p < 0.1 (.062) level of significance. It is also worth mentioning that Gender was not significant in all models. When Supervisory Status is removed, Gender becomes insignificant. Interestingly, an analysis of variance between the two showed almost no correlation. Age, another variable that some researchers (Palmore, 1998) have assumed to be associated with age bias, was found to be insignificant. This is of course consistent with Palmore's finding that education alone was a significant predictor of knowledge of aging and elders. Once again, the variation attributed to this factor may have been captured by other measures such as Experience or Supervisory Status. In summarizing the findings, it is clear that as knowledge of older adults increases, as indicated by the Knowledge score, a significant increase also takes place with how officers respond to elder mistreatment scenarios. Net-aged bias, as represented by the variable, Bias, also played a small but consistent role in how an officer responded to elder mistreatment scenarios. Training was significant, but only as it related to actual knowledge, how recently the training occurred, and whether an officer expressed interest in receiving training. The report of having received training, as it pertains to appropriate 101 response to elder mistreatment, was unimportant. Education, as indicated by whether or not an officer has a college degree (College), was significant, but not in the expected direction. If officers held a supervisory position, they were more likely to respond appropriately than officers who did not. Yet, when this factor was removed, Training Interest and Experience became significant in predicting an officer's response to the scenarios. Scenario Response Subscales In order to explore the effects of the predictor variables on the specific elements or items constituting the variable, Scenario Response, the items were analyzed individually. In doing so, two categories began to emerge (see Table 4.8) and formed six subscales: Open/Refer, Investigate, Charge/Arrest, Help/Protect, Involve APS, and Vulnerable Adult (VA) Statute. When regressions were performed independently using these scales as variables, several correlations of interest emerged. The first of three elements had to do with standard law enforcement procedures and acknowledged that a criminal act had occurred. These responses, such as opening a case for later investigation (Open/Refer), conducting an investigation (Investigate), and charging or arresting the suspect (Charge/Arrest-which was the most correct response for the given scenarios), would be expected whether or not the officer believed the case involved a vulnerable elder. In other words, when an officer encountered a scenario in which there was evidence of felony credit card fraud, any or all of these first responses would be appropriate without recognizing or invoking the vulnerable adult statutes. More than 98% of respondents indicated they would open at least one of the cases, with only three respondents indicating they would not open either case. Just over 90% 102 Table 4.8: Individual item frequencies totals from elder mistreatment scenarios 1 and 2 Individual Item Civil matter / Case not opened Open/Refer Investigate Charge/arrest Help/Protect Involve APS VA Statute .00 1.00 2.00 Total .00 1.00 2.00 Total .00 1.00 2.00 Total .00 1.00 2.00 Total .00 1.00 2.00 Total .00 1.00 2.00 Total .00 1.00 2.00 Total Frequency 131 46 3 180 3 54 123 180 17 69 94 180 103 59 18 180 92 72 16 180 114 61 7 182 105 58 19 182 Percent 65.5 23.0 1.5 90.0 1.5 27.0 61.5 90.0 8.5 34.5 47.0 90.0 51.5 29.0 9.0 90 46.0 36.0 8.0 90.0 57.0 30.5 3.5 91.0 52.5 29.0 9.5 91.0 Valid Percent 72.8 25.6 1.7 100.0 1.7 30.0 68.3 100.0 9.4 38.3 52.2 100.0 57.2 32.8 10.0 100 51.1 40.0 8.9 100.0 62.6 33.5 3.8 100.0 57.7 31.9 10.4 100.0 Cumulative Percent 72.8 98.3 100.0 1.7 31.7 100.0 9.4 47.8 100.0 57.2 90.0 100.0 51.1 91.1 100.0 62.6 96.2 100.0 57.7 89.6 100.0 indicated they would investigate one or both cases personally. However, the number who indicated a charge or arrest should be made (the ideal response) in at least one of the scenarios dropped significantly to 42.8% (see Table 4.8) suggesting a much smaller portion of officers were confident responding this forcefully. The second identified cluster were those in which the respondents indicated at some level the case involved a vulnerable older adult. The items acknowledging that the case involved a vulnerable elder included the following: addressing a safety or protection issue (Help/Protect), involving APS (Involve APS), or recognizing that vulnerable adult 103 statute applied to the case (VA Statute). Almost 50% indicated that they would take action that was intended to address safety concerns (Help/ Protect) without involving APS, such as assisting in finding a safe placement or a new caregiver. Almost 58% did not indicate that the vulnerable adult statute applied to the case, and over 62% did not mention the need to involve APS. It was not a surprise that the most aggressive or proactive response, that of charging or arresting, also was the least likely to be included in an officer's response (see Table 4.8). It is important to note that some respondents who did not view the case scenario as a crime did acknowledge the need to assist a vulnerable adult either by referring to APS or offering other forms of assistance. The first set of behaviors mentioned above was captured in the elder mistreatment rubric and are consistent with effectively responding to all criminal behaviors including those that do not necessarily involve a vulnerable adult. The second set represents behaviors, which were consistent with or mentioned in the Utah vulnerable adult statute-behaviors that both law enforcement and APS would like to see increase when deemed appropriate (see Appendix F), were the most desirable responses for both scenarios). Because UCOA conducted this survey in response to low numbers of prosecution and arrests using the vulnerable adult statute, a second regression analysis was run to see whether or not officers acknowledged the VA statute in their responses to the scenario. The highest possible score for this is 2, one point for each scenario. As noted above, 19 respondents received both possible points while 58 received one point. An OLS linear regression was run using the same five nested models from the first analysis with the elder mistreatment scenarios. This allowed for some contrast between the two 104 regressions (see Table 4.9). Knowledge was a significant factor in each subset of the model. In all subsets, the B coefficient is approximately 0.044 with a minimum significance of p < 0.1 (.06). This means that FAQ1 was predictive of whether or not officers recognized the VA statute should be applied when describing how they would respond to the scenarios. A coefficient of .044 means that for each point officers receive on the FAQ1, they will on average increase the probability of using the VA statute by 4.4%. When looking only at whether or not an officer will acknowledge the VA statute, Training was significant in the second variable subset. This was not the case when the complete scenario score was the dependent variable. In the second variable subset, the B coefficient for Training was .224 with a significance level of p < 0.1 (.072). If officers Table 4.9: Results for nested OLS regression models predicting police officer use of the VA statute in their response to scenarios of elder mistreatment. VARIABLES Knowledge Bias Experience Definition Elements College Training Interest Gender (Female) Training Recency Legal Definition Supervisory Status Training Age Constant R-squared Observations 0.0430* -0.172 0.00602 0.127 -0.0305 0.174 -0.134 0.105 -0.0283 0.331† -0.00564 -0.0051 -0.173 0.191 124 ** Significance at the p < .01 level. * Significance at the p < .05 level. † Significance at the p < 0.10 level. 0.0490* -0.0994 0.00369 0.124 -0.0078 0.181 -0.155 -0.012 0.251 0.224† -0.396 0.178 126 VA Statute 0.0470† 0.0439† -0.0986 -.00226 0.00497 0.0143† 0.1397 0.0941 -0.0105 -0.0243 0.1899 0.240† -0.15 -0.171 0.099* 0.0911* -0.0285 0.0131 0.268 -0.345 0.177 126 -0.425 0.165 132 0.0444† 0.000758 0.0143† 0.095 -0.0263 0.240† -0.17 0.0927* -0.419 0.165 132 105 had received training, they were 22.4% more likely to acknowledge the VA statute than officers who had not received training. As with the regressions above, Training Recency also was significant in the last three variable subsets. On average, the B coefficient for this variable is .09 with a significance level of p < .05 (.025). Being trained more recently did affect whether or not an officer acknowledged the VA statute. Use of VA statute increased by .09 with more recent training. Again this small but significant difference supports the hypothesis that recency of training increases the likelihood that an officer will use the VA statute when confronted with elder mistreatment. Similar to the previous regressions, when VA Statute was the only component of the dependent variable, Experience was still significant in the last two variable subsets. In both these subsets, the B coefficient for Experience is approximately .0143 with a significance level of p < 0.1 (.055). Training Interest was significant in both the fourth and fifth variable subsets. The B coefficient for Training Interest was approximately .24 for both, with a significance level of p < 0.1 (.062). Training Interest accounted for a significant difference in whether or not an officer recognized the VA Statute should be applied when confronted with the scenarios. It is worth looking at this model to determine what variables were insignificant in regard to whether or not an officer reported the need for using the VA statute when confronted with an elder mistreatment scenario. Bias, Definition Elements, College, and Gender all showed some significance in the regression when an officer's full response to the scenarios was considered rather than just one aspect. In the second regression none of these had significance. CHAPTER 5 SUMMARY This study set out to examine the effect of ageism on law enforcement officers' response to elder mistreatment. Ageism was defined as an individual holding incorrect beliefs, perceptions, or stereotypes of people based on age. The assumption of this study was that law officers holding ageist views would treat older adults differently than those who held more accurate knowledge or beliefs. This assumption was tested using officers' responses to two scenarios of elder mistreatment. Officers were asked to explain what they would do in each case. Both scenarios included situations in which the proper responses required the officer to act to protect the older person. In order to do this, the officer would ideally have described the steps taken to prevent harm by actively investigating, arresting, or charging the perpetrator using the vulnerable adult statute and making appropriate referrals. A significant positive correlation was found between officers' knowledge of older adults and whether they responded appropriately to the elder abuse scenarios. Furthermore, as officers' incorrect knowledge shifted from a negative view of elderly to an unrealistically positive view, they became less likely to include protective behaviors in their scenario responses. While research suggested that training would increase the probability that officers would include protective responses (Davies, 2010; Sever & Youdin, 2006), the results 107 showed no correlation when all behavioral elements were included in the first regression analysis. However, when the outcome variable was limited to the use of the vulnerable adult statute alone a positive correlation was found. When the outcome was viewed holistically, training did not contribute to an overall effective response. The results of this study suggest the approach-targeted, statute-specific training often used by law enforcement (Stiegel, 2006) does not mitigate the effects of ageism but may lead to an increase in the application of the statute. Discussion As an indirect measure of ageism, this study looked at knowledge of older adults operationalized by the Facts on Aging Quiz 1 (FAQ1) Knowledge and Bias scores. As defined, ageism was a significant predictor of how law enforcement officers responded to elder mistreatment. When the more direct measure of net-aged bias, Bias, was included, the results showed this added measure significantly increased confidence in the importance of ageism as a predictor of police response. Since the assumption that ageist beliefs and attitudes could lead to differential or even discriminatory behavior was supported by this study, increasing knowledge about older adults and the aging process should result in an increase in the number of protective responses by police officers when responding to elder mistreatment. In looking at the overall findings, the first issue that stands out is that the mean FAQ1 Knowledge score fell well below the expected general population average especially when level of education was included (Palmore, 1998). The average score was below what would have been expected even for those who had only completed a high school level of education. Given that a significant number of officers had completed 108 college (59), and that even of those who had not, the majority reported having attended some college, this unexpected finding highlights the importance of understanding why officers in this study had such low knowledge levels of older adults and the aging process. Studies (Hopkins, 2000; Hyde & Miller, 1997) have shown that contact with older adults changed participants' perceptions and increased their knowledge of older adults. If this idea holds true, the officers in this study may have had less interaction with older adults. According to the assumptions of the contact hypothesis, in order to change attitude and behavior toward another, it is necessary for certain conditions to be met (Allport, 1954): there must be equality in power or status between those who interact with each other, as well as a common goal and cooperative and positive interaction. When officers have had negative experiences with older adults, the assumptions of the contact hypothesis would have been violated, and no change would have be expected. Incorporating this idea, when officers have had contact, it may not have been positive. This was the case in early studies that found when police officers reported increased contact with older adults, their attitudes toward them became more negative (Goldsmith, 1976; Goldstein, 1996; Sykes, 1977). Thus, lack of interaction or negative interaction may explain the low scores. As noted previously, social learning theory holds that attitudes and beliefs are transmitted to others through social interaction with more significant others (family, peers, and persons in authority) having the greatest impact (Bandura, 1971). Previous research explained that when officers have experienced the criminal justice system as unresponsive and ineffective they are less likely to refer future cases to that failed system 109 (Daniels et al., 1999; Dubble, 2006; Johnson et al., 1994). When officers have been trained in the use of the vulnerable adult statute, they may be more likely to expect certain outcomes. When these outcomes do not occur, officers who have been trained may be more dissatisfied than those without such expectations. Research shows officers' negative experiences with elders and the criminal justice system contributed to the problem of police agencies becoming less responsive to the needs of older adults (Daniels et al., 1999; Dubble, 2006). Based on Palmore's (1998) observations as well as Sever and Youdin's (2006) findings, level of education, which was expected to be positively correlated with FAQ1 scores, had the opposite effect on the scenario response scores. Officers who reported having a college degree had almost a full point lower score on the response scenarios than officers who reported no degree (see Appendix I). When education was correlated with the FAQ1 as the dependent variable, there was no significance. This unexpected lack of relationship between general levels of education and how officers performed on the FAQ1 measure and the negative correlation between education and the scenario response scores suggests that Palmore may have overestimated the effect of education. Despite the lack of connection between education and FAQ1 score, the finding that an increase in law enforcement officers' FAQ1 Knowledge scores predicted a corresponding increase in response scores on elder mistreatment scenarios supported the hypothesis that ageist views affect law officers' responses to elder mistreatment. Because the hypothesis was supported, and education did not predict FAQ1 scores, the implication is that knowledge of older adults may be acquired differently. A closer look at Palmore's (1998) discussion of education shows that, while level of education has 110 shown a positive correlation with FAQ1 scores, the strongest correlation was expected for those who had completed a course in gerontology. In light of the low scores on the FAQ1 by officers, law enforcement agencies interested in improving response to elder mistreatment may wish to consider providing officers with a course in gerontology (Sever & Youdin, 2006). The finding that being a supervising officer moderated the focal relationship between the FAQ1 score and officers' responses to the elder mistreatment scenarios was not surprising. Commanding or supervising officers would be expected to have more experience, take more responsibility, and demonstrate greater competence. This finding is consistent with social learning theory and implies that law enforcement agencies could use supervising officers to counteract the effects of negative experiences that can occur when interacting with older adults. By identifying supervising officers who have accurate knowledge of older adults, administrators could tap them to increase the effectiveness of training. Future researchers may wish to explore learning models that help less experienced officers internalize the influence of their supervisors. Another important aspect of this study was to explore whether a more discreet measure of the FAQ1 could be used to detect police bias toward older adults (ageism). It was anticipated that officers who showed stronger bias toward older adults as measured by the FAQ1 Bias score would be less responsive to the needs of older adults exposed to elder mistreatment. Age bias taken independently added to the predictive value of the model. Having controlled for the effects of education, gender, and age as predictors of how officers will respond, an increase in officer bias or a shift from a negative to a positive bias, as indicated by Bias, was correlated with a decrease in an officer's response 111 scores to the elder mistreatment scenarios. The Bias score, as discussed previously, is a combination of both pro-aged and anti-aged bias and only includes incorrect responses. The majority of officers had a netnegative aged bias, but the range was -.68 to .58. Pro-aged beliefs, comprised of more favorable attitudes toward older adults, might even be thought to result in a compassionate response, with officers responding in a more helpful manner. However, as noted in Chapter 4, as the Bias score increased, the number of protective response behaviors to the scenarios decreased. In other words, as officers' incorrect beliefs became increasingly more unrealistically positive toward older adults, their responsiveness to their needs decreased. Palmore (1998) believed that pro-aged bias neutralized the effects of anti-aged bias. While the findings of the current study do not directly contradict this view, it was unanticipated that when officers held a net negative bias, they were more likely to respond to the needs of older adults than those who held unrealistic, positive views of older adults. This finding suggests as Bias scores increased officers were more likely to believe older adults were not vulnerable and did not require assistance. This possible bias also meant they were less likely to recognize what type of care or services was needed. This lack of recognition may have compounded older adults' feelings of vulnerability. Given that elders fear police intervention may result in a loss of their caregivers or lead to institutionalization (Faulkner, 1982), as well as the lack of acknowledgement of their vulnerable status, less police responsiveness is likely to increase elders' distrust for the system and negative attitudes toward officers. At that point, elders would be even less likely to cooperate with a system that does not respond to their needs. These findings 112 support Palmore's contention that any inaccurate view, even those thought to be favorable, are ageist and likely to result in harm to elders. This finding emphasizes the assertion that it is of critical importance for officers to have accurate knowledge about older adults. While it is difficult to uncover all potential effects, bias, as measured by the FAQ1, may be helpful in understanding how law enforcement officers respond to elder mistreatment scenarios. In the case of a vulnerable elder, it is often not enough to simply recognize that a crime may have occurred, such as credit card fraud, but rather to recognize that elders are vulnerable and may require additional measures to ensure their safety and well-being. Overall, decisions based on incorrect knowledge, regardless of how favorable the bias, result in less effective responses. Holding a pro-aged bias toward elders did not mean that officers were better responders; instead, they were worse. This study found that the goal to deter elder mistreatment via the justice system is compromised when an officer holds any inaccurate or biased beliefs. The last hypothesis tested was whether law enforcement officers who reported having had training would have higher response scores on elder mistreatment scenarios than those who reported no training. This was found to be insignificant. Yet, when Training alone was removed from the regression, Training Recency was significant at the 0.1 level. Officers who were trained recently were more likely to respond appropriately to the mistreatment scenarios. It is certainly reasonable to assume that officers who have been trained more recently are likely to retain some part of that information and to know the desired response. The real question is whether or not having access to that information leads to more desirable responses in the field. 113 Because training consistently surfaced in the literature as the means for changing how officers respond to elder mistreatment (Arcuri, 1981; Blakely & Dolon, 2000; Byers & Hendricks, 1993; Dubble, 2006; Heisler, 2004; Nerenberg, 2002; Payne, 2005; Plotkin, 1996), other indicators of training were included in the analysis. One of these was whether officers recognized the need to apply the vulnerable adult statute. As the dependent variable, use of the vulnerable adult statute (VA Statute) specifically addressed the original concern expressed by UCOA of how to increase prosecutions. When the dependent variable was limited to whether or not an officer would apply the VA Statute, whether officers indicated they had received training was significant. It was interesting that the variable, Training, was significant in predicting the use of the VA statute but none of the other elements of an effective response. It speaks to the fact that an officer who has been trained may correctly recognize when to use the statute regarding the scenarios, but may still not consider the other elements necessary to ensure the well-being of the older person when mistreatment has occurred. Within the literature, when an officer intervenes only to make a charge but does not look at the other needs, such as care and shelter, the elder may feel more vulnerable and thus be reluctant to cooperate with the criminal justice system (Faulkner, 1982). Consistent with the problem of pro-aged bias, this situation may lead to further alienation by the older person regarding involvement of law enforcement. Once again, previous research has indicated that older adults were less willing to report mistreatment when concerned about the loss of their caretaker or their living situation (Faulkner, 1982). These individuals were also less likely to press charges and more likely to view law enforcement in a negative light if they perceived the officer's response only 114 compounded their problem. Older adults may perceive officers' lack of real interest in their well-being. When a relationship of trust is not established, older adults are not confident that the criminal justice system, as characterized by the gatekeepers, will be able to improve their situation and are less likely to cooperate (Payne, 2005). Therefore, training to the VA statute without improving recognition of the overall needs of older adults should not be expected to increase prosecutions and may, in fact, exacerbate the problem. When the variable of Supervisory Status was removed in the regression model, Training Interest became significant to the 95% confidence level (see Table 4.9). Training Interest was shown to predict a significant increase in the dependent variable, Scenario Response. When officers reported they were interested in training, their scenario scores improved by almost one point. Those officers who expressed an interest in training, on average, included one additional protective behavior in response to the elder mistreatment scenarios. Because this significance occurred when Supervisory Status was removed from the regression, it was thought this might be explained by a binary correlation between Supervisory Status and Training Interest. It was possible that Training Interest was merely a result of being supervisor. Yet, no significant correlation was found (see Appendix I). Further, Training Interest alone was more significant than whether participants reported they were familiar with or had applied the statute. While actual evidence of training regarding the statute had some importance, the attitude toward that training was more significant. A positive attitude toward training may have been an indicator of a positive attitude toward working with older adults. Previous research 115 indicates that attitude towards older adults is important for police officers working with older adults (Daniels et al., 1999; Sykes, 1977). Thus, it might be that interest in training also reflected less bias toward an older adult and signaled that the officer would be more willing to work with this population. More than simply understanding how officers responded, this information could assist in identifying officers who are more prepared to work with older adults experiencing mistreatment. For this purpose, asking whether an officer is interested in training regarding elder mistreatment may be a simple screening tool for deciding whom to include in working with this population when the possibility exists to create specialized teams. In terms of the relationship of an officer's use of the vulnerable adult statute with the FAQ1, the Knowledge score was a significant predictor of whether or not officers would use the vulnerable adult statute. This finding relates directly to the question originally posed by UCOA, "Are there significant predictors of why elder mistreatment cases are not being prosecuted in the criminal justice system?" Again, officers' knowledge of older adults was positively correlated with their response to mistreatment. An increase in knowledge about older adults predicted increased use of the statute as well as other appropriate responses to the scenarios. Knowing that scores on the FAQ1 are correlated with more effective responses, this measure could be used as a quick screening tool to determine whether officers are in need of further training with respect to elder mistreatment response and more specifically whether they would be more likely to use the statute more proactively. Officers who are found to lack knowledge of older adults could be referred for the older adult specific training and thus increase the likelihood of 116 effective responses. Strengths and Limitations This study has several strengths. Most important, to date no studies have examined the interaction between ageism and law enforcement response to elder mistreatment. Despite the increase in elder mistreatment cases and the ongoing concern over lack of response, no previous studies have attempted to assess the correlation between these two factors. One of the barriers in the past to answering any question relating to law enforcement response was the lack of access to law enforcement data. This study achieved that access by drawing heavily on the collaboration between law enforcement and the Utah Commission on Aging. The data from the UCOA Survey provided the most comprehensive access to law enforcement officers' responses to date regarding elder mistreatment. With a population of approximately 275 officers at the time of the study, the response rate was almost 73% (200 participants). While not all surveys were completed, the minimum number of usable responses more than doubled the number accessed by a similar studies (Davies, 2010; Sever & Youdin, 2006). Further, the use of the Palmore FAQ1, while not unanimously agreed upon as the ideal instrument for assessing knowledge of older adults, has been the standard used in hundreds of studies for more than 30 years. Because the FAQ1 provides a more specific measure for assessing age bias, in addition to that of knowledge regarding older adults, it is the only brief measure of its kind. Thus it provided an efficient proxy variable for ageism and provided a stable measure of the extent to which the lack of knowledge of older adults identified by the FAQ1 was biased. While not conclusive, the results of this 117 study can thus be regarded as an estimate of the role ageism plays in how police respond to elder mistreatment. The study also provides a model or strategy for assessing age bias in other institutions tasked with responding to the needs of older adults. There is ample evidence that social service agencies, healthcare professionals, educators, and a host of other institutions must confront the effects of ageism. Social workers are clearly not immune to this issue and will need to continue to address ageism's effect on their practice. To date, no other brief established measure exists for this purpose although there have been some efforts to create such an instrument (Nakao, Damron-Rodriguez, Lawrance, & Volland, 2013). As with any study, there are limitations worth noting. First, the population from which the sample was drawn cannot be considered representative of the range of all law enforcement agencies. The population used for this study has its own unique attributes, and the findings and conclusions of this study should be limited to the immediate population from which the sample was taken. Any generalization to other law enforcement agencies must be done with caution. One limitation with the sample is that categories for several of the variables, such as the number of officers who were female or had only completed high school, were fairly small. This sample size limits the confidence in how representative or how generalizable findings related to these specific factors are. An additional limitation is that this was the first use of the elder mistreatment scenarios. Consequently, reliability and validity have not been established; this is the first attempt to do so. While the mechanism to analyze and assign values to each 118 officer's response is rational and consistent, having been overseen by experts in the field of criminology, there may be disagreements in how the officers' responses should be evaluated. A related concern was that the first portion of the survey focused on whether or not law enforcement officers had received training regarding the abuse, neglect, or exploitation of vulnerable adults. This section questioned officers' knowledge regarding the vulnerable adult statute, the definition of a vulnerable adult, and whether officers had applied the statute during arrests. The content of these questions could have significantly affected how officers responded to the elder mistreatment scenarios. It was surprising that officers appeared to not recognize that the scenarios drew directly on information contained in the response choices to questions in this section. Another issue of concern is the assumption that the FAQ1 is a valid and reliable measure of knowledge of older adults or age bias. The literature on this point was inconclusive and warrants caution. While this is a concern, it does not limit the confidence in any assertion that ageism as measured by the FAQ1 can predict officer response. Therefore, future research may challenge the assumptions of the current study and may choose to either validate the FAQ1 as a measure of either knowledge of older adults or bias or may choose to use a different instrument altogether. The value here is that the data from the UCOA study can still provide a realistic assessment of the link between ageist beliefs and how law enforcement officers describe potential responses to scenarios of elder mistreatment. Specific to the FAQ1, the only other study that examined how training affected officers' knowledge of older adults and aging (Sever & Youdin, 2006) found that officers responded correctly about 66% of the time on average, 119 though the instrument was different and the scoring strategy was much more complicated. Although not using the FAQ1, Sever and Youdin call into question how best to measure officers' knowledge of older adults. Future researchers will need to grapple with this inconsistency. Further, it is important to understand that while law enforcement officers constitute one of the first lines of defense with respect to preventing elder mistreatment, an increase in the number of cases that are charged and arrested under the vulnerable adults statute will not completely overturn the lack of prosecution previously noted. Clearly, this is a multifaceted problem with numerous potential solutions and avenues to pursue, and improving law enforcement response is one small piece in a much larger puzzle. The literature shows that as law enforcement officers have negative experiences with the criminal justice system's response to elder mistreatment, they become less likely to engage and, therefore, do not refer to that system (Dubble, 2006). When officers make an arrest and the prosecution subsequently drops the ball, officers become frustrated and less willing to continue working with other agencies (Dubble, 2006). The UCOA survey did not address this dynamic. There was no way to assess whether this may have been an important factor. Another limitation of the study is that it addresses only one of the many facets of the complex legal system that are required for responding to elder mistreatment. The study does very little to shed light on how cases are prosecuted once they have been passed on from law enforcement. While there have been a number of arrests reported by UCOA, these have not led to prosecutions. The instrument used in this study did not address the correlation between arrests and prosecutions. Therefore, this study can only 120 address officers' responses to the scenarios rather than answering the broader question of why elder mistreatment cases are not being prosecuted once they have entered the judicial process. Future researchers will need to explore the interaction between law enforcement and prosecution. The literature indicates that it is important to find out why prosecuting attorneys are not pursuing cases referred by law enforcement. The research by Dubble (2006) strongly suggests that this lack of prosecution negatively affects law enforcement officers' confidence in and satisfaction with the system's response to elder mistreatment and results in less responsiveness by officers. Conclusion and Recommendations As mentioned in Chapter 2, the older adult population is increasing at a rapid rate and is expected to reach more than 72 million by the year 2030. While the number of elders experiencing mistreatment is unclear, what is known is that the numbers have increased as the population has grown (Acierno et al., 2010). From the FAQ1 scores, it seems that officers not only reflect the bias of the broader society, but they are also less familiar with elders and the aging process than society as a whole and may possibly be more biased. As noted above, cognitive theories suggest that attitudes toward older adults once acquired are quite resistant to change (Wigboldus, Dijksterhuis, & Van Knippenberg, 2003). There is also a history of limited funds to help mitigate elder mistreatment issues (Elder Justice Coalition, 2011). This study suggests that training officers to use the statutes intended to resolve this problem may not be an effective nor efficient solution. Of the few studies that have directly addressed police response to elder mistreatment, all have identified training as an important mechanism for improving 121 response to elder mistreatment (Blakely & Dolon, 1989; Bonnie & Wallace, 2003; Daniels et al., 1999; Heisler & Stiegel, 2004; Quinn & Heisler, 2004; Payne, Berg, Toussaint, 2001; Plotkin, 1988). But, little has been done to evaluate whether or not training makes a difference in police response. This study suggests it is more effective to provide holistic training that educates on all aspects of older adults and leads to a greater understanding of the population rather than focusing on the statute alone. Further, targeted efforts to identify those who are interested in training and already possess accurate knowledge of elders will result in greater efficiency than blanket approaches when funding is limited. The FAQ1 could then be used as an indicator of whether training has been effective regarding the significant variable of accurate knowledge of older adults. Once officers in need of training have been identified and provided with accurate knowledge of older adults, then specialized units could be created capable of overcoming current barriers that have prevented effective police response to elder mistreatment. These specialized units would be encouraged to partner with allied agencies and the elder community as a way to encourage better relations and trust between older adult citizens and those charged with protecting them. As noted previously, Server and Youdin (2006) indicated accurate knowledge of older adults increased with any training intended to provide officers with a general knowledge of older adults. Thus, law enforcement agencies could choose from a variety of training methods to help officers increase their FAQ1 scores (Sever & Youdin, 2006) based on the needs of the agency and the constraints of budget. They could choose to give a small group in-depth training, such as a course in gerontology, or they may choose 122 to use in-service training to give all officers a better understanding of the elder population. The findings of this study suggest it would be financially prudent to offer training focused on older adult knowledge to those who are interested in such training. Officers could be concurrently trained in the use of the vulnerable adult statute and evaluated to ascertain their integration of both elder and vulnerable adult statute knowledge. The FAQ1 along with mistreatment scenarios could then be reapplied to see if there is a significant change in officer response. Ultimately each agency would need to determine which approach best meets their needs based on resources and availability. Preemptively, officers' responses to the FAQ1 could be used to anticipate how they will act when confronted with cases of elder mistreatment. By having officers complete a brief survey, similar to the one used in the UCOA study, law enforcement agencies would be able to screen officers to determine whether they are good candidates for working with older adults. Once identified, agencies could require officers deficient in knowledge of older adults to receive needed training. Implications for Social Work Practice Moving beyond the implications for law enforcement to the broader elder mistreatment response system, it is critical to examine the role of social work. Adult protective services (APS) is the other principal responder tasked with addressing elder mistreatment. In light of the pervasive nature of ageism and its effect on law enforcement officers, an examination of the effect of age bias should be considered for social workers as well. Existing evidence suggests social workers are also vulnerable to ageist views and are not significantly different in their knowledge of older adults from health care 123 professionals (Allen, Cherry, & Palmore, 2009; Kane, 2004, 2007; Wang, 2013). This research suggests that social workers working in adult protection may have the same deficiencies as police officers working with elder mistreatment. In helping to reduce the bias toward older adults, the importance of knowledge of older adults is likely to be similar for social workers. Researchers interested in the response to elder mistreatment must inevitably examine this relationship. Once again, it is important to note the relationship between law enforcement and social work with respect to elder mistreatment. Law enforcement has often looked to social work for training in how to respond to elder mistreatment. Typically that training has included information regarding the statutes and the nuances of effective response. Most often it has been void of content regarding general knowledge of older adults and of the biased and stereotypical beliefs that lead to ageist behavior. Social workers involved in the training of law enforcement officers must recognize and emphasize the importance of accurate knowledge of older adults as well as familiarity with statutes in order to effectively respond to elder mistreatment. To this end social work is well positioned. Because of its emphasis on the dignity and worth of clients and on the strengths perspective, social workers understand the critical importance of moving beyond categories and stereotypes to recognizing the value of individuals and their unique needs. Further, the principles that every client has worth and is worthy of respect and dignity could enhance the entire response system. The fact that elders have experienced ageism at the hands of social workers as well as law enforcement professionals suggests that social workers in this practice area have work to do. 124 It is fascinating that as the social work profession moves forward in understanding elder mistreatment, the findings of this study suggest that in order to effectively work with older adult clients, first establishing older adult knowledge is imperative. To that end the ongoing call for increased integration and infusion of gerontological content into the curricula of schools of social work must be answered (Berkman, Silverstone, Simmons, Volland, & Howe, 2000; Greenfield, Morton, Birkenmaier, & Rown, 2013; Rosen, Zlotnik, & Singer, 2003; Scharlach, Damron-Rodriguez, Robinson, & Feldman, 2000). There is a growing body of research that demonstrates the effectiveness of efforts to provide training specific to working with elders (Chapin, Nelson-Becker, Gordon, Landry, & Chapin, 2007; Cummings, Adler, & DeCoster, 2005; Cummings, Cassie, Galambos, & Wilson, 2006; Greenfield et al., 2013; Lee, Damron-Rodriguez, Lawrance, & Volland, 2009). Any approach must address the existing bias against older adults by both students and faculty and must focus on a greater infusion and integration of gerontological content within generalist social work programs, both at the baccalaureate and masters level (Barusch, 2009; Kropf, 2003; Paveza & VandeWeerd, 2006; Rosen et al., 2003). Without first addressing the bias and negative perceptions that exist among social work faculty and students, there will be little purpose in providing specialized training as there will continue to be little interest in such opportunities (Anderson, & Wiscott, 2004). For response to elder mistreatment to be effective, responders must not accept older adults despite their difference but because of their difference. As long as professionals lack correct knowledge of older adults, a medium for bias flourishes that perpetuates a response system that mirrors the ageist values of society. In that climate, change will not 125 occur, reporting will not increase, and prosecution will remain the exception. Once again, it is worth noting that whether an officer reported having received training was less predictive than an interest in training as to how they responded to the elder mistreatment scenarios. In other words, attitude is critical. Moreover, because attitude is so important, it is worth considering how ageism is not only affecting officer response, but also all elder mistreatment stakeholders. These findings indicate that future studies should explore how ageism affects social workers and other members of the elder mistreatment response system. This would include the elders who rely on it for safety and protection as well as other community agencies such as APS and other members of the criminal justice system. The original question of this study, "Does ageism affect law enforcement response to elder mistreatment?" has been answered affirmatively. This study concludes that the most likely mechanism for improving police response to elder mistreatment scenarios is to ensure officers have accurate knowledge of older adults. It is equally important for other members of the elder mistreatment system to have accurate knowledge of older adults as a first step to root out the ageist views that pervade the response system. It is the conclusion of this study that until ageism is addressed, more training and laws will not be effective. Social learning theory suggests that as effective role models emerge, the culture of an agency and the overall response system may move away from an ageist view of older adults toward a more realistic view. Social workers are well positioned to lead in providing educational resources and modeling those behaviors that will lead to effective responses to the problems experienced by elders. Reducing the ageism that occurs as law 126 enforcement officers, and other agents and agencies respond to elder mistreatment will ultimately have positive repercussions throughout society. This in turn will help to produce the most critical outcome of all, an increase in the safety and well-being for those the system is intended to serve, older adults. APPENDIX A DEFINITIONS 128 SEC. 144. PREVENTION OF ABUSE OF OLDER INDIVIDUALS. (a) DEFINITIONS.-S~C~3~0O2 ~of the Older Americans Act of 1965 (42 U.S.C. 3022), as amended by section 136(a), is amended by adding at the end the following: ''(15) The term 'abuse' means the willful- "(A) infliction of injury, unreasonable confinement, intimidation, or cruel punishment with resulting physical harm or pain or mental anguish; or "(B) deprivation by a caretaker of goods or services which are necessary to avoid physical harm, mental anguish, or mental illness. "(16) The term 'elder abuse' means abuse of an older individual. "(17) The term 'caretaker' means an individual who has the responsibility for the care of an older individual, either voluntarily, by contract, receipt of patient for care, as a result of family relationship, or by order of a court of competent jurisdiction. "(18) The term 'exploitation' means the illegal or improper act or process of a caretaker using the resources of an older individual for monetary or personal benefit, profit, or gain. "(19) The term 'neglect' means the failure to provide for oneself the goods or services which are necessary to avoid physical harm, mental anguish, or mental illness or the failure of a caretaker to provide such goods or services. "(20) The term 'physical harm' means bodily pain, injury, impairment, or disease. APPENDIX B UCOA SURVEY 130 UCOA Survey Abuse, neglect, or exploitation of a vulnerable adult Utah Statute 76-5-111 A Survey for Law Enforcement Officers Prepared For: Utah Commission on Aging-Public Safety Work Group Tanta Lisa Clayton-Chair Draft designed by: Wendy M. Thomas University of Utah-Gerontology Interdisciplinary Program 131 We would like to begin by asking you some specific questions that relate to elder abuse/neglect/exploitation of older adults in the state of Utah. Please select the one response that you think is the best answer. 1. Have you received training regarding the abuse, neglect or exploitation of a vulnerable adult? a. yes, no further training on this topic is needed b. yes, and would like further training on this topic c. no, not interested in training on this topic d. no, but would like training on this topic 2. If you answered "Yes" to the prior question, please approximate when the training took place. a. within last six months b. six months to one year ago c. one to two years ago d. two to three years ago e. over three years ago 3. Are you familiar with the legal definition of a vulnerable adult? a. Yes b. No c. I think so, but am not certain 4. Are you familiar with Utah Statute 76-5-111, Abuse, neglect, or exploitation of a vulnerable adult? a. no 132 b. somewhat familiar c. yes, but have not applied it during arrests d. yes, have applied it during arrests 5. If you have applied 76-5-111 in an arrest, in which location(s) did the arrest take place? a. private Residence b. Long Term Care Facility (nursing home, assisted living, etc.) c. hospital d. other (Please indicate): _____________________________________ e. I have not applied 76-5-111 Please indicate which answers below best complete the legal definition of "vulnerable adult" as defined in UCC 76-5-111. Circle all answers that apply. 6. "Vulnerable Adult" means an elder adult (65+) or an adult 18 years of age and older who has a mental and physical impairment which substantially affects the person's ability to: a. Provide personal protection b. Provide necessities such as food, shelter, clothing, or medical or other health care c. Obtain services necessary for health, safety, or welfare d. Carry out the activities of daily living (ADL) e. Manage the adult's own resources f. Comprehend the nature and consequences of remaining in a situation of abuse, neglect, or exploitation 133 Next, we will ask you to complete four case studies. 7. James is a 79 year-old male who is blind. He sends his neighbor Sue to the store with his credit card to buy his cigarettes. Family members discovered $1,500 charged to his card at Wal-Mart for gas, fast food, and a television. James says he never gave anyone permission to use his card to buy anything other than his cigarettes. James and his family report this to the proper authorities for investigation. What should the responding officer do? 8. Eric is a 45 year-old, who functions on a 12-year old level. He lives with his older brother, Frank. Eric and Frank got into an argument and Frank punched Eric in the face with a closed fist. Eric develops a black eye from the punch. The injury is called in to property authorities when observed at Eric's day program. What should the responding officer do? 9. Mary is a 58-year-old who has severe MS and is bed-bound but she is mentally alert and she often talks to friends on the phone. A home-health agency cares for Mary, but she got angry at an aide and fired the agency. The agency was providing all her needs including food, medications, and toileting. After three days without care, Mary calls 911 and reports that the agency is committing criminal neglect. What should the responding officer do? 10. Ethyl is an 83-year old grandmother who lives with her granddaughter, Megan. A Meals-on-Wheels driver recognized Ethyl walking down the block from her home on a busy street and pulled up next to her. Megan arrived shortly after the driver found Ethyl and said, "You better not be f***ing leaving the house or I will put you in a nursing home." She yelled and swore at Ethyl for five minutes, while 134 Ethyl cried and shook. When Adult Protective Services (APS) interviewed Ethyl, she said she was humiliated. APS contacted law enforcement. What should the responding officer do? Listed below are general questions about aging and older adults. Please select the one response you think is the best answer. 11. The proportion of people over 65 who are senile (have impaired memory, disorientation, or dementia) is: a. about 1 in 100 b. about 1 in 10 c. about 1 in 2 d. the e. majority 12. The senses that tend to weaken in old age are: a. sight and hearing b. taste and smell c. sight, hearing, and touch d. all five senses 13. The majority of old couples: a. have little or no interest in sex b. are not able to have sexual relations c. continue to enjoy sexual relations d. think sex is only for the young 14. Lung vital capacity in old age: 135 a. tends to decline b. stays about the same among non-smokers c. tends to increase among healthy old people d. is unrelated to age 15. Happiness among old people is: a. rare b. less common than among younger people c. about as common as among younger people d. more common than among younger people 16. Physical strength: a. tends to decline with age b. tends to remain the same among healthy old people c. tends to increase among healthy old people d. is unrelated to age 17. The percentage of people over 65 in long-stay institutions (such as nursing homes, mental hospitals, and homes for the aged) is about: a. 5% b. 10% c. 25% d. 50% 18. The accident rate per driver over age 65 is: a. higher than for those under 65 b. about the same as for those under 65 136 c. lower than for those under 65 d. unknown 19. Most workers over 65: a. work less effectively than younger workers b. work as effectively as younger workers c. work more effectively than younger workers d. are preferred by most employers 20. The proportion of people over 65 who are able to do their normal activities is about: a. one-tenth b. one-quarter c. one-half d. three-fourths 21. Adaptability to change among people over 65 is: a. rare b. present among about half c. present among most d. more common than among younger people 22. As for old people learning new things: a. most are unable to learn at any speed b. most are able to learn, but at a slower speed c. most are able to learn as fast as younger people d. learning speed is unrelated to age 137 23. Depression is more frequent among: a. people over 65 b. adults under 65 c. young people d. children 24. Old people tend to react: a. slower than younger people b. at about the same speed as younger people c. faster than younger people d. slower or faster than younger people, depending on the type of test 25. Old people tend to be: a. more alike than younger people b. the same as younger people in terms of alikeness c. less alike than younger people d. more alike in some respects and less alike in others 26. Most old people say: a. they are seldom bored b. they are sometimes bored c. they are often bored d. life is monotonous 27. The proportion of old people who are socially isolated is: a. almost all b. about half 138 c. less than a fourth d. almost none 28. The accident rate among workers over 65 tends to be: a. higher than among younger workers b. about the same as among younger workers c. lower than among younger workers d. unknown because there are so few workers over 65 29. The proportion of the U.S. population now age 65 or over is: a. 3% b. 13% c. 23% d. 33% 30. Medical practitioners tend to give older patients: a. lower priority than younger patients b. the same priority as younger patients c. higher priority than younger patients d. higher priority if they have Medicaid 31. The poverty rate (as defined by the federal government) among old people is: a. higher than among children under age 18 b. higher than among all persons under 65 c. about the same as among persons under 65 d. lower than among persons under 65 32. Most old people are: a. employed 139 b. employed or would like to be employed c. employed, do housework or volunteer work, or would like to do some kind of work d. not interested in any work 33. Religiosity tends to: a. increase in old age b. decrease in old age c. be greater in the older generation than in the younger generations d. be unrelated to age 34. Most old people: a. are seldom angry b. are often angry c. are often grouchy d. often lose their tempers 35. The health and economic status of old people (compared to younger people) in the year 2010 will: a. be higher than now b. be about the same as now c. be lower than now d. show no consistent trend 140 Law Enforcement Profile 1. Years in Law Enforcement:___________ 2. Highest position served: ___________________________________ 3. Locations where served: ___________________________________ 4. What is your age? __________ 5. Gender? _________ Male ___________ Female 6. What is the highest level of your education? (please circle your response) e. High School Graduate f. Partial College (at least one year) or specialized training g. Standard College or University graduation h. Graduate professional training (Graduate degree) APPENDIX C VULNERABLE OLDER ADULT SCENARIOS 142 1. James is a 79 year-old male who is blind. He sends his neighbor Sue to the store with his credit card to buy his cigarettes. Family members discovered $1,500 charged to his card at Wal-Mart for gas, fast food, and a television. James says he never gave anyone permission to use his card to buy anything other than his cigarettes. James and his family report this to the proper authorities for investigation. What should the responding officer do? 2. Ethyl is an 83-year old grandmother who lives with her granddaughter, Megan. A Meals-on-Wheels driver recognized Ethyl walking down the block from her home on a busy street and pulled up next to her. Megan arrived shortly after the driver found Ethyl and said, "You better not be f***ing leaving the house or I will put you in a nursing home." She yelled and swore at Ethyl for five minutes, while Ethyl cried and shook. When Adult Protective Services (APS) interviewed Ethyl, she said she was humiliated. APS contacted law enforcement. What should the responding officer do? APPENDIX D PALMORE FACTS ON AGING QUIZ 144 Palmore Facts on Aging Quiz (FAQ1) Multiple Choice 1. The proportion of people over 65 who are senile (have impaired memory, disorientation, or dementia) is: a. About 1 in 100 + b. about 1 in 10 c. about 1 in 2 * - d. the majority - 2. The senses that tend to weaken in old age are: a. sight and hearing + b. taste and smell + c. sight, hearing, and touch + d. all five senses * 3. The majority of old couples: a. have little or no interest in sex - b. are not able to have sexual relations c. continue to enjoy sexual relations - * d. think sex is only for the young - 4. Lung vital capacity in old age: a. tends to decline * b. stays about the same among non-smokers c. tends to increase among healthy old people d. is unrelated to age + 5. Happiness among old people is: a. rare - + + 145 b. less common than among younger people - c. about as common as among younger people d. more common than among younger people * + 6. Physical strength: a. tends to decline with age * b. tends to remain the same among healthy old people + c. tends to increase among healthy old people d. is unrelated to age + + 7. The percentage of people over 65 in long-stay institutions (such as nursing homes, mental hospitals, and homes for the aged) is about: a. 5% * b. 10% - c. 25% - d. 50% - 8. The accident rate per driver over age 65 is: a. higher than for those under 65 - b. about the same as for those under 65 - c. lower than for those under 65 * d. unknown 0 9. Most workers over 65: a. work less effectively than younger workers b. work as effectively as younger workers * c. work more effectively than younger workers d. are preferred by most employers + - + 146 10. The proportion of people over 65 who are able to do their normal activities is about: a. one-tenth - b. one-quarter c. one-half - - d. three-fourths * 11. Adaptability to change among people over 65 is: a. Rare - b. present among about half c. present among most - * d. more common than among younger people + 12. As for old people learning new things: a. most are unable to learn at any speed - b. most are able to learn, but at a slower speed * c. most are able to learn as fast as younger people d. learning speed is unrelated to age + 13. Depression is more frequent among: a. people over 65 - b. adults under 65 * c. young people 0 d. children 0 14. Old people tend to react: a. slower than younger people * b. at about the same speed as younger people + + 147 c. faster than younger people + d. slower or faster than younger people, depending on the type of test 15. Old people tend to be: a. more alike than younger people - b. the same as younger people in terms of alikeness c. less alike than younger people 0 + d. more alike in some respects and less alike in others * 16. Most old people say: a. they are seldom bored * b. they are sometimes bored c. they are often bored d. life is monotonous - - - 17. The proportion of old people who are socially isolated is: a. almost all - b. about half - c. less than a fourth d. * almost none 18. The accident rate among workers over 65 tends to be: a. higher than among younger workers - b. about the same as among younger workers c. lower than among younger workers - * d. unknown because there are so few workers over 65 19. The proportion of the U.S. population now age 65 or over is: - + 148 a. 3% 0 b. 13% * c. 23% 0 d. 33% 0 20. Medical practitioners tend to give older patients: a. lower priority than younger patients * b. the same priority as younger patients + c. higher priority than younger patients + d. higher priority if they have Medicaid + 21. The poverty rate (as defined by the federal government) among old people is: a. higher than among children under age 18 b. higher than among all persons under 65 - - c. about the same as among persons under 65 - d. lower than among persons under 65 * 22. Most old people are: a. employed + b. employed or would like to be employed + c. employed, do housework or volunteer work, or would like to do some kind of work * d. not interested in any work - 23. Religiosity tends to: a. increase in old age 0 b. decrease in old age 0 c. be greater in the older generation than in the younger generations * 149 d. be unrelated to age 0 24. Most old people: a. are seldom angry * b. are often angry - c. are often grouchy - d. often lose their tempers - 25. The health and economic status of old people (compared to younger people) in the year 2010 will: a. be higher than now * b. be about the same as now c. be lower than now - d. show no consistent trend - - APPENDIX E SCENARIO SCORING RUBIC 151 Scenario Scoring Rubric The rubric represents how the responses were scored and reflects the importance responsive, thorough and appropriate police work regardless of whether the responding officer recognized that a case represented a violation of the laws protecting vulnerable older people. Also if the officer recognized that vulnerability did he take steps to protect and insure safety, acknowledge the need to include APS, and recognize that the vulnerable adult statute should be applied to the case. The law requires that when a law enforcement officer becomes aware of a case of vulnerable adult mistreatment, he must notify APS. 1. Case was viewed as a civil matter requiring at most documentation of the report. 0 pts. 2. The scenario represents a criminal case which should be investigated. 1pt 3. Officer recognized the situation warranted an immediate and personal investigation of the facts rather than passing it on to another investigator. 1pt 4. The officer recognized that perpetrator needed to be charged or arrested. 1pt 5. The officer recognized the need to involve or work with APS. 1pt 6. Steps to protect and insure the victim were taken independent of involving APS. 1pt 7. The officer's response recognized that the case violated the vulnerable adult statute. 1 pt Elements Point value Officer response 0 Matter viewed as a civil requiring at most a report. 1 Case viewed as criminal requiring investigation. 1 situation warranted an immediate and personal investigation 1 Recognized perpetrator needed to be charged or arrested. 1 Recognized the need to involve or work with APS. 1pt 1 Steps taken to protect and insure safety of the victim in addition to APS. 1 recognized that the case violated the vulnerable adult statute APPENDIX F REPORTING REQUIREMENTS FOR LAW ENFORCEMENT 153 Title 62A Utah Human Services Statute Chapter 3 Aging and Adult Services Section 305 Reporting requirements-Investigation-Immunity-Violation-Penalty- Nonmedical healing. 62A-3-305. Reporting requirements-Investigation-Immunity-Violation-Penalty- Nonmedical healing. (1) A person who has reason to believe that a vulnerable adult has been the subject of abuse, neglect, or exploitation shall immediately notify Adult Protective Services intake or the nearest law enforcement agency. When the initial report is made to law enforcement, law enforcement shall immediately notify Adult Protective Services intake. Adult Protective Services and law enforcement shall coordinate, as appropriate, their efforts to provide protection to the vulnerable adult. (2) When the initial report or subsequent investigation by Adult Protective Services indicates that a criminal offense may have occurred against a vulnerable adult: (a) Adult Protective Services shall notify the nearest local law enforcement agency regarding the potential offense; and (b) the law enforcement agency may initiate an investigation in cooperation with Adult Protective Services. APPENDIX G STRATEGY FOR RESPONSES TO THE TRAINING ITEM CONCERNING RANK, "HIGHEST POSITION SERVED" 155 Strategy for Responses to the Training Item Concerning Rank, "Highest Position Served Strategy for responses to the training item concerning rank, "highest position served" Actual Responses Frequency Collapsed Groupings and Frequency Chief Chief (Police Chief) Captain Lieutenant 1 1 2 4 Patrol Sergeant/shift Supervisor Sergeant Supervisor First line supervisor Corporal Senior Officer 1 22 1 1 1 1 Deputy/Detective Detective Detective/Officer Patrol/Detective Patrol/Investigations 2 16 1 1 1 Detectives and Investigation 21 Deputy Deputy Sheriff Deputy/Officer Officer Line Officer Deputy/Patrol Patrol 87 7 5 2 2 2 12 Deputy /Officer / Line Chiefs, Captains and Lieutenants 8 Final Grouping and Frequency Commanding and Supervising Officers 35 Sergeants, Corporals, Senior Officers and Supervisors 27 Non-Commanding Officers 138 117 APPENDIX H RESULTS FOR LINEAR (OLS) REGRESSION USED FOR PREDICTING POLICE OFFICER RESPONSE TO SCENARIOS OF ELDER MISTREATMENT- BOTH FOCAL AND CONTROL VARIABLES 157 Results for linear (OLS) regression used for predicting police officer response to scenarios of elder mistreatment -- both focal and control variables. FAQ1 Knowledge Bias Training Yes / no Interest Recency Familiar with legal definition Familiar with the Utah Vulnerable Adult Statute Yes familiar with statute vs. no or only somewhat Familiar-have not applied Applied statute "Vulnerable Adult" affects the person's ability Years in law enforcement Rank Highest rank Supervisory Status role vs. rank and file Commanders, Supervisors and Detectives vs. Deputies Age Gender Education Highest level College degree vs. No degree ** Significance at the p < .01 level. * Significance at the p < .05 level. † Significance at the p < 0.10 level. B Constant RSquare Significance Valid n .200** -.840 2.937 5.128 .055 .009 .004 .255 149 152 .376 1.010** .162 .250 5.051 4.509 5.048 4.974 .008 .050 .012 .008 .224 .003 .144 .225 180 180 180 180 .390* 4.821 .028 .024 180 .660† 5.069 .021 . 055 179 .417† 5.099 .017 .086 179 .459 .701* 5.207 4.922 .004 .028 .406 .024 179 200 .040† 4.677 .022 .058 165 .354† .887* 4.974 4.992 .024 .030 .052 .030 158 158 .665† 4.963 .022 .062 158 .035† .296 3.791 5.170 .018 .003 .091 .481 157 165 .-.159 -.433 5.586 5.362 .003 .010 .511 .216 161 161 APPENDIX I CORRELATION TABLE FOR THE PREDICTOR VARIABLES Correlation Table for the Predictor Variables Net Bias Knowledge Knowledge 1.0000 Net Bias .1826* Training Training Interest Training Recency 0.0431 0.2139* 1.0000 0.0787 Training Training Interest Training Recency Legal Def. VA Statute Experience Supervisory Status Age Gender 1.0000 0.0968 College 1.0000 0.0081 1.0000 0.0594 0.0602 0.0602 0.8315* 0.0128 1.0000 Legal Def. 0.0022 0.0309 0.2317* -0.0852 0.2431* 1.0000 VA Statute 0.0029 0.1145 .1947* -0.0775 0.1529* Experience Supervisory Status 0.0563 0.0136 0.0462 0.1141 -0.0946 .3641* 0.0017 0.1673* 0.0799 0.0240 0.1458 -0.0106 Age 0.0995 0.0326 -0.0278 0.1035 -0.0673 0.0169 0.0687 Gender -0.1144 0.1491* 0.0049 0.1561* College 0.0017 0.0900 0.0522 0.0356 0.1052 0.0637 1.0000 1.0000 0.0970 0.0822 0.0110 0.0231 0.0115 0.0981 0.0617 .4948* 1.0000 .8268* .4137* 0.0436 -0.0062 1.0000 0.0126 0.0501 0.0591 0.0753 1.0000 ** Significance at the p < .01 level. * Significance at the p < .05 level. † Significance at the p < 0.10 level. 159 REFERENCES Abu-Bader, S. H. (2011). Using statistical methods in social science research: With a complete SPSS guide. Chicago, IL: Lyceum Books. Acierno, R., Hernandez, M., Amstadter, A. B., Resnick, H. S., Steve, K., Muzzy, W., & Kilpatrick, D. G. (2010). Prevalence and correlates of emotional, physical, sexual, and financial abuse and potential neglect in the United States: The national elder mistreatment study. American Journal of Public Health, 100(2), 292-297. doi:10.2105/AJPH.2009.163089 Acierno, R., Hernandez-Tejada, M., Muzzy, W., & Steve, K. (2009). Final report: National Elder Mistreatment Study (Document No. 226456). Washington, DC: National Institute of Justice. Allen, P.D., Cherry, K.E., & Palmore, E. (2009). Self-reported ageism in social work practitioners and students. Journal of Gerontological Social Work, 52(2), 124- 134. doi:10.1080/01634370802561927 Allport, G. W. (1954). The nature of prejudice. Reading, MA: Addison-Wesley. Anderson, D., & Wiscott, R. (2004). Comparing social work and non-social work students' attitudes about aging: Implications to promote work with elders. Journal of Gerontological Social Work, 42(2), 21-36. doi:10.1300/J083v42n02_03 Anetzberger, G. J. (2000). Caregiving: Primary cause of elder abuse. Generations, 24(2), 46-51. Anonymous. (2002a). Protecting older Americans: A history of federal action on elder abuse neglect, and exploitation. Journal of Elder Abuse and Neglect, 14(2/3), 9- 31. doi:10.1300/J084v14n02_03 Anonymous. (2002b). Senator John Breaux: The elder justice proposal of 2002. Journal of Elder Abuse and Neglect, 14(2), 37-85. doi:10.1300/J084v14n02_05 Arcuri, A. F. (1981). The police and the elderly. In D. Lester (Ed.), The elderly victim of crime (1st ed., pp. 106-128). Springfield, IL: Charles C. Thomas. Barusch, A. S. (2009). Love and Ageism-A Social Work Perspective. Social Work Today, 9(1), 12. Retrieved from 161 http://www.socialworktoday.com/archive/011909p12.shtml Bayley, D. H., & Bittner, E. (1984). Learning the skills of policing. Law and Contemporary Problems, 47(4), 35-59. Retrieved from http://www.jstor.org/stable/1191686 Beard, H., & Payne, B. K. (2005). The portrayal of elder abuse in the national media. American Journal of Criminal Justice, 29, 269-284. Berkman, B., Silverstone, B., Simmons, W. J., Volland, P. J., & Howe, J. L. (2000). Social work gerontological practice: The need for faculty development in the new millennium. Journal of Gerontological Social Work, 34(1), 5-23. doi:10.1300/J083v34n01_02 Bergeron, L. R. (2001). An elder abuse case study: Caregiver stress or domestic violence? You decide. Journal of Gerontological Social Work, 34(4) 47-63. doi:10.1300/J083v34n04_05 Blakely, B. E., & Dolon, R. (1998). A test of public reactions to alleged elder abuse. Journal of Elder Abuse and Neglect, 9(4), 43-65. doi:10.1300/J084v09n04_03 Blakely, B. E., & Dolon, R. (2000). Perceptions of adult protective services workers of the support provided by criminal justice professionals in a case of elder abuse. Journal of Elder Abuse and Neglect, 12(3/4), 71-94. doi:10.1300/J084v12n03_04 Blakely, B. E., & Dolon, R. (2003). Another look at the helpfulness of occupational groups in the discovery of elder abuse and neglect. Journal of Elder Abuse and Neglect, 13(3), 1-23. doi:10.1300/J084v13n03_01 Block, M. R., & Sinnott, J. D. (Eds.) (1979). The battered elder syndrome: An exploratory study. College Park, MD: University of Maryland Center on Aging. Bonnie, R. J., & Wallace, R. B. (Eds.). (2003). Elder mistreatment: Abuse, neglect, and exploitation in an aging America. Washington, DC: National Academies Press. Brownell, P. J. (1994). Family crimes against the elderly and the criminal justice system: A study of elder abuse and the New York City police department. Doctoral dissertation. Available from ProQuest Dissertations and Theses database. (UMI No. 304111219) Brownell, P. J. (1998). Family crimes against the elderly: Elderly abuse and the criminal justice system. New York, NY: Garland Publishing. Butler, R. N. (1989). Dispelling ageism: The cross-cutting intervention. Annals of the American Academy of Political and Social Sciences, 503(1), 138-147. doi:10.1177/0002716289503001011 162 Butler, R. N. (1999). Warning signs of elder abuse. Geriatrics, 54(3), 3-4. Retrieved from Academic Search Premier. Butler, R. N. (2005). Ageism: Looking back over my shoulder. Generations, 29(3), 84- 86. Retrieved from Butler, R. N. (2008). Combating ageism. International Psychogeriatrics, 21(02), 211. doi:10.1017/S104161020800731X Butler, R. N., Kim, K., Curran, M., Estrine, J., Gallagher, J., Knapp, K., . . . O'Brien, N. (2006). Longevity research, policy, and education: New York, NY: International Longevity Center. Byers, B., & Hendricks. J. E. (1993). Adult protective services: Research and practice. Springfield, IL: Charles C. Thomas. Chapin, R., Nelson-Becker, H., Gordon, T., Landry, S. T., & Chapin, W. B. (2007). Responding to the Hartford Geriatric Social Work Initiative: A multilevel community approach to building aging competency. Journal of Gerontological Social Work, 50(1-2), 59-74. doi:10.1300/J083v50n01_05 Chermak, S. (1993). Adult protective services and the criminal justice system. In B. Byers, & J. E. Hendricks, (Eds.) Adult protective services: Research and practice (pp. 109-136). Springfield, IL: Charles C. Thomas. Chu, L. D., & Kraus J. F. (2004). Predicting fatal assault among the elderly using the national incident-based reporting system crime data. Homicide Studies, 8(2), 71- 95. doi:10.1177/1088767903262396 Clarke, J. B. (2007). Evaluating the effectiveness and benefit-cost of Michigan background check program using crime opportunity theory. Doctoral dissertation. Available from ProQuest Dissertations and Theses database. (UMI No. 304798924) Connolly, M. T. (2010). Where elder abuse and the justice system collide: Police power, parens patriae, and 12 recommendations. Journal of Elder Abuse and Neglect, 22(1), 37-93. doi:10.1080/08946560903436338 Crystal, S. (1987). Elder abuse: The latest crisis. Public Interest, 1(88), 56-66. Available from http://www.ncbi.nlm.nih.gov/pubmed/10283243 Cuddy, A. J. C., Norton, M. I., & Fiske, S. T. (2005). This old stereotype: The pervasiveness and persistence of the elderly stereotype. Journal of Social Issues, 61(2), 267-285. doi:10.1111/j.1540-4560.2005.00405.x Cummings, S. M., Adler, G., & DeCoster, V. A. (2005). Factors influencing graduate- 163 social-work students' interest in working with elders. Educational Gerontology, 31(8), 643-655. doi:10.1080/03601270591003382 Cummings, S. M., Cassie, K. M., Galambos, C., & Wilson, E. (2006). Impact of an infusion model on social work students' aging knowledge, attitudes, and interests. Journal of Gerontological Social Work, 47(3-4), 173-186. doi:10.1300/J083v47n03_11 Cupello, M. S. (2010). Law enforcement knowledge of elder laws. (Master's thesis). Retrieved from http://content.lib.utah.edu/utils/getfile/collection/ehslGerInt/id/52/filename/49.pdf Daly, J. Jogerst, G., Brinig, M., & Dawson, J. (2003). Mandatory reporting: Relationship of APS statute language on state reported elder abuse. Journal of Elder Abuse and Neglect, 15(2), 1-22. doi:10.1300/J084v15n02_01 Dauenhauer, J. A., Mayer, K. C., & Mason, A. (2007). Evaluation of adult protective services: Perspectives of community professionals. Journal of Elder Abuse and Neglect, 19(3/4), 41-57. doi:10.1300/J084v19n03_03 Daniels, R. S., Baumhover, L. A., Formby, W. A., & Clark-Daniels, C. L. (1999). Police discretion and elder abuse: A nested model of observation, reporting, and satisfaction. Journal of Criminal Justice, 27(3), 209-225. doi:10.1016/S00472352(98)00055-5 Davies, M. (2010). Law enforcement officers' understanding and response to elder abuse. (Master's thesis, California State University, Long Beach). Available from ProQuest Dissertations and Theses. Retrieved from http://search.proquest.com/docview/862563612?accountid=9817. (UMI No. 862563612) Doerner, W. G., & Lab, S. P. (2012). Victimology. Boston: Anderson Publishing Dolon, R., & Blakely, B. E. (1989). Elder abuse and neglect: A study of adult protective services workers in the United States. Journal of Elder Abuse and Neglect, 1(3), 31-49. doi:10.1300/J084v12n03_04 Dolon, R., & Hendricks, J. E. (1989). An exploratory study comparing attitudes and practices of police officers and social service providers in elder abuse and neglect cases. Journal of Elder Abuse and Neglect, 1(1), 75-90. doi:10.1300/J084v01n01_07 Dong, X., Simon, M., Mendes de Leon, C., Fulmer, T., Beck, T., Hebert, L., . . . Evans, D. (2009). Elder self-neglect and abuse and mortality risk in a communitydwelling population. The Journal of the American Medical Association, 302(5), 517-526. doi:10.1001/jama.2009.1109 164 Dubble, C. (2006). A policy perspective on elder justice through APS and law enforcement collaboration. Journal of Gerontological Social Work, 46(3), 35-55. doi:10.1300/J083v46n03_03 Dunlap, E. E., Golding, J. M., Hodell, E. C., & Marsil, D. F. (2007). Perceptions of elder physical abuse in the courtroom: The influence of hearsay witness testimony. Journal of Elder Abuse and Neglect, 19(3), 19-39. doi:10.1300/J084v19n03_02 Elder Justice Coalition. (2011). One year anniversary of the elder justice act. Retrieved from http://www.elderjusticecoalition.com/ Falk, U. A., & Falk, G. (1997). Ageism, the aged, and aging in America: On being old in an alienated society. Springfield, IL: Charles C. Thomas. Faulkner, L. R. (1982). Mandating the reporting of suspected cases of elder abuse: An inappropriate, ineffective and ageist response to the abuse of older adults. Family Law Quarterly, 16(1), 69-91. Retrieved from HeinOnline. Federal Interagency Forum on Aging-Related Statistics. (2010). Population. Older Americans 2010: Key indicators of well-being. Washington, DC: U.S. Government Printing Office. Retrieved from http://www.agingstats.gov Forst, B. E., & Hernon, J. C. (1985). The criminal justice response to victim harm. Washington, DC: U.S. Department of Justice. Greenfield, E. A., Morton, C., Birkenmaier, J., & Rowan, N. L. (2013). Optimizing geriatric social work education: Program and individual characteristics that promote competencies. Journal of Gerontological Social Work, 56(4), 356-377. doi:10.1080/01634372.2013.771807 Gibbs, L. M., & Mosqueda, L. (2007). The importance of reporting abuse of the elderly. American Family Physician, 75(5), 628. Retrieved from http://www.aafp.org/afp/2007/0301/p628.html Goldsmith, J. (1976). Police and the older victim-keys to changing perspective. Police Chief, 43(2), 23. Goldstein, H. (1963). Police discretion: The ideal versus the real. Public Administration Review, 23(3), 140-148. doi:10.2307/973838 Goldstein, J. (1960). Police discretion not to invoke the criminal process: Low-visibility decisions in the administration of justice. The Yale Law Journal, 69(4), 543-594. doi:10.2307/794445 Hamel, R. (1979). Assisting the elderly victim. In A. P. Goldstein, W. J. Hoyer, & P. J. Monti (Eds.), Police and the Elderly (pp. 67-85). New York, NY: Pergamon 165 Press. Harbison, J. (1999). Models of intervention for "Elder abuse and neglect": A Canadian perspective on ageism. Journal of Elder Abuse and Neglect, 10(3), 1-17. doi:10.1300/J084v10n03_01 Harrell, F. E. (2001). Regression modeling strategies: With applications to linear models, logistic regression, and survival analysis. New York, NY: Springer. Harris, D., Changas, P., & Palmore, E. (1996). Palmore's first facts on aging quiz in a multiple-choice format. Educational Gerontology, 22(6), 575-589. doi:10.1080/0360127960220605 Heisler, C. J. (2000). Elder abuse and the criminal justice system: New awareness, new responses. Generations, 24(2), 52. Retrieved from http://search.proquest.com/docview/212219012?accountid=9817 Heisler, C. J. (2012). Elder abuse and the criminal justice system: An uncertain future. Generations, 36(3), 83-88. Retrieved from http://generations.metapress.com/content/20h8n5651h533513/ Heisler, C. J., & Quinn, M. J. (1995). A legal perspective. Journal of Elder Abuse and Neglect, 7(2-3), 131-156. doi:10.1300/J084v07n02_08 Heisler, C. J., & Stiegel, L. A. (2002). Enhancing the justice system's response to elder abuse: Discussions and recommendations of the "improving prosecution" working group of the national policy summit on elder abuse. Journal of Elder Abuse and Neglect, 14(4), 31-54. doi:10.1300/J084v14n04_05 Hodge, P. D. (1998). National law enforcement programs to prevent, detect, investigate, and prosecute elder abuse. Journal of Elder Abuse and Neglect, 9(4), 23-41. doi:10.1300/J084v09n04_02 Hoffman, J. P. (2010). Linear regression analysis: Assumptions and applications. Retrieved from https://sociology.byu.edu/Hoffmann/SiteAssets/Hoffmann%20_%20Linear%20R egression%20Analysis_%20second%20edition.pdf Holtzman, J. M., & Beck, J. D. (1979). Palmore's facts on aging quiz: A reappraisal. The Gerontologist, 19(1), 116-120. doi:10.1093/geront/19.1.116 Jirik, S., & Sanders, S. (2014). Analysis of elder abuse statutes across the United States, 2011-2012. Journal of Gerontological Social Work, 57(5), 478-97). doi:10.1080/01634372.2014.884514 Jogerst, G., Daly, J. M., & Ingram, J. D. (2001). National elder abuse questionnaire: 166 Summary of adult protective service investigator responses. Journal of Elder Abuse and Neglect, 13(4), 59-64. doi:10.1300/J084v13n04_04 Jogerst, G. J., Daly, J. M., Brinig, M. F., Dawson, J. D., Schmuch, G. A., & Ingram, J. G. (2003). Domestic elder abuse and the law. American Journal of Public Health, 93(12), 2131-2136. doi:10.2105/AJPH.93.12.2131 Jogerst, G. J., Daly, J. M., Dawson, J. D., Brinig, M. F., & Schmuch, G. A. (2004). Required elder abuse education for Iowa mandatory reporters. Journal of Elder Abuse and Neglect, 15(1), 59-73. doi:10.1300/J084v15n01_05 Johnson, A. (2002). Geriatric education for elder justice. Journal of Elder Abuse and Neglect, 14(2/3), 185-190. doi:10.1300/J084v14n02_08 Johnson, I. M., Sigler, R. T., & Crowley, J. E. (1994). Domestic violence: A comparative study of perceptions and attitudes toward domestic abuse cases among social service and criminal justice professionals. Journal of Criminal Justice, 22(3), 237-248. doi:10.1016/0047-2352(94)90020-5 Kane, M. N. (2004). Ageism and intervention: What social work students believe about treating people differently because of age. Educational Gerontology, 30(9), 767- 784. doi:10.1080/03601270490498098 Kane, M. N. (2007). Social work and criminal justice students' perceptions of elders. Journal of Social Service Research, 34(1), 13-26. doi:10.1300/J079v34n01_02 Kite, M. E., & Wagner, L. S. (2002). Attitudes toward older adults. In T. D Nelson (Ed.), Ageism: Stereotyping and prejudice against older persons (pp. 129-161). Cambridge, MA: MIT Press. Klein, A., Tobin, T., Salomon, A., & Dubois, J. (2008). A statewide profile of abuse of older women and the criminal justice response, summary (Summary No. 222460). Sudbury, MA: Advocates of Human Potential. Retrieved from https://www.ncjrs.gov/pdffiles1/nij/grants/222460.pdf Kline, T., & Kline, D. (1991). The association between education, experience, and performance on two knowledge of aging and elderly questionnaires. Educational Gerontology, 17(4), 355-361. doi:10.1080/0360127910170405 Kropf, N. P. (2003). Strategies to increase student interest in aging. Journal of Gerontological Social Work, 39(1-2), 57-67. doi:10.1300/J083v39n01_06 Lachs, M. S., Williams, C. S., O'Brien, S., Hurst, L., & Horowitz, R. (1996). Older adults: An 11-year longitudinal study of adult protective service use. Archives of Internal Medicine. 156, 449-453. doi:10.1001/archinte.1996.00440040127014 167 Lachs, M. S., Williams, C. S., O'Brien, S., Pillemer, K. A., & Charlson, M. E. (1998). The mortality of elder mistreatment. The Journal of the American Medical Association, 280(5), 428-432. doi:10.1001/jama.280.5.428 Lee, S. E., Damron-Rodriguez, J., Lawrance, F. P., & Volland, P. J. (2009). Geriatric social work career tracking: Graduates of the Hartford Partnership Program for Aging Education (HPPAE). Journal of Gerontological Social Work, 52(4), 336- 353. doi:10.1080/01634370802609262 Leedahl, S. N., & Ferraro, F. R. (2007). Why is elder abuse overlooked? Media and ageism. Psychology and Education, 44(1), 1-9. Lieberman, T. (2013). The enduring myth of the greedy geezer. The Columbia Journalism Review. Retrieved from http://www.cjr.org/united_states_project/the_enduring_myth_of_the_greed.php?p age=all 51 Martens, A., Greenberg, J., Schimel, J., & Landau, M. (2004). Ageism and death: Effects of mortality salience and perceived similarity to elders on reactions to elderly people. Personality and Social Psychology Bulletin, 30(12), 1524-1536. doi:10.1177/0146167204271185 Miller, R B., & Dodder, R.A. (1980). A revision of Palmore's facts on aging quiz. The Gerontologist, 20(6), 673-679. doi:10.1093/geront/20.6.673 Morgan, E., Johnson, I., & Sigler, R. (2006). Public definitions and endorsement of the criminalization of elder abuse. Journal of Criminal Justice, 34(3), 275-283. doi:10.1016/j.jcrimjus.2006.03.004 Mullaly, R. (2002). Challenging oppression: A critical social work approach. Toronto, ON: Oxford University Press. Nakao, K. C., Damron-Rodriguez, J. A., Lawrance, F. P., & Volland, P. J. (2013). Examination of the psychometric properties of the knowledge of aging for social work quiz. Educational Gerontology, 39(10), 761-771. National Center on Elder Abuse. (2011a). Elder abuse/mistreatment defined. Retrieved from http://www.ncea.aoa.gov/NCEAroot/Main_Site/FAQ/Basics/Definition.aspx National Center on Elder Abuse. (2011b). Major types of elder abuse. Retrieved from http://www.ncea.aoa.gov/ncearoot/Main_Site/FAQ/Basics/Types_Of_Abuse.aspx National Committee for the Prevention of Elder Abuse (2005). 2005 White House conference on aging: Post-event summary report. Retrieved from http://www.preventelderabuse.org/whcoaging2005.html 168 Nelson, T. D. (2005). Ageism: Prejudice against our feared future self. Journal of Social Issues, 61(2), 207-221. Nelson T. D. (Ed). (2002). Ageism: Stereotyping and prejudice against older persons. Cambridge, MA: MIT Press. Nelson T. D. (2009). Ageism. In T. D Nelson (Ed.), Ageism: Handbook of prejudice, stereotyping, and discrimination (pp. 431-440). [e-book]. New York: Psychology Press: Retrieved from http://site.ebrary.com/lib/alltitles/docDetail.action?docID=10285315&p00=ageis m%3A%20handbook%20prejudice%2C%20stereotyping%2C%20discrimination Nerenberg, L. (2002). The national policy summit issue briefs. Journal of Elder Abuse and Neglect, 14(4), 71-104. doi:10.1300/J084v14n04_07 Nusbaum, N. J., Mistretta, M., & Wegner, J. (2007). An educational intervention for police and firefighters for elders at risk: Limits of education alone as a strategy for behavior change. Educational Gerontology, 33(10), 801-809. doi:10.1080/03601270701568798 O'Shaughnessy, C. V. (2010). The elder justice act: Addressing elder abuse, neglect, and exploitation. Retrieved from http://www.nhpf.org/library/details.cfm/2836 Older Americans Act: Amendments of 1987. (1987). Report from the committee on labor and human resources (to accompany S. 887), 100th congress, 1st session. Washington, DC: Congress of the U.S., Senate Committee on Labor and Human Resources. Retrieved from http://www.eric.ed.gov/ERICWebPortal/contentdelivery/servlet/ERICServlet?acc no=ED288108 Palmore, E.B. (1980). The facts on aging quiz: A review of findings. The Gerontologist, 20, 669-672. Palmore, E. B. (1998). The facts on aging quiz. New York, NY: Springer. Palmore, E. B. (1999). Ageism: Negative and positive. New York, NY: Springer. Parnas, R. I. (1973). Prosecutorial and judicial handling of family violence. Criminal Law Bulletin, 9, 733-769. Paveza, G. J., & VandeWeerd, C. (2006). Elder mistreatment and the role of social work. In Berkman, B., & Harootyan, L. (Eds.), Social work and health care in an aging society, (pp. 245-268). New York, NY: Springer. Payne, B. K. (2001). Understanding differences in opinion and 'facts' between ombudsmen, police chiefs, and nursing home directors. Journal of Elder Abuse 169 and Neglect, 13(3), 61-77. doi:10.1300/J084v13n03_04 Payne, B. K. (2002). An integrated understanding of elder abuse and neglect. Journal of Criminal Justice, 30(6), 535-547. doi:10.1016/S0047-2352(02)00175-7 Payne, B. K. (2003). Justification for punishing crimes against the elderly: Perceptions of police chiefs, nursing home professionals, and students. Journal of Offender Rehabilitation, 38(1), 33-51. doi:10.1300/J076v38n01_04 Payne, B. K. (2005). Crime and elder abuse: An integrated perspective. Springfield, IL: Charles C. Thomas. Payne, B. K., & Berg, B. L. (1999). A qualitative examination of the similarities and differences of elder abuse definitions among four groups: Nursing home directors, nursing home employees, police chiefs and students. Journal of Elder Abuse and Neglect, 10(3), 63-85. doi:10.1300/J084v10n03_04 Payne, B. K., & Berg, B. L. (2003). Perceptions about the criminalization of elder abuse among police chiefs and ombudsmen. Crime and Delinquency, 49(3), 439-459. doi:10.1177/0011128703049003005 Payne, B. K., Berg, B. L., & James, L. D. F. (2001). Attitudes about sanctioning elder abuse offenders among police chiefs, nursing home employees. International Journal of Offender Therapy and Comparative Criminology, 45(3), 363-382. doi:10.1177/0306624X01453008. Payne, B. K., Berg, B. L., & Toussaint, J. (2001). The police response to the criminalization of elder abuse: An exploratory study. Policing: An International Journal of Police Strategies and Management, 24(4), 605-626. doi:10.1108/EUM0000000006500 Penhale, B. (1993). The abuse of elderly people: Considerations for practice. British Journal of Social Work, 23(2), 95-112. Retrieved from http://bjsw.oxfordjournals.org.byui.idm.oclc.org/content/23/2/95.full.pdf+html Pillemer K., & Finkelhor D. (1989). Causes of elder abuse: Caregiver stress versus problem relatives. American Journal of Orthopsychiatry, 59, 179-187. doi:10.1111/j.1939-0025.1989.tb01649.x Plotkin, M. R. (1988). Time for dignity: Police and domestic abuse of the elderly. Washington, DC: Police Executive Research Forum (PERF). Retrieved from http://www.ncjrs.gov/App/Publications/abstract.aspx?ID=115805 Plotkin, M. R. (1996). Improving the police response to domestic elder abuse victims. Aging Magazine, 1(367), 28-30. Retrieved from http://search.proquest.com/docview/220734635?accountid=9817 170 Quinn, M. J., & Heisler, C. J. (2004). The legal response to elder abuse and neglect. Journal of Elder Abuse and Neglect, 14(1), 61-77. doi:10.1300/J084v14n01_04 Quinn, M. J., & Tomita, S.K. (1997). Elder abuse and neglect (2nd ed.). New York, NY: Springer. Ragan, A. M., & Bowen, A. M. (2001). Improving attitudes regarding the elderly population: The effects of information and reinforcement for change. The Gerontologist, 41(4), 511-515. doi:10.1093/geront/41.4.511. Ramsey-Klawsnik, H., Teaster, B. P., Mendiondo, S. M., Marcum, L. J., & Abner, L.E. (2008). Sexual predators who target elders: Findings from the first national study of sexual abuse in care facilities. Journal of Elder Abuse and Neglect, 20(4), 353- 376. doi:10.1080/08946560802359375 Ray, S., Sharp, E., & Abrams, D. (2006). Ageism: A benchmark of public attitudes in Britain. London, UK: Age Concern England. Reed, J., Cook, M., Cook, G., Inglis, P., & Clarke, C. (2006). Specialist services for older people: Issues of negative and positive ageism. Aging and Society, 26(6), 849- 865. doi./10.1017/S0144686X06004855 Reiss, A. J. (1984). Consequences of compliance and deterrence models of law enforcement for the exercise of police discretion. Law and Contemporary Problems, 47(4), 83-122. doi:10.2307/1191688 Roby, J. L., & Sullivan, R. (2001). Adult protection service laws: A comparison of state statutes from definition to case closure. Journal of Elder Abuse and Neglect, 12(3-4), 17-51. doi:10.1300/J084v12n03_02 Rohan E. A., Berkman B., Walker, S., & Holmes, W. (1995). The geriatric oncology patient: Ageism in social work practice. Journal of Gerontological Social Work, 23(1-2), 201-222. doi:10.1300/J083V23N01_1 Rosen, A. L., Zlotnik, J. L., & Singer, T. (2003). Basic gerontological competence for all social workers: The need to "gerontologize" social work education. Journal of Gerontological Social Work, 39(1-2), 25-36. doi:10.1300/J083v39n01_04 Rosenblatt, D. E., Cho, K.H., & Durance, P.W. (1996). Reporting of mistreatment of older adults: The role of physicians. Journal of the American Geriatrics Society, 44(1), 65-70. Schack, S., & Frank, R. S. (1978). Police service delivery to the elderly. Annals of the American Academy of Political and Social Science, 438(1), 81-95. doi:10.1177/000271627843800109 171 Scharlach, A., Damron-Rodriguez, J., Robinson, B., & Feldman, R. (2000). Educating social workers for an aging society: A vision for the 21st century. Journal of Social Work Education, 36(3), 521-538. doi:10.1080/10437797.2000.10779026 Schneider, D., Mosqueda, L., Falk, E., & Huba, G. (2010). Elder abuse forensic centers. Journal of Elder Abuse and Neglect, 22(3/4), 255-274. doi:10.1080/08946566.2010.490137 Schuller, N. (2006). Older people, crime and justice. Community Safety Journal, 5(4), 37 -43. doi:10.1108/17578043200600033 Sedlak, A.J., Mettenburg, J., Basena, M., Petta, I., McPherson, K., Greene, A., & Li, S. (2010). Fourth national incidence study of child abuse and neglect (nis-4): Report to Congress. Washington, DC: U.S. Department of Health and Human Services, Administration for children and Families. Retrieved from http://www.acf.hhs.gov/programs/opre/abuse_neglect/natl_incid/index.html Seefeldt, C., Jantz, R., Calper, A., & Seroc, K. (1977). Children's attitudes toward the elderly: Educational implications. Educational Gerontology, 2(3), 300-310. doi:10.1080/0360127770020308 Sever, B., & Youdin, R. (2006). Police knowledge of older populations: The impact of training, experience, and education. Professional Issues in Criminal Justice, 1(2), 35-54. Retrieved from http://www.picj.org/docs/issue2/PICJ%20Number2.pdf#page=46 Sherman, L. W. (1980). Causes of police behavior: The current state of quantitative research. Journal of Research in Crime and Delinquency, 17(1), 69-100. doi:10.1177/002242788001700106 Siegel, L. (2010). Criminology: The core. 4th edition. New York, NY: Wadsworth. Smith, M. R., & Alpert, G. P. (2007). Explaining police bias: A theory of social conditioning and illusory correlation. Criminal Justice and Behavior, 34, 1262- 1283. doi:10.1177/0093854807304484 Sorgman, M., & Sorensen, M. (1984). Ageism: A course of study. Theory into Practice, 23(2), 117-123. doi:10.1080/00405848409543101 Stein, K. F. (1991). A national agenda for elder abuse and neglect research: Issues and recommendations. Journal of Elder Abuse and Neglect, 3(3), 91-108. doi:10.1300/J084v03n03_05 Stein, K. F. (2006). What does the elder abuse field need? The Gerontologist, 46(5), 701- 704. doi:10.1093/geront/46.5.701 172 Steinmetz, S. K. (1988). Elder abuse by family caregivers: Processes and intervention strategies. Contemporary Family Therapy, 10(4), 256-271. Stiegel, L. (2000). The changing role of the courts in elder abuse cases. Generations, 24(2), 59-64. http://search.proquest.com/docview/212184244?accountid=9817 Stiegel, L. (2006). Recommendations for the elder abuse, health, and justice fields about medical forensic issues related to elder abuse and neglect. Journal of Elder Abuse and Neglect, 8(1), 41-81. doi:10.1300/J084v18n01_03 Stiegel, L., & Klem, E. (2008). Information about laws related to elder abuse. Retrieved from http://www.ncea.aoa.gov/NCEAroot/Main_Site/Library/Laws/InfoAboutLaws_08 _08.aspx Sykes, R. (1977). The urban police function in regard to the elderly: A special case of police community relations. In J. Goldsmith & S. S. Goldsmith (Eds.), Crime and the Elderly (pp. 128-131). Lexington, MA: Lexington Books. Street, D., & Cossman, J. S. (2006). Greatest generation or greedy geezers? Social spending preferences and the elderly. Social Problems, 53(1), 75-96. doi:10.1525/sp.2006.53.1.75 Tatara, T. (1990). Summaries of national elder abuse data: An exploratory study of state statistics based on a survey of state adult protective service and aging agencies. Washington, DC: National Aging Resource Center on Elder Abuse. Tatara, T. (1993). Summaries of the statistical data on elder abuse in domestic settings for FY90 and FY91 a final report. Washington, DC: National Aging Resource Center on Elder Abuse. Tatara, T., & Kuzmeskus, L. M. (1997). Elder abuse: Trends in elder abuse in domestic settings. Elder Abuse Information Series, 2. Retrieved from http://www.ncea.aoa.gov/ncearoot/Main_Site/pdf/basics/fact2.pdf Tatara, T., Kuzmeskus, L. B., Duckhorn, E., & Bivens, L. (1998). The National Elder Abuse Incidence Study (NEAIS): Final report. Washington DC: National Center on Elder Abuse. Teaster, P. B., Dugar, T., Mendiondo, M., Abner, E., Cecil, K., & Otto, J. (2006). Abuse of adults 60 years of age and older: The 2004 survey of adult protective services. Washington, DC: National Center on Elder Abuse (NCEA). Retrieved from http://www.ncea.aoa.gov/NCEAroot/Main_Site/pdf/021406_60PLUS_REPORT.p df Teaster, P.B., Roberto, K.A., Duke, J.O., & Kim, M. (2000). Sexual abuse of older 173 adults: Preliminary findings of cases in Virginia. Journal of Elder Abuse and Neglect, 12(3/4), 1-16. doi:10.1300/J084v12n03_01 Treas, J. (1995). Older Americans in the 1990s and beyond. Population Bulletin, 50(2), 1-46. Washington, DC: Population Reference Bureau. Turner, B. S. (1998). Aging and generational conflicts: A reply to Sarah Irwin. The British Journal of Sociology, 49(2), 299-304. Retrieved from http://www.jstor.org/stable/591314 U. S. Census Bureau. (2012). 2012 National Population Projections. Retrieved from http://www.census.gov/population/projections/data/national/2012.html United States. Congress. Senate. Special Committee on Aging. (1995). Society's secret shame: Elder abuse and family violence: Hearing before the Special Committee on Aging, United States Senate, One Hundred Fourth Congress, first session. Washington DC: Government Printing Office. Utah State Division of Aging and Adult Services. (2012). 2011 annual report. Retrieved from http://www.hsdaas.utah.gov/pdf/2011-annual-report.pdf Vincent, G. K., & Velkoff, V. A. (2010). Next four decades: The older population in the United States: 2010 to 2050. Population estimates and projections. Washington, D.C.: Bureau of the Census. Retrieved from http://www.census.gov/prod/2010pubs/p25-1138.pdf Walsh C. A., Olson, J. L., Ploeg, J., Lohfeld, L., & MacMillan, H. L. (2010): Elder abuse and oppression: Voices of marginalized. Elders, Journal of Elder Abuse and Neglect, 23(1), 17-42. doi:10.1080/08946566.2011.534705 Wang, D. (2013). Social workers' attitudes toward older adults: A review of the literature. Journal of Social Work Education, 49(1), 149-172. doi:10.1080/10437797.2013.755104 Wasik, J.F. (2000). The fleecing of America's elderly. Consumers Digest, (March/April), 77-83. Wigboldus, D. H. J., Dijksterhuis, A., & Van Knippenberg, A. (2003). When stereotypes get in the way: Stereotypes obstruct stereotype-inconsistent trait inferences. Journal of Personality and Social Psychology, 84(3), 470-484. doi:10.1037/00223514.84.3.470 Wilkinson, J. A., & Ferraro, K. F. (2002). Thirty years of ageism research. In T. Nelson (Ed.), Ageism: Stereotyping and prejudice against older persons (pp. 339-358). Cambridge, MA: MIT Press. 174 Willard, F. R. (2011). The elder justice act: A big step towards protecting older Americans. Proceedings (Baylor University Medical Center), 24(1), 48. Retrieved from http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3012290/ Wolf, R. S. (1996). Elder abuse and family violence: Testimony presented before the U.S. senate special committee on aging. Journal of Elder Abuse and Neglect, 8(1), 81-96. doi:10.1300/J084v08n01_05 Wolf, R. S. (2003). Elder abuse and neglect: History and concepts. In R.J. Bonnie & R.B Wallace, (Eds.), Elder mistreatment: Abuse, neglect, and exploitation in an aging America (pp. 238-248). Washington, DC: National Research Council. Wolfe, D. A. (2003). Elder abuse intervention: Lessons from child abuse and domestic violence initiatives. In R. J. Bonnie & R. B. Wallace (Eds.), Elder mistreatment: Abuse, neglect, and exploitation in an aging America (pp. 501-526). Washington, DC: National Research Council. Wright, L. (1996, January 17). Bill focuses on elder abuse legislation aims to curb elder abuse. The Salt Lake Tribune, pp. A.1.