The influence of desire for control on internet purchasing selection

Since individuals began utilizing the Internet to purchase travel, travel agencies have suffered financially and in many cases went out of business. It has been suggested that Internet technology has contributed to the demise of these agencies. Travel agencies may have been unprepared for technological growth or failed to complete the necessary research; regardless, it is evident that travel agencies are confused about what Internet travel purchasers expect when a travel purchase is made. Utilizing desire for control theory, individual control characteristics were measured and examined in Internet travel purchasers. The instrument for the study employed several combinations of measurement to assist in predicting a person's probable Internet purchasing method and its relationship to desire for control: (a) 17 questions regarding demographics and purchasing behaviors, and (b) 20 questions regarding desire for control. Questionnaire invitations were distributed to 1,034 current email-address holders from a local Internet travel company/agency. Four hundred individuals responded to the questionnaire, resulting in a 38.6% response rate. The demographic portion of the questionnaire revealed that a majority of participants were female (57.1%), had completed a college education (55.6%), and earned $30,000 or more annually (83.7%). When making a travel-purchase selection over the Internet, 77.8% of the participants felt that price was extremely important to very important, participants indicated that the amount of control over the actual purchasing decision was extremely important to very important (78.8%). Logistic regression was utilized to determine a predictive set of variables to predict the respondents' preferred method of purchasing travel over the Internet, but the predictive set of variables never developed from the identified list. The study failed to support the research hypothesis and indicated that desire for control was not a significant variable in determining Internet purchasing decisions. Therefore, travel agency marketers need not focus on one's desire for control in developing websites for travel purchasing. Further research may shed more light on Internet travel-purchase decision making and assist travel agencies in their quest to further develop needed Internet services for their clients and hopefully maintain their place within the travel marketplace.

THE INFLUENCE OF DESIRE FOR CONTROL ON INTERNET PURCHASING SELECTION by Frederick J. Inter A dissertation submitted to the faculty of The University of Utah in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Parks, Recreation, and Tourism The University of Utah August 2013 Copyright © Frederick J. Inter 2013 All Rights Reserved T he U ni versit y of Ut ah Grad uat e S chool STATEMENT OF DISSERTATION APPROVAL The dissertation of Frederick J. Inter has been approved by the following supervisory committee members: , Chair Linda S. Ralston 3-12-2013 Date Approved , Member Gary D. Ellis 3-12-2013 Date Approved , Member Edward J. Ruddell 3-12-2013 Date Approved , Member Michael K. Gardner 3-12-2013 Date Approved , Member Alison E. Regan 4-3-2013 Date Approved and by the Department of Daniel L. Dustin Parks, Recreation, and Tourism and by Donna M. White, Interim Dean of The Graduate School. , Chair of ABSTRACT Since individuals began utilizing the Internet to purchase travel, travel agencies have suffered financially and in many cases went out of business. It has been suggested that Internet technology has contributed to the demise of these agencies. Travel agencies may have been unprepared for technological growth or failed to complete the necessary research; regardless, it is evident that travel agencies are confused about what Internet travel purchasers expect when a travel purchase is made. Utilizing desire for control theory, individual control characteristics were measured and examined in Internet travel purchasers. The instrument for the study employed several combinations of measurement to assist in predicting a person's probable Internet purchasing method and its relationship to desire for control: (a) 17 questions regarding demographics and purchasing behaviors, and (b) 20 questions regarding desire for control. Questionnaire invitations were distributed to 1,034 current email-address holders from a local Internet travel company/agency. Four hundred individuals responded to the questionnaire, resulting in a 38.6% response rate. The demographic portion of the questionnaire revealed that a majority of participants were female (57.1%), had completed a college education (55.6%), and earned $30,000 or more annually (83.7%). When making a travel-purchase selection over the Internet, 77.8% of the participants felt that price was extremely important to very important, participants indicated that the amount of control over the actual purchasing decision was extremely important to very important (78.8%). Logistic regression was utilized to determine a predictive set of variables to predict the respondents' preferred method of purchasing travel over the Internet, but the predictive set of variables never developed from the identified list. The study failed to support the research hypothesis and indicated that desire for control was not a significant variable in determining Internet purchasing decisions. Therefore, travel agency marketers need not focus on one's desire for control in developing websites for travel purchasing. Further research may shed more light on Internet travelpurchase decision making and assist travel agencies in their quest to further develop needed Internet services for their clients and hopefully maintain their place within the travel marketplace. iv The completion of this dissertation is dedicated to my wife, whom I love very much; without her this would have never been completed. TABLE OF CONTENTS ABSTRACT…………………………………………………….………………………. iii LIST OF TABLES……………………………………..……………………………….. viii ACKNOWLEDGEMENTS…………………………………………………..………… ix CHAPTER 1 INTRODUCTION……………………..……………………………………………… 1 2 LITERATURE REVIEW……………………………….…………………………….. 4 Online Travel Purchasing…………………………………….......................... Defining Travel Provider Groups…………………...………………………… Control………………………………………………………........................... Human Personality……………………………………………………………. The Purchasing Process………………………..……………........................... Conclusion……………………………………………………………………. Hypothesis………………………….………………………………………… 5 19 24 26 46 57 59 3 METHOD……………………………………………………………........................... 63 Study Setting…………………………………………..……........................... 63 4 RESULTS……………………..………………………………………………………. 68 Population Profile…………………………………………….......................... Population Computer Purchasing Procedures……..…………………………. Hypothesis Testing…………………………………………………………… 68 69 70 5 DISCUSSION………………..………………………………………………………... 78 Purpose of Study…………………..………………………………………….. 78 Internet Travel Purchaser Profile…………………..…………………………. Hypothesis Testing…………………………………………………………… Future Research……………………………………………………………….. Conclusion…………………………………………………………………….. 78 80 81 81 APPENDIX……………………………………………………………………………… 83 REFERENCES…………………………………………………………………………... 89 vii LIST OF TABLES Table 1. Page Annual Percentage of Southwest Airlines Ticketless Travel and Online Purchases…………………………………………………………………….... 60 2. Annual Percentage of Continental Airlines E-Ticket Purchases……………… 60 3. Internet Travel Purchasers Income (N = 398)……………………………..…. 72 4. Importance Regarding the Length of Time Searching the Internet for Travel (N = 93)…………………………………………………………...…………. 73 5. Hours Spent Surfing the Internet……………………………………………... 73 6. Important Travel-Purchasing Factors………………………….……………… 74 7. Computer Location Access (N = 400)…………………………….………….. 74 8. Vacation Length of Stay (N = 400)……………………………..…………….. 75 9. Desire for Control Predictive Set…………………………….……………….. 76 10. Importance of Control Predictive Set……………………….………………… 77 ACKNOWLEDGEMENTS I wish to acknowledge the efforts of individuals and organizations that contributed to the completion of this dissertation. First, a special thanks to my wife and mother, for their continual determination to push me through this process. Additional appreciation is expressed to my father, who protected me from the two determined women in my life. Further appreciation is given to Dr. Terry Long and committee members Dr. Gary Ellis, Dr. Edward Ruddell, Dr. Allison Reagan, and Dr. Michael Gardner, for their participation in the project. An extra special thank-you goes to Dr. Linda Ralston, who saw me through the dissertation process, and Sabrina Rood who spent countless hours reviewing this document. Lastly, I would like to send a special thanks to the Morris Murdock Travel Marketing Team and the technological support they provided for this study. CHAPTER 1 INTRODUCTION Lack of understanding Internet travel purchases is a serious problem for travel agencies who market on the World Wide Web, commonly referred to as the Internet. Many of these agencies once booked the majority of travel purchased in the United States. Yet, since the late 1990s, revenues have declined for these more traditional suppliers (Shattuck, 2003) and now stands at around 20% of what it was once (Amarante, 2002). One reason for this decline is competition resulting from the availability of online travel suppliers. Internet travel suppliers (as opposed to Internet travel agencies) are a conglomerate of travel-supplier sites, search-engine sites, and destination sites. Their key advantage is the ability to provide travel-purchase services in a quicker and less costly manner. Internet travel purchasing has increased more than 300%, having grown from $3.1 billion in 1998 (Forrester Research, 1999) to $14.8 billion in 2001 (Schaal, 2002). The bulk of this increase has been associated with Internet suppliers. With the Internet as a new mode of doing business and with Internet suppliers occupying most of this market, the most affected businesses are the smaller travel agencies. These are defined as those agencies earning $1 million or less in revenues annually (Lazar, 1999). Since 1994, one in every three travel agencies that previously existed has disappeared (Winstead, 2002). If these smaller agencies wish to remain viable, they will be faced with the challenge of discovering market and economic solutions that match their agencies' size and competitive position (Lassiter, 1998; Luhrman, 2001). Loss of such travel 2 agencies would reduce or eliminate travel-purchase services that cater to those individuals who are unable to use or choose not to use computers and those seeking more "human contact" in their travel-purchase selection. Based on the assumption that lack of understanding Internet travel purchases is a serious problem for travel agencies who market on the Internet, increased understanding of Internet purchase behavior might help such agencies remain viable. Desire for control may be an important variable in such understanding. People desire varying levels of control over the events in their lives (Burger, 1984, 1985, 1986, 1987, 1989, 1992). Control over one's travel purchases may be among the set of such events. Increased control provided by travel-supplier websites coupled with the "new style" consumer's tendencies (lack of time, attention span, and trustworthiness) has inspired many travel consumers to substitute Internet travel suppliers for previously used travel agents. Utilization of the Internet for travel-purchase selection has increased consumer control by affording choice among multiple travel websites that enable consumers to select from hundreds to thousands of different travel options. While some travel consumers may appreciate such control, others may not. These latter individuals may prefer that the travel agent do this for them. Thus, a great understanding of desire for control may provide valuable insight into consumer travel purchases and the recent shift in online travel markets. Desire for control may be irrelevant to how people make online travel-purchase selection because control is not important to them. Travel purchasers might prefer purchasing online travel because it is convenient or because they are not limited by options and have several choices. Another possibility is that people enjoy receiving instant notification that their travel purchase has been made, rather than waiting for a travel agent to return with a few itineraries later that day or the next, thus changing the focus away from control to 3 available time. Control requires, or implies, taking the time to find the travel selection that best suits the individual purchasers' needs, but as the consumer changes, the amount of available control is becoming more of an issue. Lastly, the entire online travel-purchasing process could be entirely driven by monetary value. Therefore, the purpose of this study was to explain the association between travel-purchase selection and desire for control in a population of Internet travel purchasers. CHAPTER 2 LITERATURE REVIEW Over the past decade the travel industry has experienced significant fluctuations, with the introduction of the Internet causing a positive fluctuation (Modahl, 2000; Peters, 1996) and the September 11, 2001 event causing a negative fluctuation (Schaal, 2002). For 2002, the travel industry is recovering, reporting a 71% increase over 2001 travel revenues (Schaal, 2002), but the travel industry is still some $3 billion below revenues from 2000 ("On-line Travel Trends," 2001). However, the one area of travel purchasing that seems to be on the rise is the sector of online purchasing (Schaal, 2002). Online travel purchasing has grown from $3.1 billion in 1998 (Forrester Research, 1999) to $14.8 billion in 2001 (Schaal, 2002), and it is expected to exceed $60 billion by 2006 (Schaal, 2002). Thus, the following review of literature will examine online purchasing, online travel selection, consumer desire for control, and the relationship between control and online travel selection. The population utilized in this study consisted of Internet users who were interested in purchasing travel using the World Wide Web. With travel purchases increasing through the Internet (Durbin, 2002), it is essential that researcher and professionals gain a better understanding of the services that individuals both seek and expect. 5 Online Travel Purchasing Technology Enters the Hospitality Industry Apparently two large London hotels were the earliest to successfully utilize computers in the hospitality industry. These hotels operated what they called the Leo reservation system, the DEC PDP8, and served the Palace Hotel (450 rooms) and the Cumberland (800 rooms) (Cooper, 1993). Around the same time that the Leo was introduced, International Business Machines Corporation (IBM) and Hilton opened the world's first fully computerized hotel, the New York Hilton (Cooper, 1993). Unfortunately, the hotel experienced several major complications and closed down the computer system several days after opening for operation. Failure of this system caused many United States organizations to lose interest in computer technologies and drastically slowed the computer's progression for the American hotel industry. Even though the Leo and its DEC PDP8 reservation system experienced some success, very few companies were willing to take the necessary steps to restructure their business and become totally computer automated (Anderson, 2000). It was not until the late 1970s and the further development of the microcomputer that computer-based technology began to take off and become beneficial to the hospitality industry (Gamble, 1993). The world's first Property Management System (PMS) was installed in 1978 by Hoskyns Services in the Seven Hills Hotel in Cobham, England (Chervenak, Keane, & Co., 1988). The system, operating with a few bugs, was generally accepted by most hotel guests. However, fears still remained and many hotel companies continually refused to make technological accommodation to their businesses in the United States until after 1980 (Chervenak, Keane, & Co., 1988). One can observe that during mid-1983 there were some 300 hotels using this PMS system. By 1987, there were more than 30,000 PMS systems in 6 use throughout the United States (Chervenak, Keane, & Co., 1988). Travel Technologies Travel companies did not start utilizing computerized reservation systems until a few years after World War II (Chervenak, Keane, & Co., 1988). Prior to the development of computer reservation systems, airline companies utilized several manual techniques to keep track of seating schedules. During the 1930s, air carriers handled all airline reservation transactions through central reservation locations manned by agents, who provided flight information and reserved seats on flights (Anderson, 2000). This cumbersome process involved agents who answered phone calls regarding flight information. The agents responded to airline inquiries and acknowledged flight-availability information to other agents, who in turn informed potential travel purchasers. Many of these reservation centers used large wall-sized boards to keep track of information in order to maintain an updated reservation list that allowed them to prevent overbooking. Reservation agents would manually maintain the board, close the board when a flight sold out, and send telexes to all other reservation offices to terminate the selling of airline tickets (Sheldon, 1997). The system experienced overbooking and reservation complications because there were usually several central reservation centers booking the same flights. This method, commonly referred to as the Lazy Susan system, was one of the first steps toward automation and, at the time, was considered a major innovation (Anderson, 2000). The reservation system was continuously examined in an attempt to improve it, but despite numerous attempts to control seating information, the reservation process remained a tedious task. In the 1940s, new forms of communications deployed as a result of World War II helped to streamline airline operations (Anderson, 2000). Telex machines, for example, were used to record flights, messages, and whereabouts of the airplanes in war. When the war 7 ended, these telex machines became a useful tool for communications between reservation offices. However, availability and scheduling information were still stored on cards in the Lazy Susan. It was not until after 1952 that American Airlines began developing the first computer system that could handle airline reservations, ticketing, scheduling, seat inventories, and passenger-name records (Anderson, 2000; Sheldon, 1997). This computer marvel was first referred to as the Magnetronic Reservoir and became the model for future reservation systems, including the Semi-Automatic Business Research Environment (SABRE) system. However, until the mid-1970s, airline travel, in most cases, had to be purchased directly from the air carriers. As a result, airlines were sometimes unable to continually fill their flights, yet reservation centers were operating nonstop at virtually ultimate capacity. Realizing that they were not meeting the demands of the travel market, many airline companies began to see the relevancy of a person outside their organization who would sell reservations and relay important travel information on to the central reservation locations. Travel agents quickly began making airline reservations from outside airline organizations and found success. Nevertheless, airline companies shortly thereafter discovered complications in their reservation system that limited the efficiency of their automated teleticketing machines. Few travel agencies could distribute tickets through these machines, limiting their effectiveness (Godwin, 1987). Because the reservation systems had various limitations, an attempt was made to create a central reservation system that would give travel agents access to airline information in 1975. The effort was referred to as the Joint Industry Computer Reservation System, which brought together major carriers to plan an effective travel system specifically for travel agent usage (Copeland & McKenney, 1988). The plan failed. The effort, however, pushed 8 companies into developing other user-friendly reservation systems. One such system was developed by IBM and United Airlines. This system took its basic ideas from the Joint Industry Computer Reservation System, which was supposed to engineer an industry-wide travel-agent-friendly system. The result was the Random Access Memory Accounting Machine (RAMAC), which could store and memorize 5 million units of information for periods of 100 days (Anderson, 2000). Each system provided travel agents with the opportunity to rent their equipment and provide direct access to their computers. Upon observing the success of both automated reservation systems, several others attempted to design their own systems. By the mid-1980s, travel agents wanting to automate could select from five different systems. In 1964, the first SABRE reservation system was developed by American Airlines and IBM (Anderson, 2000). The SABRE prototype had to overcome several major barriers before becoming operational. These challenges were similar to those the previous reservation systems had faced. First, the system had to be able to provide information in real time, because travel agents required immediate responses to questions pertaining to seat availability and fares. Second, the system had to have a computer data communications network that was able to connect all travel-agent reservation locations to a central site. This was one of the largest problems to overcome, as few computer networks existed that could handle such a task. Third, the system had to be accessible 24 hours a day because of the different time zones and purchasing behaviors. Fourth, the system had to be able to receive a multitude of information and decipher it quickly because the system would be receiving thousands of reservation bookings and fare inquiries daily. The network had to be a large online computer system able to process large volumes of information that were connected to other similar computer systems, in order to provide immediate and accurate responses 9 regarding travel inquiries. The SABRE system took over a decade to perfect, thus in 1976 becoming the first fully automated reservation system to be installed in a travel agency (Anderson, 2000). With the rise of such dynamic reservation systems, two types of airlines emerged: those affiliated with a reservation system used by travel agents for multiple carriers such as Global Distribution Systems (GDS) and those not affiliated (Computer Reservation Systems, CRS) (Vellas & Becherel, 1999). Controversy swirled in the United States pertaining to the competitiveness and profitability from being affiliated with GDS. The fact that some airlines could have reservations made by any number of travel agents, while others were limited to their own reservation systems, caused a great deal of friction in the airline industry. Companies who utilized CRS felt like they were at a disadvantage because they were associated with nonvendor airlines and were without direct communication with travel agents. After a great deal of legislation, rules were implemented by the airline commission to fairly and equally represent all airlines on the GDS screens and reduce any perceived unfairness (Sheldon, 1997). GDS Terminal Technologies The first and most prevalent technology available to the travel agent was the Webbased platform Global Distribution Systems (GDS) terminal system (Sheldon, 1997). GDS differs from simple information systems because it is far more comprehensive and provides important travel and reservation information, allowing reservations to be registered and sales transactions to be exchanged and processed over the Internet directly through the system (Vellas & Becherel, 1999). Today, GDS terminals are the major information and booking tools used by travel agents for all types of travel. The computer hardware that is most appropriate for a particular travel agency 10 depends on the size of the organization and the desired versatility the agency requires (Anderson, 2000). Every hardware system begins with at least one GDS terminal. At first these terminals were leased from some terminal vendors, but after legislation by the Department of Transportation in 1992, travel agents were able to purchase the terminals from outside sources rather than having a lease system. In order to operate each system, a gateway is required if there is more than one terminal in an office (Harris, 1996). Additionally, a modem and printer are required as well as local connecting network software. The server is then connected through a gateway, modem, and phone line to a central mainframe. Because the agencies are no longer bound to leasing the terminals from GDS vendors, the hardware configurations have become more flexible (Harris, 1996). In addition to the travel agency's GDS technologies, travel agents have had to familiarize themselves with the Internet, viewed as both a threat and a business opportunity (Sheldon, 1997; Vellas & Becherel, 1999). The reason for the fear was that the Internet opened up the travel market for all consumers and could remove much of the travel agents' business (Anderson, 2000); others saw the potential of using the World Wide Web. These agents used the Web to gather research to protect themselves, realizing that in the future some travel products may only be available through the Internet. Furthermore, it is understood that many people who purchase travel have very little free Internet user time, thereby creating a market for specialists to develop special travel itineraries and reservations. It is expected that Internet inquiry and special booking services are likely to add value to the customer's experience in the form of more convenience, better information, and cost savings when purchasing from a travel agent. Currently, airlines have had the luxury of bypassing the traditional travel agent and 11 have opened new marketing and purchasing opportunities through the World Wide Web (Schaal, 2001). The Internet has enabled these airlines to send information out into the world in the way of advertisements and company Web pages that are unique and profitable. Airlines and other travel suppliers have not only found that the Internet has enabled them to reach a greater number of persons through advertisements, but that people are more inclined to purchase travel specific to their own needs; thus, airline companies have generated large revenues from not only travel agents, but individual households as well. Introduction of the Internet The Internet was developed during the 1960s to provide the Department of Defense's computers with the opportunity to communicate with other global computers in the event of a nuclear war, assuming the normal communications network with a central hub would be destroyed if such a situation were to occur (Peters, 1996). In 1969, four computers across the United States could successfully exchange information over the Internet-type network which became known as the Advanced Projects Research Agency. In 1985, this network was made available to the public; more than 1,000 computers were connected to the network that year (Anderson, 2000). With computers becoming more accessible to family households, this number began to increase dramatically. By 1995, the Internet had attracted an excess of 1 million computers (Bonn, Furr, & Susskind, 1999). The most familiar and widely used Internet resource is the World Wide Web, commonly referred to as the Web (Sheldon, 1997). Developed in 1989, the basic operation of the Web involved the composition of documents comprised of pages specially formulated into a formatted text called Hyper Text Markup Language (HTML), which allows individuals and organizations to display information on a computer screen visible through the means of viewing software, called browsers. These Web browsers enable people to send and receive e- 12 mail, participate in discussion groups, purchase goods and services, and conduct most Internet functions 24 hours a day. Technology and Travel Consumers Technology has had a great effect on travel companies, but it has also affected the travel consumer (Middleton, 2001). Although the means to engage in travel-related activities have been available to the public for at least 13 years, it was not until around 1995 that people began to truly become interested in using the Internet as a means to purchase travel products and services. Prior to 1995, travel consumers could only purchase travel one of two ways: either through a telephone conversation with a travel company or supplier or through a face-to-face transaction with an agent who had the proper equipment to complete the purchase. Regardless of how the travel purchase occurred, the consumer often had limited options regarding the selection of travel packages and itinerary (Anderson, 2000; Jones, 2000). Purchasers had to select from travel packages that the travel agencies had available. If one wanted to develop a specific travel itinerary, then it would realistically take a day or two for the travel agent to develop the requested travel plans. The consumer would then pay large commission fees to the individual or agency for the time the travel agent spent developing the travel itinerary. However, as technology developed so did the techniques for purchasing travel. Travel consumers first saw purchasing behavior change slightly when the government put restraints on airline reservation systems (Jones, 2000). Travelers were no longer as concerned as they once were when using travel agents to make reservations. All agents were able to make reservations on the same system, allowing the consumers some peace of mind that they had made a good travel selection, not to mention that the travel agent was able to 13 provide the travel consumer with more travel options and minimize the time it took to develop specialized itineraries. Though it was nice to know that one could reserve a flight on a particular airline or at a particular hotel, a shopper had to be wary when agreeing on a price for a travel experience. A handful of travel agents attempted to take advantage of the system and tended to overlook many of the low-cost effective travel items available to the public in order to identify two or three of the higher-priced travel arrangements (Anderson, 2000). They were able to do this because travel suppliers offered them incentives to book certain travel packages. Basically, travel agents were able to receive more money for booking certain travel itineraries while ensuring that the airlines and hotels who provided the best incentive packages had full seats and available room, thereby demonstrating that it was more advantageous for travel agents to sell travel packages than it was for the consumer to purchase one. These types of travel dealings were the norm for many years until people became familiar with the Internet (Reynolds & Olson, 2001). With this newfound knowledge, people saw and experienced the benefits of using the Internet to search for and purchase travel. By having the ability to research travel sites on the internet, consumers were able to obtain important information regarding travel options such as cost, schedule, and services provided. Travel consumers began to do more in depth research, collecting as much information as they felt necessary in order to make a travel decision (Lewis & Bridger, 2000). For the first time they were able to book the reservation themselves or call the travel agent to book the travel arrangements. What consumers discovered created an awareness that had never before been available. People discovered how much the travel itinerary actually cost and measured the actual costs against the fee the travel agencies were charging, resulting in many 14 travel purchasers wanting to take the time to save a little money to purchase travel on their own. To some people, paying the extra fee seemed outrageous, while others just paid the extra fees because that was what they were accustomed to or that is how their business wanted them to purchase travel. However, as the Internet continued to advance, people became increasingly familiar with the technology and have made more and more reservations on their own, so much so that the traditional role of a travel agent has changed and, in many circumstances, has been eliminated completely (Modahl, 2000). The Internet and Travel During the last decade, the Internet has changed the world as we know it. Technology has advanced at a remarkable rate, thus making it difficult to track the impact of these advancements. One area of particular concern is the impact the Internet has had on the traditional travel agent (Durbin, 2002). The increasing availability of tourism information on the Internet, coupled with direct bookings and the development of convenient payment facilities, could possibly explain why the Internet has been detrimental to traditional travel agencies (Bennett, 1993; Stride, 1996; Yesawich, 1996). Some have even suggested that the end of the travel agent is near (Durbin, 2002; Oppermann, 1999). Therefore, satisfying or meeting the needs and expectations of a customer has never been quite as important for travel agencies as it is today (Oppermann, 1999). Consumers, empowered by the readily available information provided over the Internet, are becoming more "travel smart" (Bruner, 1998; Lodziak, 2002). As a result, traditional patrons of travel agencies have ventured out on their own in an attempt to find cheaper, faster, simpler, and more accessible means of purchasing travel, mainly through the Internet. Consequently, by using the Internet to search for and purchase travel, consumers 15 have found more information about potential travel products and services, better travel rates, greater selections, and a multitude of traveling times to fit the most demanding schedules, without having to pay the extra costs associated with using a travel agency. Because of this, travel agencies have been inclined to adjust their traditional travel-purchasing and marketing methods to address the changing necessities of their patrons. The travel agents' transformation has been caused by the development of independent Internet travel-supply companies (Schaal, 2002) like Expedia.com and Travelocity, as well as other primary travel suppliers such as Southwest Airlines and Continental Airlines (Dorsey, 1999). The above companies have developed new, less time-consuming, and less expensive means of purchasing travel that allow instant reservation purchasing. These companies' innovations have forced travel agents and agencies to reform their methods of marketing and travel purchasing to focus on the development of specific travel packages that include all possible travel arrangements (airline tickets, hotel reservations, automobile rentals, etc.). These techniques of purchasing travel over the Internet took years to develop, and travel agents were not the only ones looking in from the outside (Anderson, 2000). So it comes as no surprise that several other sectors of the travel industry have resisted the temptation to use modern technology, but as time progresses more and more of these companies have made the move to improve their industries through technological advancements. Growth in online travel sales has developed so rapidly compared to other products that it could represent nearly half of all e-commerce within the next 2 to 3 years (Luhrman, 2001). Many companies have developed businesses utilizing this technology, but few have been prepared for the unexpected growth that has accompanied it, thereby making it difficult to measure. For example, many airline agencies have been utilizing the Internet for a number 16 of years, and reports from sources indicate that millions of people access airline Web sites annually and billions of dollars are spent on travel purchases (Jones, 2001). However, after reviewing several of the top airlines' annual reports, stockholders were not informed of any online purchase totals until 2000. Though all of the reviewed airline companies reported the advancement of Internet technologies prior to 2000 and even discussed how many hits their Web sites are receiving on an average day, these companies revealed little information pertaining to their online travel-purchasing history during the initial stages of their Web development. With such rapid growth in Internet technologies, and recent reports concerning the enormity of reported purchases during the past couple of years, airline companies and a few secondary travel providers were able to develop some foresight concerning the prospect of the Internet's future growth. Southwest Airlines has been conducting business over the Internet since 1996 (Southwest Airlines, 1995) but made little indication about how the Internet had affected their business until their 2001 annual report. In 2000, however, they reported that over 31% of their entire 2000 revenues, $1.67 billion, was the result of tickets purchased over the Internet (Southwest Airlines, 2001). Southwest is one example of an Internet success story, but Southwest is very different from other travel companies. Southwest continually tried to develop technology to enhance their service and had been working with technological innovation for some time before entering the Internet age. They used this technology and introduced the first ticketless travel program in 1995 (Southwest Airlines, 1994). The ticketless travel program was the ideal service to use over the Internet. The service enabled Southwest to modify their traditional marketing approach and begin to market and advertise on the Internet, but the airlines were not aware of the impact it would have on the company over the next 7 years. 17 Since beginning the program in 1995, ticketless travel has accounted for a large portion of Southwest's overall ticket purchases. In 1995, it accounted for 40% of all tickets purchased (Southwest Airlines, 1995). In 1996, it accounted for 50% of all tickets purchased (Southwest Airlines, 1996), and in 1997 it accounted for 60% of all tickets sold (Southwest Airlines, 1997). Southwest's ticketless travel program has grown every year since it began and currently represents over 80% of all tickets purchased through the airline (see Table 1). This is important because all the airlines have attempted to develop Internet business and some have had more success than others. From examining other airlines, it seems that those who have strong ticketless travel programs have had more success with Internet business than those who did not have to initially develop a strong ticketless travel program. Another airline company that experienced success in Internet travel purchasing was Continental Airlines, who developed an e-ticket program during 1997 (Continental Airlines, 1998). Continental experienced continual annual growth in Internet sales every year since its development and realized its success in 1999 when e-ticket purchases increased 173% from the previous year, accounting for 47% of all the company's ticket purchases (Continental Airlines, 1999). Furthermore, in 2000 Continental sold 54.1% of their tickets as e-tickets and sold 59.5% in 2001 (Continental Airlines, 2001), thereby demonstrating its importance to the company (see Table 2). In contrast, Delta Air Lines, who did not develop as strong a ticketless travel program and who did not achieve the level of success that Southwest and Continental had, admitted that early on much of their equipment needed to be updated when the Internet race began (Delta Air Lines, 1997). However, in their most recent annual report, they revealed to stockholders that their Internet business was on the rise, and with new equipment and technological advancements within their system, Internet purchases represented 9% of all 18 tickets purchased during 2001 (Delta Air Lines, 2001). Other companies also developed successful travel business over the Internet and further threatened traditional travel agencies. In 1996, both Travelocity and Expedia were introduced. As years progressed, people became more comfortable booking online. By 1998, $2.6 billion in travel was sold online, and by 1999 that figure escalated to more than $7 billion (Sanfilippo, 2000). However, information from a variety of sources indicates that consumers benefit from using a travel agent, especially for business travel and on high ticket items, but consumers are discovering new travel intermediaries through which to purchase travel (Del Rosso, 2001; Featherly, 2000; "On-line Travel Trends," 2001; Schaal, 2001). This train of thought may explain why many travel agents can only be found on the World Wide Web as specialists who provide full travel services. Studies reveal that consumers who use the Internet to research trips still use travel agents to check the relevancy of the information and book travel. Often, purchasers want an outside party to take responsibility or are fearful of purchasing travel through the Internet on a credit card. In the future, travel intermediaries will have to continuously redefine their operations and prepare for consumers who wish to further the travel process by purchasing travel on their own time (Featherly, 2000). This cause for redevelopment could be a result of several factors: a consumer wanting more control over the travel-purchase process, saving money by purchasing one's own travel without a third party, or purchasing travel through the Internet by way of a fail-safe process, which provides the consumer with more confidence when purchasing travel with a credit card. Regardless, technology's rapid advances will ultimately help ease purchasers' minds about their travel concerns. 19 Defining Travel Provider Groups Travel Weekly, a major travel research organization, has studied consumer onlinepurchasing preferences and reported that people purchase most online travel through one of four purchasing selection methods ("On-line Travel Trends," 2001). The most preferred method of online travel-purchase selection is through travel-supplier sites. These are online companies that supply one or more travel amenities to the consumer. The most recognized members in this classification are airlines, car rentals, and cruise lines. The second most preferred method of online travel-purchase selection is through online travel agents ("On-line Travel Trends," 2001). These travel organizations have evolved from the traditional brick and mortar travel agencies who were service oriented (Shattuck, 2003). Traditionally, they met or communicated with travel purchasers and assisted them by providing a few sets of travel itineraries and amenities necessary to have a successful travel experience (e.g., providing passports or medical information). Brick and mortar agencies were commission-driven travel-itinerary bookers utilized by travel-supplying companies to sell reservations for their companies. When the Internet developed, many of these brick and mortar travel agencies were unprepared for the transition to an online market. Thus, after losing market share, these brick and mortar companies developed online segments of their business to keep pace with the new travel companies achieving success over the Internet. This shift in business strategy enabled the travel agent to enter into online enterprise. As online travel agents they still provided the same services they provided prior to the Internet, but looked toward extending their market to less service-oriented facets. The third most popular method of travel-purchase selection is through the utilization of travel search engines ("On-line Travel Trends," 2001). Travel search engines are one of the fastest growing markets in the travel industry (Sanfilippo, 2000). Travel search engines 20 such as Expedia and Travelocity are Web sites who pay travel destinations and suppliers to list their travel opportunities on the Web sites. People access these sites and put their travel interests, dates of travel, and destination into a categorical search engine, at which point the search engine will list all the possible options they have in their listings, based on price and/or time of arrival. The fourth most popular method of travel-purchasing selection is the destination ("On-line Travel Trends," 2001). These Web sites are developed by the actual destination to provide potential visitors with information and an opportunity to make a reservation. Some examples are Walt Disney World, Las Vegas, or Hilton Hotels. For this study, the methods of online travel purchasing will be categorized into two groups based on services offered: Internet travel agents or Internet travel suppliers (Featherly, 2000). The Internet travel-agent group serves the needs of the typically underserved markets, often considered "on the margin." These online companies want customers to feel as though they are communicating with an actual person during the entire purchasing process, providing the human touch, typically communicating through e-mail before and after the purchase of selected travel itinerary, as well as providing some additional services, such as obtaining passports and passport pictures and notifying the purchaser of destination events and activities. Internet travel suppliers are a conglomerate of the travelsupplier sites, search-engine sites, and destination sites. The categorization of this group is characterized by the limited services they provide the travel purchaser during the purchasing process. This group represents approximately 60% of the entire online travel-purchasing market (Del Rosso, 2001). 21 Increased Competition Research groups such as PhoCus Wright, Forrester Research, and Worldspan expect online travel revenues to exceed $60 billion by 2006 (Schaal, 2002). As the online travel industry grows, so too do the number of online travel companies, suppliers, and search engines interested in entering the online travel market (Featherly, 2000). A Gomez analyst by the name of Pappas said that the "big three" in the industry-Expedia, Travelocity, and Preview Travel-maintain about 40% of the online travel market, while the remaining online travel companies are struggling to gain market share (Featherly, 2000). What many supply companies have recently come to terms with is that they no longer have to pay commissions to travel agents, because consumers are searching for and purchasing travel on their own (Mandelbaum, 2002). Airlines were the first to begin this trend, accounting for 60% of total online travel purchases (Featherly, 2000). By eliminating commissions, travel-supply companies pressured travel agencies to compete based on price rather than service. The price-sensitive consumer reacted by rapidly adopting online travel purchasing through the readily available travel-supplier websites. The recent additions of online travel agencies have not seen the tremendous growth as experienced by the online travel suppliers. Research is needed to understand the dynamics of the changing consumer to further explain this shift in travel purchasing selection. The Changing Consumer Since World War II consumerism is just one of a handful of changes that affected travel purchasing. Literature indicates that during the 1950s and 1960s, most customers were largely satisfied with the products they received and the prices they paid. It was not until the late 1960s that consumers began to feel concerned about their buying habits. The revolution was triggered by the Thalidomide affair in the 1960s, when a scandal arose over a 22 tranquilizer that was given to pregnant women. Women were encouraged to take a tranquilizer to assist with morning sickness associated with pregnancy (Sjostrom & Nilsson, 1972). The drug prescribed to these women often caused severe physical and mental birth defects. The affair received worldwide publicity and began to create doubts in the eyes of the consumer concerning the safety of that product and other products that were available to the public. From that time onward, suspicions and doubts plagued products within the consumer markets (Sjostrom & Nilsson, 1972). These doubts and concerns were reinforced by investigations that revealed that advertisers were utilizing illusionary techniques and misrepresenting product features. Consumers began to feel that all companies were attempting to persuade them to purchase unnecessary and valueless items. Controversy swirled when Ralph Nader, a well-known consumer advocate/activist, questioned the safety of General Motor's vehicles, thus questioning the quality of their products. Although the hoopla did not establish validity in Nader's statements, it did reinforce the current state of consumer belief (Zummuto, 1982). The system was changing and becoming more beneficial to the consumer, but as much as it changed, it remained the same. Consumers could hold the manufacturers accountable for the safety of the products, but they were quite limited in selection and product availability. Manufacturers basically made a product available, and if the product functioned similar to what was marketed, then people purchased it. Therefore, every major aspect of a purchasing transaction through price and distribution channels was dictated by the manufacturers and suppliers (Lewis & Bridger, 2000). Due to the conflict that resulted from the 1960s, new consumer attitudes were evolving. This new consumer pushed the marketplace to take a closer look at what was being produced and hold producers accountable for the products they made available for sale. 23 Companies were reverting back to these early motivational theories and were attempting to become more familiar with consumer behaviors. As a result, over the last several decades, the primary emphasis in consumer-behavior research has been psychological (Eagly & Himmelfarb, 1978). Old Consumers Versus New Consumers Drucker (1994) described how every few centuries, society crosses what he termed "a divide." Divides are periods of time when society rearranges itself and people born then cannot even imagine the world in which their grandparents lived and their parents were born. During the early 1990s, such a divide occurred and was responsible for many dramatic changes to consumer behavior. The divide was an implication of the Internet and how society revolved around it. The old customers, who had limited discretionary funds, product choice, and product availability, found themselves looking in on a new world. These old consumers were left with the choice to maintain their current methods of product purchasing or transcend into the role of the new consumer. The new consumer evolved from the development of technology. These individuals are products of a new society that is limited by time constraints, lack of attention span, and trust (Lewis & Bridger, 2000; Underhill, 1999) rather than income, choice, and availability. New consumers are individualistic and highly involved with the process of production and consumption. They long to interact with the product or service and gain essential information about it. This new consumer wants to be informed, and media like the Internet have made information cheaper and more readily available than ever. The attraction of information partly lies in the fact that information over the Internet allows new consumers greater control over their spending (Lewis & Bridger, 2000; Modahl, 2000; Peters, 1996; Sutherland & Sylvester, 2000; Underhill, 1999), therefore making attributes like control, financial stability, 24 and education critically important to understanding the travel consumer. Control Internet travel consumers are changing, thus becoming more involved and informed when purchasing travel and travel products (Peters, 1996). With people becoming more experienced with online travel purchasing, they develop a desire for control over their purchases (Burger, 1992; Lewis & Bridger, 2000). Unfortunately, desire for control has not been examined in a consumer setting, let alone in online purchasing settings. Thus the following review of literature will discuss desire for control as an attribute and the role it plays in the purchasing process. Early Control Philosophy Researchers initially viewed early control philosophy as the equivalent of choice (Friedman & Lackey, 1991). In order for a researcher to gain an understanding of the concept, they had to test the effects that loss of control presented in relation to choice. Researchers limited the number of choices available to each individual and tested the effects of increased control throughout the experiment. The experimenters discovered that increasing one's control further increased the number of new choices the respondent had available, thus demonstrating that in order to increase control, a person must have more options in order to make a decision (Glass & Singer, 1972; Langer & Rodin, 1976; Perlmuter & Montey, 1979). However, as the area of control was further explored, it became evident that this early understanding was not sufficient in explaining the control construct (Skinner, 1995). An example of this inadequacy is in relation to how a person purchased travel over the Internet. If one contacts a travel agent over the Internet and provides them with their basic information regarding their preferences, the travel agent will put together a limited group of packages for 25 that destination. By having just a few options, the travel purchaser may perceive fewer choices, but the question remains as to how they perceive an abundant number of alternatives, such as those provided by an Internet travel company able to search a multitude of travel information. This plethora of choices could be viewed as overwhelming and, at least in this instance, may leave the purchaser with the feeling that he has no control. One might expect that the more choices an individual has, the greater the control, but a sizeable number of travel options may have the opposite effect and the consumer perceives he has lost control. People may feel they lose control for a number of reasons. In this instance, purchasers may have a time restraint limiting their ability to review all the possible travel sites prior to making travel arrangements or lack the experience or confidence in their decision. Both are issues of control. Regarding the issue of time, Kiesler (1966) discovered that individuals normally spend more time selecting between 2 options than 4, which leads the researcher to believe that a person perceives there to be an optimum number of possible options. Too many alternatives may become overwhelming and confusing, but too few options may cause a person great difficulty in selecting the best suited option. In short, there is a preferred point that individuals regard as the optimum number of selections, but these actual effect points are unknown and need further attention in future research. Control has also been equated with power, but in fact control is not an equivalent to power (Langer, 1983). To support this statement about control and power, imagine that you have been asked to be president of a major corporation. It is your first day of work. For most people, the promotion (an increase in power) would be met with increased feelings of anxiety and insecurity or loss of control. Power is the ability to influence tangible outcomes, whether that power is achieved 26 through money, muscle, knowledge, or anything else that may affect an individual or others. People vary in their desire and need for power (McClelland, 1987), and the need within individuals varies across one's life span (Erikson, 1974; Levinson, 1986). Control is a more intrinsic variable that is less concerned with the current state of the external world. The need for control remains stable, although aspects of the external environment threaten that perception of control and are likely to change the experience, thus limiting control as power. The psychology of control centers on the control of one's perceptions of reality (Langer, 1983). The exercise or nonexercise of control is significant in virtually all facets of an individual's life. When one believes his control is threatened, the result could be severely incapacitating, because the need for control is the reason we search the environment for information (Heider, 1958). It is this basic interpretation of human personality and its associated attributes that enables us to explain our behaviors (Jones & Davis, 1965; Langer & Dweck, 1973). Human Personality Matthews and Deary (1998) reported that Aristotle was one of the earliest to look at personality characteristics in the "Nicomachean Ethics," and that much of the initial work surrounding personality traits developed out of folklore. One of the first areas of folk psychology that developed these viewpoints was based on astrology. Astrologists developed 12 personality traits based on sun signs that could be used to define human characteristics. Another early aspect of human personality was developed from the Chinese. The Chinese believed that certain personality traits were common within individuals who were born during a specific year. These traits were then defined by using animalistic traits to explain the individual's personality. For example, those who were born during the Year of the Cow are said to be contentious and hard-working. Personality-trait research has a very 27 unique development in that it can be traced from early folklore medicine and natural language. Among the earliest contributors to personality-trait research were Hippocrates (ca. 460 BC) and Galen of Pergamum (AD 130-200) (Matthews & Deary, 1998). The Hippocratic conception was a medical theory based upon humours, or blood fluids such as blood, phlegm, or vomit. It was Galen who developed a terminology focusing on these fluids. He defined the terms melancholic, choleric, phlegmatic, and sanguine, which are still used today in some areas of the medical profession. Kant redefined these terms to represent dimensions of feeling and activity (Matthews & Deary, 1998). His typology emphasized the psychological nature of the temperaments and became known as the four humoral temperaments. These terms, later used by the founder of modern psychology, Wilhelm Wundt, were described in terms of two dimensions: weak emotions and changeable-unchangeable activity, which are very similar to the present-day works of Eysenck and Eysenck (1985) and Goldberg (1993), which concerns dimensions of extroversion and neuroticism. Early work surrounding personality traits was from researchers like Galton (1886), who hypothesized that personality trait might be represented in natural language terms and identified some human-characteristic-type descriptive terms, later dubbed the "Lexical Hypothesis." However, it was not until a series of works completed by Dutch psychologists, Heymans and Wierma between 1906 and 1909, that psychologists began rating individual characteristics in an attempt toward trait classification (Matthews & Deary, 1998); thus developing a conceptual method similar to factor analysis, which introduced three dimensions of personality: emotionality, extroversion, and introversion. Others followed these works, but statistical procedures during this time were crude at 28 best and offered limited findings. The researchers who followed, however, using the earlier theoretical trait developments, believed much of the work of the time to be common sense driven. It was not until 1938 that Carr addressed many of the issues behind personality traits. Carr was the first to emphasize the predictive nature of traits in regard to a person's likely future behavior. Furthermore, Carr (1938) stated that personality traits were not directly observable and may only be inferred from behavior, expressing a need for reliable trait scales in order to compare and classify individual traits (Pervin, 1994). Measuring Personality Traits Due to the diversity among personality traits, it has been difficult to develop a method to accurately measure them. The simplest recognized method of measuring personality is by asking a person to rate how well different trait adjectives apply to himself or herself (Matthews & Deary, 1998). Furthermore, supporting information regarding one's personality can be obtained by asking questions that relate to a person's observed behavior. When measuring personality traits, there is no guarantee regarding the accuracy of the information collected when using only self-descriptors or objective behaviors. Studies that have attempted to use only one of these techniques for a study were found to have limited success and little validation (Kline, 1993). However, Carr (1938) stated that many core issues of trait psychology at a conceptual level are predictive by the nature of traits. After all, it is the general assumption of the trait approach that people do in fact tend to behave consistently in different settings, thus further explaining that by knowing an individual's personality traits one can often provide predictive information regarding the likelihood of future behavioral tendencies. Carr (1938) further articulated that traits may only be inferred for behaviors and emphasized the need for trait scales in order to compare individuals on different characteristics. 29 When developing an instrument to measure traits, one must take great care in the composition of questionnaire items; all items on the questionnaire must be easy to understand, applicable to all respondents, and unlikely to cause offense (Angleitner & Wiggins, 1986). In addition, it is important to check that the items are not strongly influenced by response sets or biases, which can occur when testing trait adjectives that contain similar meanings. By following these basic rules, researchers in this area have experienced the most success and have been able to obtain strong predicative validity that the trait measured is correlated with some independent index of quality associated with the trait. However, the ultimate goal of personality-trait research is to establish construct validity (Matthews & Deary, 1998). The construct validity is the correlation between the trait and external criteria that are established in advance from a scientific theory. The difficulty with construct validity is that of establishing scientific truth, hence construct validity is somewhat provisional. The previous method can be utilized to obtain an effective scale for measuring a single item, but one cannot arrive at an acceptable model of multiple personality scales simply by accumulating different traits. The technique most widely used for multiple-trait identification is factor analysis (Bentler, 1995; Cattell, 1971; Gorsuch, 1983; Joreskog, 1973; Kline, 1994). Factor analysis is used to develop a correlation matrix between listed items utilized on the questionnaire. The goal of the correlation matrix is to enable the identification of one or more underlying dimension or factors which account for a majority of the variance in the individual's item scores, and the factors are defined by the items that correlate with them. The matrix of a simple structure can be rotated, as the most meaningful solution is when the factor interpretation is most apparent. However, simple structures can only suggest a hypothesis. Another approach that allows for hypothesis testing is confirmatory factor analysis. 30 Confirmatory factor analysis enables the pattern of factor loading for a given set of items tested with the sample and specified in advance (Joreskog, 1973). In this case, the factor analysis calculates the solution which is closest to the hypothesized matrices, determining whether or not the collected information provides an acceptable initial hypothesis. The researcher can also select an oblique rotation, which enables derived personality factors to be correlated. If the factors are correlated, another factor analysis can be calculated between the correlations and the factors themselves. The second factor analysis will identify the secondary factors, which enables the researcher to obtain a broader set of personality factors by narrowing personality-trait measures. Many instruments have been developed and utilized for assessing personality traits, but few have been used as frequently as the Sixteen Personality Factor Questionnaire (16PF). Often referred to as the Cattellian Project, this questionnaire is recognized as one of the most ambitious instruments ever undertaken in psychology (Eysenck, MacLeod, & Matthews, 1987; Matthews & Deary, 1998; Mischel, 1968). Cattell's questionnaire originally began with 23 functional trait factors, from which the sixteen most robust factors were selected for the 16PF. However, the 16PF received some speculation because of the instrument's relatively low internal consistency. In an attempt to improve the measure's internal consistency, Conn and Rieke (1994) developed the 16PF 5 where the numeral 5 refers to 5 second-order factors. The new personal-factor scale's internal consistency was moderately better, calculating a mean Cronbach's alpha of .74. However, by increasing its internal consistency it lost much of its comparability to the original 16PF, thereby causing some researchers to doubt the predictive validity of the measure as well as its construct validity (Kline, 1994). Several other personality-trait measures have also been developed to assess 31 personality traits, but have more severe problems than those exhibited by the 16PF (Kline, 1994). One such method is the California Psychological Inventory (Gough, 1987). This widely used measure assesses 18 different traits with moderately good reliability; however, factor analysis was not used to determine these 18 factors. Rather, the method of criterion keying was used. This method selected items on the basis of their ability to discriminate criterion groups. The problem with this method is that the scales may not correspond well with those utilizing factor analysis, and with the absence of systematic experimental studies, the construct validity is lacking (Kline, 1993). Another personality scale developed to measure personality traits was the Occupational Personality Questionnaire (Saville, Holdsworth, Nyfield, Cramp, & Mabey, 1984). This scale measured for 31 different relevant traits for personnel recruitment, career development, and training. The scale is reasonably reliable, but the model is not explicitly based on factor analysis, thus making it difficult to correspond with other personality measures. Two of the more highly regarded personality models are the Gigantic Three, developed by Eysenck (1967), and the Big Five, developed by Costa and McCrae (1988). Eysenck's model focused on three personality dimensions: psychoticism, introversionextroversion, and neuroticism. According to Eysenck, the Gigantic Three model is a conglomerate of trait descriptors that fall into one of the three classifications. Each classification has its own scale, in order to best determine a person's most common human characteristic. Eysenck and Eysenck (1985) explained that a typical extrovert will score high on the introversion-extroversion scale. These individuals are generally sociable, crave excitement, take chances, and are moderately unreliable. On the other hand, the characteristics of an introvert describe someone who is quiet, fond of books, and very 32 serious, with high ethical standards. The high neuroticism personality refers to someone who experiences anxiety and depression, often worries, and has difficulty sleeping. Unlike the high neuroticism, the low scorer recovers quickly after an upsetting experience and is usually calm and unworried. The individual high in psychoticism is solitary, often troublesome, sometimes cruel, aggressive, and has unusual traits (Eysenck & Eysenck, 1985). Costa and McCrae's (1988) 5 trait factor model consists of neuroticism, extroversion, openness, agreeableness, and conscientiousness. The scale is made up of 240 questions, with 48 representing each classification. The high neuroticism is reflected by high anxiety, hostility, depression, and vulnerability. High extroversion is characterized by warmth, assertiveness, and excitement seeking. The openness personality centers around fantasy, aesthetics, feelings, and values. People within the agreeableness personality group are trusting, straightforward, modest, and compliant, while the conscientiousness personality focuses on competence, order, achievement, and self-discipline. These two models are the most frequently used in trait research, but there are criticisms revolving around them. These are broad-based conceptions of personality and it is commonly seen that the narrower trait concepts can fall into more than one of the dimensions listed within each personality-trait type (Zuckerman, Kuhlman, Thornquist, & Kiers, 1991). Other theorists argue that 5 factors are too few to represent the major dimensions of personality (Hogan, 1986). Narrower Trait Concepts For this study, the large personality scales and models will not be utilized because the researcher is interested in a narrower trait concept. Narrower concepts do not profess to cover the main areas of human interindividual differences and make no attempt to explain the broad-based conception of personality (Kluckhohn, 1953). Instead, they are to assist the 33 researcher in explaining individual personalities such as desire for control, locus of control, Type A behavior, and self efficacy. In this study, the researcher is interested in the trait concept of control and the role it plays in the purchasing process. The following literature will examine the history of control and introduce some of the more recognized theories regarding the control constructs. A review of control constructs is needed to identify the most appropriate method for this study. Control Constructs One of the main difficulties in examining control is determining what you are interested in studying. "Are all control constructs similar?" or are they simply "control by another name" (Rodin, 1990)? Are they completely different, examining separate facets of the experience of control? The literature suggests that they are distinctive in nature (Burger, 1992; Rotter, 1966; Wright, von Bussmann, Friedman, & Khoury, 1990) but also interrelated, perhaps even overlapping. Psychologists have focused on various aspects of control and have suggested that control is neither a simple choice nor power. Averill (1973) separated control into three categories: behavioral, a direct action taken in the environment to influence events; cognitive, how control relates to the interpretations of events; and decisional control, the opportunity to select from possible actions. According to Averill's (1973) research, these control categories are independent of each other, but can have some overlapping tendencies within variable definitions. Furthermore, Langer (1983) explained that most researchers perhaps utilize several of these definitions to guide their work. Accordingly, all the control beliefs are part of a larger system of knowledge, referred to as the competence system, which focuses on the regulating and interpreting of goaldirected interactions with the environment (Connell, 1990; Connell & Wellborne, 1991; 34 Harter, 1978; Skinner, 1995). Many related control theories explain how control experiences contribute to the construction of beliefs and how these beliefs promote or undermine effective interactions (Skinner, 1995). The competence system depicts a sequence of events in which individual beliefs contribute to action outcomes, which influence the success of the actual performance, thus impacting future beliefs. Control can be examined under the focus of perceived control through four main theories: self-efficacy, locus of control, learned helplessness, and causal attributions. In 1977, Bandura developed a new expectancy construct: self-efficacy. He later added that self-efficacy refers to one's belief about his or her ability to perform a specific behavior (Bandura, 1986) and is a construct based on cognitive and behavioral concepts described through an individual's perception of his or her skill and abilities and whether the skills/abilities produce effective and competent actions. In the formulation of self-efficacy, Bandura (1986) explained that a majority of theories regarding perceived control center on beliefs about the effectiveness of responses in causing outcomes. Skinner (1995) supported this view. However, these responses should have no effect on behavior unless one also has conviction that they can successfully exhibit the behavior necessary to produce the outcome (Bandura, 1977). Therefore, it is important for people to believe that they can exercise some control over stressful situations, thus becoming less impaired in their performance (high efficacy). However, if a person lacks personal control he will often be unable to put the situation in the proper perspective, thereby possibly encountering difficulties when attempting to carry out the most situationally proper actions. This implies that having control is better than not having control, at least in stressful or adverse situations. The term locus of control is often used interchangeably with self-efficacy (Skinner, 1998). While self-efficacy focuses on the perception of ability to act competently and 35 effectively, locus of control focuses on the perception of control (Bandura, 1977). Rotter's locus of control was originally introduced as an expectancy term (Langer, 1983), but as the theory evolved it was viewed almost exclusively as an individual difference. Locus of control (Rotter, 1966) refers to one's belief in his or her abilities to control life events. An individual with internal locus of control believes that outcomes are related to his or her behavior, while an individual with external locus of control believes that outcomes are not related to his or her behavior but to external forces beyond his or her control. Individuals with an external locus of control may perceive life events to be controlled by luck, chance, fate, or powerful others. Individuals with an internal locus of control are more likely to change behavior following reinforcement than individuals with external locus of control (Marks, 1998). Uncertainty of success has pushed people to understand that there is an overlap between skill and luck (Langer, 1983). Skill situations, in general, occur when there is a common link between behavior and outcome, thus, skill tasks are controllable. Luck, however, is a fortuitous happening which is usually a situation out of one's control. This lack of control is often referred to as chance or uncontrollability. If chance truly exists, and people understand that fortuitous situations are not controllable, why is it that so many individuals act as though these events are subject to control? Moghaddam and Studer (1998) stated that humans seem to alter their ideas or actions so that they are in harmony with one another. The process can be done subconsciously, but inevitably some level of control occurs either by thought or action, leading individuals to believe that they are somewhat in control of the event. Moghaddam and Studer (1998) used the following example: A woman friend had married a man we will refer to as David, who was one of two original suitors for the woman. The woman was originally interested in the other man and planned to marry him. During the 36 planning of the wedding, the honeymoon planning, and the purchase of a new home, the other man received a promotion at work that obligated him to move from the area, thereby causing him to leave her. The woman then became interested in David and later married him. She had explained that it happened just as she planned, that she never expected to marry the other man and that she and David were meant for one another. The story demonstrates just how individuals believe they control a situation and how they justify their control, even though in this situation it was evident that the woman was not really in full control of her fate. The end result, however, was favorable and she believed that she controlled the situation, thereby eliminating any outside chance. Hence, the event was predictable and the final result had been anticipated, so the event was controlled. Whether an event is determined by skill or chance may depend on factors that are relatively independent from the actual contingency. Several researchers have studied this tendency for people to attribute desirable outcomes to internal factors and blame external factors such as chance for failures (Cohen, 1964; Feather, 1969; Langer & Roth, 1975; Streufert & Streufert, 1969). They have determined that valence of the outcome would not be a potent factor if people distinguished between chance and skill on the basis of an objective contingency. As a result, people have developed an illusion of control or think they are in control of the situation, even though they are not. In 1975, Seligman's research provided a conceptual framework within which one could manipulate important parameters of the control process. The work examined what has become known as learned helplessness. The key proposition was that when people and other animals (e.g., dogs and rats) experience adverse events that occur independently of their own responses, they learn that no effective responses exist, that they are essentially helpless. According to Seligman (1975), a person gains control when a desired outcome occurrence is 37 dependent on that person's response. The outcome will not occur without the individual making some response to bring about the action. It is believed that perceived independence between one's response and forthcoming outcomes results in learned helplessness and depression (Klien & Seligman, 1976; Seligman & Maier, 1967). Furthermore, Abramson, Seligman, and Teasedale (1978) suggested that for people, the effects of uncontrollability are dependent not only on the belief that one's response does not matter, but also on the explanation one provides for why they believe this to be so. Hence, treatments based on these models direct helpless individuals to focus their beliefs and responses on controllable outcomes. Additionally, one must be mindful of control for one to perceive control, and it is the person's perception of the action that is crucial rather than one's overt responses (Langer, 1983). Therefore, seeing an animal as friendly and nonthreatening does not produce feelings of control unless the animal was previously perceived as dangerous or threatening. One must have a conscious appreciation of the alternative in order to experience a meaningful interaction with the environment. It must be remembered that even though the environment always allows for decisional control, at least one can choose between taking an action and not taking an action. That being said, unless the individual sets his mind to noticing possible distinctions in the environment, control will not be perceived (Averill, 1973). Causal attributions, though somewhat discussed in the purchasing section of this document, take issue with Rotter's (1966) locus of control, namely the link between internal and external shifts. Weiner (1985) insisted that the causes used to distinguish the internalexternal dimensions differed on another dimension: stability. Weiner (1985) noted that when something negative or unexpected happens, people want to know why it happened. The cause of the event can be attributed to a number of items. Weiner has placed the attributes into four 38 different dimensions: internality, stability, controllability, and intentionality; thereby explaining that these dimensions predict many important outcomes, such as emotions, behaviors, and motivation. Research regarding control depends on recognition of how control affects a person's individual behavior regarding internal and external environments. The early works by Bettelheim (1943) and Cohen (1964) were focused on the role of internal and external environments. Rotter (1966) solidified the locus of control theory using a similar model including internal and external control. The Rotter publication was one of the first to advocate that people differ in the amount of control they feel they have over the positive and negative events they experience within their lives (Mannell & Kleiber, 1997). The Rotter (1966) document argued that people could be placed under observation and that the typical events of their lives could be categorized as situations under their individual control or under the control of external forces. More precisely, people with an internal locus of control believe that fate is largely the result of their action, but people with external locus of control believe that good or bad luck, the environment, or someone else is responsible for their fate. Because Internet travel purchasing focuses on internal locus of control, as described by Rotter (1966), external locus of control was not examined for this study. As Averill (1973) previously explained, control must be defined in one of three ways: behavioral, cognitive, or decisional. For this study, control will be defined as behavioral and will require the researcher to examine the control behavior from a narrower perspective, primarily high desire of internal control. Type A Behavior The Type A behavior concept is important to examine due to its high desire of internal control. Type A behavior represents a conglomerate of extreme behaviors that can 39 contribute to personal health concerns. Friedman and Rosenman (1974) define a Type A pattern [as an action-emotion complex that can be observed in any person who is aggressively involved in a chronic, incessant struggle to achieve more and more in less and less time, and if required to do so, against the opposing efforts of other things or other persons. (p. 67)] In other words, the normal or classic case of Type A behavior is competitive, pushy, time-urgent, and carries around a great deal of hostility, especially when confronted by barriers that inhibit the opportunity to achieve. The absence of these characteristics is termed a Type B pattern (Houston & Snyder, 1988). The early interest in the Type A pattern was in the medical field, because a lot of evidence supported that the Type A personality was an independent risk factor for heart disease. With the completion of more research, evidence is shifting viewpoints away from that of risk toward the subcomponent of behavior and other personality factors. According to Glass (1977), Type A behaviors attempt to maintain and/or regain control over the stressful parts of the environment. Therefore, Type A personalities are thought to desire more control than those with Type B patterns, who tend to lose control during difficult events. The Type A pattern has been associated with functions other than medical. Psychologists have examined it in academic performance and vocational success and for psychological tendencies. One instrument developed to assist in these areas was developed by Wright et al. (1990) and deemed the way of life scale. The reason for its development resulted from inconsistencies of other scales, such as the locus of control scale. The way of life scale was compared with the locus of control scale and the desire for control (Wright et al., 1990). Furthermore, Wright et al. (1990) were surprised to discover that the way of life 40 scale correlated well with the desire for control scale but did not correlate well with the locus of control scale. The way of life scale was developed to study control, particularly in those with unusual levels of desire for control (Wright, May, & Jackson, 1991; Wright & Newman, 1994), which directs this study away from an unusual level of control to that of individual desire for control. Desire for Control Desire for control differs from Rotter's (1966) concept because locus of control focuses on an individual's perceived control, while desire for control is concerned with the extent to which people want control (Burger & Cooper, 1979). A person can have an external locus of control and have a high desire for control (Burger, 1992). In other words, an individual may be controlled by some external element and have little control over situations, but it does not mean that the person does not desire to overcome this control (Skinner, 1995). For example, a person may be bound to purchase travel through a travel agent due to time constraint. As a result, the individual did not receive as many selections as he would like, but because of the constraint he must select from what the agent had made available. A person with a strong desire for control wants more selection but is limited because of some external factor. Therefore, for future travel purchases, to increase travel selection the individual may identify some travel-purchasing sites on the Internet and surf those sites in order to increase selection opportunities, then present these options to the travel agent to increase the option for future travel. The opposite is true as well. A person may not have a strong desire for control. The travel agent is an appropriate way to reserve travel without maintaining a high level of control. The individual receives a basic selection, makes a decision, and rests easy, knowing that responsibility falls into someone else's hands. Furthermore, a person with a very high 41 level of control may not want any assistance from an outside source, thus designing and purchasing one's own travel package directly through the Internet. Something may have happened in the individual's past to cause this direction of purchase, possibly overcharging or lack of selection opportunity. Therefore, the individual feels it is necessary to take complete control of the situation and take responsibility for his fate. Researchers have recognized that people do not always prefer to take control (Burger, 1989; Folkman, 1984; Thompson, Cheek, & Graham, 1987). Sometimes the increased control of an event means increased responsibility and increased pressures to perform well. While high pressures can lead to increased performance; it may not be desired, thereby creating problems and in some cases danger (Burger, 1987). Some individuals may prefer to have another person take control because they feel others can do a better job or are able to avoid negative outcomes. In short, many psychologists have made their case that people generally prefer to have some level of control over what happens to them (Burger & Cooper, 1979; Friedman & Lackey, 1991; Langer, 1983). Control varies from person to person and the individual will respond in a way that is familiar from past experiences. Therefore, the desire for control construct will be further examined as a key attribute. Control as an Attribute A key connection for this study is the designation of control as an attribute. Therefore it must be determined where control falls in the attribution process of purchasing. Since the earliest writings of Heider (1944, 1958), Jones and Davis (1965), and Kelley (1967), numerous studies have been completed focusing on causal explanations for given events (Jackson, White, & Schmierer, 1996; Zelen, 1988). One way of describing an event is by stating what caused the event. It should be noted that there are considerable differences between scientific inquiry and common sense, but the similarities are significant enough to 42 believe that modern methods of scientific inquiry can be used to understand an explanation of everyday events (Jaspars, Hewstone, & Fincham, 1983). The foremost influential study regarding this theory was completed by Heider in 1944 and furthered by him in 1958; his study focused on the theory of naïve analysis of action. Jones and Davis (1965) are credited with furthering attribution research with the inclusion of intentions and dispositions. Lastly, Kelley's studies (1967, 1973) discussed and explained attributional covariation and configuration. Heider is considered to be the founder of attribution theory, with his early writings in 1944 and 1958 (Jackson et al., 1996; Jaspars et al,. 1983; Mischel, 1968; Weiner, 1986; Zelen, 1988). Heider's naïve analysis of action theory attempted to explain the process of how an untrained observer makes sense of the actions regarding others. The process has been referred to as cause and effect, or actor and act. Heider's interests directed him toward understanding the varying degrees of similarity between the two parts of the unit, cause and effect or actor and act. In Heider's (1958) study, he revealed that certain factors, such as similarity and proximity, were determined to be valuable components of the locus of attribution (Hewstone, 1983). In other words, if two events were similar to each other, or proximate, then one is likely to be seen as the cause of the other. To take it a step further, when observing the actor-to-act relationship, a person's attribution is more likely to be the cause of an event rather than the situation. Heider (1958) further acknowledged that people do experience consequences for their actions and hence have and want some control over their causal experience. Heider (1958) stated, "If a person believes that the lines in his palm foretell his future, this belief must be taken into account in order to explain certain expectations and actions" (p. 5). Therefore, naïve analysis of action deals with how observable behavior is linked to unobservable causes, 43 thus explaining the cause of an activity as either internal or external attribute. Internet attributes are factors within the person (e.g., effort, intention, and ability) and external attributes lie outside the person (e.g., type of task or luck). By understanding the factors and behaviors of another person, one can make one's own world more predictable and provide oneself with a sense of control over one's experiences (Hewstone, 1989). Jones and Davis's attribution-theory research (1965) was an attempt to formulize how individuals make inferences about intentions and dispositions. They reasoned that human actions are often explained by attributes and relatively invariant dispositions to what they commonly refer to as the actor (Jaspars et al., 1983). In other words, it is the contention by the observer to decide on the effects of an observed action and determine which were intended by the actor (Hewstone, 1983). Jones and Davis argued that the attribution of intention to the actor was the precondition for the dispositional attribution and proceeded to develop sets of fundamental standards for other attributional steps (Hewstone, 1983). The model developed by Jones and Davis (1965) demonstrated one of the few attempts to deal explicitly with the use of different rules for making different kinds of attributions. Kelley's (1967, 1973) process-oriented attribution theory dealt explicitly with the problem of attributing events, usually behaviors, to their causes. From this theory, an abstract model regarding the process was developed in which such attributions are created (Hewstone, 1983, 1989; Jaspars et al., 1983). The model included covariation principles such as consensus, consistency, and peculiarity, in addition to configuration principles like discounting, augmentation, and compensation. The importance of Kelley's work follows that principles have been used to account for many diverse causal phenomena but have never been applied to the actor's responsibility for action and outcome (Jaspars et al., 1983). While there have been other significant developments involving attribution theory, 44 these three theories are considered to be the major early contributors that set the foundations of most attribution research conducted today (Jackson et al., 1996; Jaspars et al., 1983; Mischel, 1968; Weiner et al.,1971; Zelen, 1988). These studies show that a variety of items and factors get attributed in psychological attribution theory, yet different types of attribution, particularly causation and responsibility, depend on different principles which have not been addressed in the research. The attribution-theory fundamentals established by these three works guided Weiner et al. (1971) to further make inquiries regarding causal relationships. In their 1971 study, they discovered that attributes in successful and unsuccessful experiences take on both cognitive and emotional responses. Furthermore, supporting research by Weiner (1986) confirmed that people use information about the situation they have participated in to form an explanation for the successful or failed experience, thereby explaining that the individual's explanation is the result of the factors that are attributed to a person's success and failure. Once the attributions have been determined, it has certain predictable emotional and behavioral characteristics (Weiner, 1985). However, much of the literature regarding attributions in success and failure scenarios tends to revolve around laboratory studies and educational settings (Antaki & Brewin, 1982; Elig & Frieze, 1979; Frieze, Bar Tal, & Carroll, 1979; Frieze, Moss, & Olson, 1988). Unfortunately, most of that work has been subject to criticism resulting from significant but subjective performance ratings. Regardless, because there were some encouraging results, researchers have decided to study attribution success and failure outside controlled laboratory and education-type settings (Jaspars et al., 1983). One such study completed by Frieze et al. (1988) looked at attributes that individuals determined were necessary to be successful at work and attributes that companies determined 45 were necessary to have successful employees in a work setting. The study revealed that most good work settings are developed by companies attempting to have good employees a nd those employees were the result of well-developed company programs and philosophies. In order for a company to be successful in this process, the company must match the attributes that the employees deem necessary with those attributes that the company deems necessary. The problem with the system is that some people buy into the philosophy and others do not, so the company must sell the program to its employees and explain the benefits associated to the success of such a program. The result is that companies who succeed in developing good work settings must market to their employees (see Figure 1). First, both the individual and the company must define what they deem necessary to be successful. In the marketing of a product or service, buyers define what they are looking to buy and the marketer develops a definition of the product or service they are looking to sell. Second, in order to be successful in the workplace, the employee needs to be able to find information regarding what the company wants from a successful employee, plus the company needs to be able to access pertinent information explaining their success. When purchasing a product or service, the marketer must provide information to the potential buyer in the form of Web pages or newspapers to search through to make a proper decision. Marketers search these forms of media for information regarding what consumers need and desire. Third, both employees and companies make attributions for why the program was successful, and in the purchasing scenario the marketer and the buyer both make attributions for the reason why the product or service was purchased. In conclusion, the two systems are closely linked because they are both marketing programs designed to sell something. 46 Granted, success in the workplace may not seem to be closely related to marketing, but when you look at it from a process perspective, they are very similar. After all, the company invested money into the success in the workplace program hoping to develop a strong and efficient staff. The company is essentially trying to purchase a successful program, similar to how a person purchases a product or service. However, the purchasing process seems to be a little more complicated than what has been presented by the model regarding the workplace, thus it is important to discuss several definitions and steps involved in the purchasing process. The Purchasing Process Prior to explaining the connection between the attribution process and the purchasing process, a working definition of the purchase needs to be introduced. According to MerriamWebster's dictionary (2002), a purchase is to obtain something by paying money or equivalent. The key word used to define this process is obtain. One cannot purchase a product, or in some cases a service, without knowing where to find it. Someone must position the product or service in a location where others can find it in order to purchase it. The task of positioning the product or service is termed marketing. Marketing is explained by Kotler, Bowen, and Makens (1996) as business philosophy that is more than a function carried out by a marketing department for advertising purposes; rather, it is a way of doing business that is focused on meeting the needs and desires of customers, whether in company philosophy or in product development. Therefore, when someone purchases a product or service, the marketer and business generally believes that they have succeeded in meeting their goal, and when a product or service is not purchased, the marketer and business believe they have failed to meet their goal, hence the beginning of the relationship between purchasing and attribution theory. 47 The purchasing process consists of five steps and a set of outcomes as listed in Figure 2. The first step is the definitional matching that occurs between an individual purchaser and marketer. A demonstration of how a person purchases a travel itinerary through the Internet may better illustrate the model. First, the potential purchaser must select or define a destination to visit. Once the individual has determined a destination, he must decide on the preferences he would prefer during the travel opportunity. Often this phase is referred to by the marketer as consumer behavior, but to the buyer it should be termed purchasing behavior. These preferences can be related to any number of individual characteristics. For example, a person desires a certain level of control over the purchase or has certain preferences due to a previous visit, thus matching his own specific travel needs with options made available by travel marketers. Next, the person searches for a travel itinerary that best suits the needs that have been singled out. Once the decision regarding the travel itinerary and travel options is made, the person must then select a method of travel purchasing. When purchasing over the Internet, potential buyers have the opportunity to purchase travel one of two ways: (a) through Internet travel agent or agency, or (b) directly from an Internet travel supplier such as Southwest Airlines, Continental Airlines, Expedia.com, or Travelocity. Lastly, the purchaser develops an internal perception for why he purchased a specific travel product or service. Maybe the cost of the travel product was low or the person needed to get away from life's frantic pace, and this package was the one that best met the perceived needs. Regardless, the entire process provides the purchaser with an experience, and from that a purchaser develops feelings of either satisfaction or dissatisfaction toward a product, service, company, or the purchasing process. In turn, these feelings evolve into the beliefs of the purchaser. It is from these beliefs that have occurred during the purchasing process that 48 consumer behaviors develop and influence future purchasing decisions. The travel marketer must be aware of this purchasing process and develop methods to meet the needs of its consumer, thus, they must first define or develop travel itinerary or travel packages for destinations. During this process they need to develop several different travel options to meet the needs of potential purchasers. Many travel purchasers have unique preferences, but several choices are common with regard to consumer behavior. For instance, marketers may realize that a relationship exists between a potential purchaser's computer proficiency and computer accessibility in regard to purchasing travel over the Internet. Developing Internet travel-agent sites that require little to no computer ability in order to purchase a travel product or service would therefore be ideal and meet the basic needs of many potential travel purchasers. Once a travel agency has developed these packages, it must try to identify the unique consumer behaviors and target those specific markets. When using the Internet, marketers develop Web pages as forms of advertisement to reach possible clients. It is through Web pages that consumers can access any number of sites with the hope of finding the best travel opportunity. In an attempt to appeal to the potential purchaser, marketers make available the different methods of purchasing the travel product or service. Upon completion of the sale, the marketer then assesses the travel purchases and tries to further understand why the person made that particular selection and determines potential travel markets. When the company succeeds in a sale of a product or service, the company's revenues increase, opportunity for the travel company to develop new travel packages becomes available, and/or the company is able to offer unique travel promotions due to increased revenues. So what has transpired between the travel company and travel buyer is a meeting of the minds. If the two do not arrive at a consensus, then the purchaser will look for other 49 travel opportunities and the marketer will hope to attract other potential buyers. However, if the travel company has a product that appeals to the purchaser's defined service and matches the needs that the individual expects, then the purchaser should be able to find the product or service when conducting his travel search based on the purchaser's behavioral characteristics. For this study, the researcher is particularly interested in the step between the purchaser's information search and the travel-purchase selection. In times prior to the Internet, the purchase was simply considered a given phase of the purchasing process (Pizam & Mansfield, 1999), but with increased consumer control that has changed. It is between these steps that the new consumer has developed and where desire for control exists. The control construct is different than traditional purchasing behavior research, because purchaser and marketer attributional behavior matching occurs prior to the information search. This control of purchase behavior is recognized as occurring after the information search. Traditional Purchaser and Consumer Behaviors Purchaser and consumer behaviors are virtually identical, with individual differences occurring within terms of observation perspectives. The purchaser's behavior demonstrates certain characteristics which marketers assess and categorize (Modahl, 2000; Underhill, 1999). Astute businessmen have long known that in order for a company to be successful they must create customers (Drucker, 1954). These consumers are responsible for purchasing goods and services that will ultimately determine the level of success that an organization will reach. Usually within these companies, individuals are designated to a marketing department. The main task of this department, whether operated by an individual or group of individuals, is to develop a customer base and track preferences and behaviors of purchasers. By demonstrating how behavioral preferences are formed and modified, professionals are able to gain an understanding of several aspects pertaining to consumer behavior. 50 Researchers have completed analysis on preferences, their origins, and their modifications by examining the cognitive aspects of preferences and choices (Nozick, 1990). A paradigm commonly used in categorizing consumer-behavior research has been the following: (a) specify the attributes of objects and products, (b) find the values of these attributes, and (c) determine how these attribute values combine to produce an overall preference (Abelson & Levi, 1985). Modern consumerism has undergone many changes before identifying such methods of defining individual preferences. It is generally accepted that consumer behavior theory is a post-World War II phenomenon that resulted from the emergence of motivation research, but other theories were developed prior to the war. Most of these early theories evolved from psychological and sociological studies, more specifically, human motivation. Human Motivation, Drive Theory It is from the works of Thorndike (1899) that the study of motivation begins. Thorndike began his work in the area of animal behavior in an attempt to bypass some of the barriers experienced by other scientists when studying human behaviors. From his study, Thorndike (1899) was interested in the effect of hunger on the activities of animals, from which he developed the term impulse to describe the animal's motive of action. An impulse would drive the animals toward a particular reward, or valence. Later psychologists who followed in the behaviorist tradition substituted the word impulse for drive (McClelland, 1987) and additionally developed the terms motive and desire to describe impulse. In 1911, Thorndike continued his work and identified three functions of drive. He stated that drive energizes behavior, orients behavior toward a goal or reward, and selects behavior by leading toward the valence; in other words, it causes learning. Additionally, Thorndike (1911) identified the concept of a "satisfying state of affairs" (p. 280), which 51 describes a sought-after reward or reinforcement from an objective point of view. This has frequently been alluded to as an incentive, because the anticipation of it leads to making an appropriate response. Furthermore, animal drives and human motivations were compared, and it was widely accepted by behaviorists that human and animal behaviors have similarities. Thorndike assumed that whatever principles were found to govern the behavior of lower animals would apply to humans. The discoveries from these two studies have been the building blocks for subsequent human-motivation theories. It must be remembered that Thorndike's work has never been really questioned (Melton, 1941) and that his observations are considered accurate accounts of current beliefs regarding impulse and reward behavior today. Hull's Model Most psychologists, following Thorndike's work, spent their time attempting to understand how drive facilitates learning. Thorndike's work described the drive process in a completely objective way without recourse to terms suggesting subjective states of mindlike feelings, hopes, fears, and purpose, as did most of the psychologists who followed. Hull (1943) furthered this work by developing an objective model which eliminated subjective states. He thought it best to think of organism, animal and human, as a truly self-maintaining robot. With this model, Hull could avoid giving the robot human subjective qualities. The robot could move around freely as long as it had enough fuel. To ensure that there was enough fuel for movement, the robot was equipped with a fuel indicator which would activate a pump to collect fuel. Hull (1943) imagined this robot moving throughout the work where fuel was located in various locations. He reasoned that the robot world would come across some of these fuels by chance. When the robot did so, the pump would begin withdrawing the fuel from the pool 52 into the fuel tank, the indicator would rise and would turn off the pump, causing increased activity. Because of the experience, Hull additionally equipped the robot with a memory device which would lead the robot to fuel more quickly whenever the fuel supply ran low, similar to the concept developed by Thorndike. In short, Hull (1943) was able to translate Thorndike's drive theory into mechanical terminology through a motivational sequence. Drive reduction automatically reinforces the connection between the stimuli and responses in the sequence, so that the next time the robot would receive an internal cue, it associated low fuel with the site of the fuel pool. As a result of the cue, the responses of fueling occur promptly and efficiently. An example would be when an individual is thirsty and sees a drinking fountain. A connection is made in one's mind that getting a drink from the drinking fountain promptly fulfills the desire to overcome their thirst. Therefore, a drive stimulus impels action, and any situation could become a drive if it is made strong enough. The stronger the stimulus, the more drive exists. While every stimulus may become strong enough to act as a drive, certain special classes of stimuli seem to be the primary basis for the greater portion of motivation, while other stimuli with weaker drives are secondary. Rewards Rewards are defined as anything that reduces drive (McClelland, 1987) such as food, water, and sex. Behaviorists recognize that virtually any item has reward value, even though the drive or drives are not evident. According to Hullian theory, objects could even acquire reward values by being associated with primary-drive reductions (Hull, 1943). These are called secondary or acquired rewards, to contrast them with primary rewards like food. For example, one acquires money to pay for food, so the acquisition of money is a secondary 53 reward (Miller & Dollard, 1941). Secondary rewards can also be perceived from a situation they believe they were rewarded, such as the value one places on a lucky charm. For instance, a person may have been involved in an extremely dangerous situation while wearing a particular charm. Somehow they avoided the danger and believed the avoidance of danger to be associated with wearing the charm; thus the person wears the charm on any threatening occasion to reduce anxiety. Rewards can be primary or secondary, but just because some rewards are secondary does not mean there is a weak drive (Miller & Dollard, 1941). Many secondary rewards are labeled secondary because we do not necessarily need to have them, but something within an organism determines that it is desired. For instance, a cat loves catnip, but is the enjoyment of catnip a primary reward? The answer is no, because if a cat is enjoying catnip and becomes hungry, the cat will quit playing with the catnip and fulfill its primary desire to eat, thus categorizing catnip as a secondary reward or incentive rather than a primary one. Travel can be described as a secondary reward, but the drive associated with the activity could be strong for a person. Furthermore, depending on how controlling the individual is, the purchasing process could be a crucial component to the travel purchaser when making a final purchasing decision. The feeling that they are in control of the transaction and the confidence associated with the purchasing act may be a reward in itself. Thus, a better awareness of travel behavior is necessary to guide this research toward understanding the relationship between the travel consumer and the travel purchasing reward. Travel Theories and Models The literature associated with consumer behavior and the travel industry is centered on the old-style consumer, with little attention paid toward the new-style consumer (Lewis & Bridger, 2000). Travel consumers have had a difficult time establishing quality markets. One 54 of the main problems in the early travel industry, especially from a travel agent's perspective, was the lack of trust between airline industries and retail travel providers. This problem was only solved when airlines began to realize the importance of retail travel providers because they could reach more potential travel buyers (Jones, 2000). In an attempt to improve upon such an advancement, many researchers in the travel and tourism industry have spent a vast amount of time developing theories that explain and group tourist travel motivations and behaviors in order to assist the industry with future growth. Mill and Morrison (1992) described travel as a method to satisfy one's needs, and they linked Maslow's (1943) hierarchy of human needs with travel motivation. They chose to describe tourism as a system that consists of four different parts: the market, travel, destination, and marketing. The traveler can be understood by a market segment of the system and the decision to travel is made if the individual has learned that travel will satisfy his needs. Once the person decides to travel he must decide upon his travel preferences. The second segment describes and analyzes the choices. The destination mix consists of the attractions and services provided for the traveler. The destination mix is dependent on how well they satisfy the traveler, and from that satisfaction the traveler formulates a feeling toward the destination. It is the formulation of these feelings that enables the marketer to further develop marketing techniques to encourage people to travel to their destinations. Similarly, Dann (1977) concluded that travel motivation was also related to Maslow's hierarchy of human needs and argued that there are two factors in the travel decision-making process, consisting of push and pull factors. The push factors were those that made a person want to travel; pull factors were those that influenced where to travel. Dann (1977) proposed seven categories of travel motivation: (a) travel as a response to what is lacking, yet desired: People are part of an anomic society which fosters a need for social interaction that is 55 missing from their everyday home life; (b) destinations pull in response to motivational push; c) motivation as a purpose (i.e., visiting friends and family); (d) motivation as a fantasy; (e) motivational typologies or categories; (f) motivation and tourism experiences; and (g) travelers defining their situations will provide greater understanding of travel motivation than observing their behavior. Plog (1974) developed classifications and categorized travelers into three dimensions: allocentric, midcentric, and psychocentric. According to Plog, travelers must be classified along these dimensions. Travelers who are more allocentric are thought to prefer exotic destinations while psychocentrics are thought to prefer familiar destinations. Lastly, Plog (1974) found that the midcentric was somewhere in the middle of the two personality types. His typology of tourism has been the springboard for many researchers attempting to develop other travel typologies (Cohen, 1974; Gallup Organization, 1989; Lowyck, Van Langenhove, & Bollaert, 1992), as well as developing theories and/or models pertaining to the travel decision-making process (Howard & Sheth, 1969; Mathieson & Wall, 1982; Mayo & Jarvis, 1981; Middleton, 1988; Moutinho, 1987; Nicosia, 1966; Schmoll, 1977). The researchers did a reasonable job explaining the travel-buyer behavior process, but as seen in these previous models, the purchasing component of the traveler's travel-selection process is notably absent. Until this decade, very little attention has been given to the actual purchase of travel products and services during the purchase process. Basically, if a person wanted to purchase a travel product or service, he had to do it in one of two ways: directly through an airline by way of the telephone or through a travel agent in a face-to-face encounter. However, during the last decade this has all changed with the development of the new Internet consumer. As previously indicated, few consumer-behavior theories in the area of travel and tourism mention the purchasing step in their consumer-buying models (Howard & Sheth, 56 1969; Mathieson & Wall, 1982; Middleton, 2001). Others refer to purchase as ownership (Andreason, 1965) and/or decision (Nicosia, 1966). It seems as though many have overlooked the entire purchasing transaction as an important component of the consumerbehavior process. Prior to the advent of the Internet, consumers had little choice in their travel-purchase selection. The dominance of the Internet and online travel purchasing requires greater understanding of the purchasing procedure and consumer desires. Factors such as the amount of time an individual has available to develop travel itineraries; differing types of purchasing methods available to the consumer (credit card, cash, or personal check); safety of the purchase transaction; individual selection of travel routes; and the travel agent/purchaser relationships are more important than ever ("On-line Travel Trends," 2001). By looking closely at travel purchasers, travel researchers may gather important information regarding new purchasing markets. These travel-purchasing markets may provide insights about the types of people who utilize different Internet travelpurchasing techniques and assist future travel marketers when attempting to market to potential travelers. Marketers must look for every advantage possible with the rise of competition and assess greater amounts of accurate marketing information. Every company that sells travel has a database full of information regarding those who have purchased products from them. The travel companies who have appropriately used these databases have developed their own client profiles and preferences to a more in depth understanding of their consumers. They have used this information to help determine market segments, position marketing promotions, and target advertisements to possible buyers. A traveler's purchasing decisions over what to buy and why are based on core constructs that can be found in their personalities (Lewis & Bridger, 2000) as well as in everyday life's routines and experiences. 57 Market Segmentation Mill and Morrison (1992) explained that market segmentation is recognized and globally accepted as a way of examining tourism markets. Crossely and Jamison (1993) stated that the market segments are made up of people with common characteristics that influence purchasing behavior, in particular demographic, psychographic, and behavioral characteristics. Thus, market segmentation is the process of grouping together individuals with similar behaviors and characteristics (Mill & Morrison, 1992) for the purpose of communicating with them as consumers, commonly referred to as target marketing. Conclusion With recent developments in travel-purchasing procedures, it is hypothesized that a relationship exists between travel-purchasing behaviors and the level of control that a consumer desires throughout the purchasing process. The level of control that one desires is directly linked to the method in which a person prefers to purchase travel through the Internet. There is little research in the travel and tourism literature that examines this hypothesis, but the literature from other disciplines indicates that a relationship exists. Therefore it makes sense to examine this relationship further and determine the markets associated with the two purchasing methods. Additionally, several studies examining an individual's control over a process have utilized and developed context-specific measures of desire for control. Specific desire-forcontrol scales were developed regarding healthcare (Anderson, DeVellis, Boyles, & Feussner, 1989; Smith, Wallston, Forsberg, & King, 1984) and dental procedures (Logan, Baron, Keeley, Law, & Stein, 1991). In each of these studies, the desire-for-control scales revealed moderate to high levels of reliability and validity. Therefore, a desire-for-control method can be developed to sufficiently measure the context-specific area of travel purchase. 58 Furthermore, the other measures may provide some insight to other control issues as well as defining possible marketing segments that could assist the future of the travel agent. Internet travel agents must take a good look at their business operations and decide if it is still important to provide full service to a potential traveler, or whether the only way they can succeed as a profession is to work in high-volume markets such as those of the Internet travel supplier (Schaal, 2001). The desire-for-control construct is different from both the Type A behavior and locus of control constructs. A person's desire for control can change, usually because of past experience (Burger, 1992; Wright et al., 1990). Most of those who have a high desire for control feel comfortable with the selection process of their travel, but this feeling of comfort can change as a result of several factors. Several of those factors have been addressed such as one's familiarity with a particular destination. A person traveling abroad may feel more confident utilizing a travel agent because he has never been to a specific location before, an African safari, for instance. Another factor influencing desire for control may be the issue of travel safety. Has there recently been a tragedy that may affect an individual's perception of safety? There are many internal and external factors that influence the level of control one desires. It is expected that a person's desire for control influences several aspects of the travel-purchasing selection. It is evident that very little research has been completed in the areas of point of purchase sale, Internet travel-purchasing marketing segments, and control within the recreation and travel industry. The purpose of this study is twofold: (a) to examine purchasing behaviors and determine how control impacts the travel-purchasing process, and (b) to determine what markets exist for both the internal and external factors that affect a purchaser's desire for control of his life. Different profiles of those who purchase travel 59 based on the consumer control, internal and external circumstances, and other personal preferences will influence how a person purchases travel over the Internet. Therefore, the purpose of this study is to determine which group, those who purchase travel through an Internet travel agent or through an Internet travel supplier, possesses the greater amount of desire for control. Hypothesis Individuals with a higher desire for control are more likely to purchase online travel through an Internet travel supplier and individuals with a lower desire for control are more likely to purchase travel through an Internet travel agent when controlling for the effect of market-segmentation variables. 60 Table 1 Annual Percentages of Southwest Airlines Ticketless Travel and Online Purchases Year Ticketless Travel 1997 60% 1998 70% 1999 79% 2000 80% Online Purchases 31% Data Source: Southwest Airlines (1997, 1998, 1999, 2000) Table 2 Annual Percentages of Continental Airlines E-Ticket Purchases Year Electronic Ticket Purchases 1999 47% 2000 54.1% 2001 59.9% Data Source: Continental Airlines (1999, 2000, 2001) 61 Individual Definition of Success Definitional Matching Occupational Definitions of Success Individual's Definition of Event Success Perceived Success Matching Organizational Definition of Event as Success Individual's Search of Information to Explain Success Individual Makes an Attribution for the Causes of Success Informational Matching Attributional Matching Organizational Search of Information to Explain Success Organization Makes an Attribution for the Causes of Success Figure 1 An Attributional Process Model for Success in the Marketing Process. Outcomes -Job Satisfaction -Future Expectations for Success Outcomes -Increased Salary -Promotions 62 Individual Defines Travel Services to Purchaser Purchaser Behavior Individual's Search for Information on Travel Products and Services Independent Purchase of Travel Individual Ma kes Attributions for the Ca usality Tra vel Products a nd Services purchased Definitional Matching Perceived Purchasing Behavior Matching Informational Matching Time of Purcha se Ava ila bility Attributional Matching Marketers define travel products and services for purchase Marketers target markets according to purchase behavior Marketers provide search Information Availability of travel purchase Ma rketers ma ke a ttributions for the ca usality of tra vel products a nd services Figure 2 The Purchasing Process. Outcomes -Expectations -Satisfaction Outcomes -Increased purchases -More Promotions CHAPTER 3 METHOD Study Setting The intention of this study was to determine an individual's Internet travel-purchasing selection in relation to individual consumer preferences based on seven measures. Specifically, the measures consisted of: 1. Desire for control 2. Income 3. Time to purchase 4. Education level of purchaser 5. Computer accessibility 6. Length of stay 7. Cost of purchase It was anticipated that combinations of these seven variables would accurately predict the probability of an Internet travel purchaser selecting to purchase travel through an Internet travel agent or Internet travel supplier. Research Participants The study examined individuals who utilized the Internet to purchase travel, primarily Internet travel agencies and Internet travel suppliers. The individuals selected to participate in the study were drawn from an Internet travel company's e-mail distribution list 64 representing individuals who have purchased travel from the organization. The agency provided both an Internet travel agent and Internet travel supplier with the e-mail list consisting of 3,000 past purchasers. Sampling Procedure The study utilized simple random sampling from the individual travel organization providing both Internet travel agency and Internet travel-supplier services, previously defined in the literature review. The study accessed to two separate e-mail databases, each containing 3,102 electronic addresses of past travel purchasers. A random starting point was selected within each population and every third address thereafter was selected to participate in the study. This resulted in 1,034 individuals receiving an electronic invitation to access the questionnaire. Questionnaire Design The study employed combinations of measurements to assist in predicting a person's probable Internet purchasing method: (a) 17 questions regarding demographics and purchasing behaviors, and (b) 20 questions regarding desire for control (Burger, 1992) (see Appendix). The demographic section included gender, income, and level of education as well as other questions related to the previously mentioned variables. The answer to these questions ranged from yes and no responses to open-ended responses and Likert-scale response scoring. Because control is a personality measurement, the construct is somewhat complex. Therefore, the desire-for-control scale was utilized to measure individual control. Participants were asked to rate their personal desire for control according to a 7-point Likert scale, where number 1 represented "statement does not apply to me at all" to number 7 which represents "the statement always applies to me." 65 The 20-statement single-dimension desire-for-control scale (DC scale) was developed by Burger in 1984, but its foundational basis was established by Nunnally in 1967. Since 1984, Burger further tested the scale (Burger, 1984, 1985, 1986, 1987, 1989), as have others who maintained the same DC scale or developed new scales similar to the DC scale (Anderson et al., 1989; Logan et al., 1991; Smith et al., 1984). Example items on the questionnaire would read, "When it comes to orders, I would rather give them than receive them" or "I try to avoid situations where someone tells me what to do." Reliability For this scale, reliability was evaluated by two questions: First, were all test items similarly distributed or appear to measure the same construct? Second, did the scale scores remain consistent over repeated applications? Historically, the DC scale has indicated that it is reliable in measuring the same general construct control. Originally, Burger (1992) selected the 20 items for the DC questionnaire because they generated a high internal consistency with the sample but might not for another sample. In 1992, Burger tested the DC scale on another sample and computed a Kuder-Richardson coefficient of .81 on the final 20 item personality scale. In addition, several investigators have reported good internal consistency coefficients, which provided support regarding the reliability of Burger's instrument. For example, Smith, Woodward, Wallston, Rye, and Zylstra (1988), who administered the DC scale to various samples of adults, reported alpha reliability coefficients of .74, .81, and .78, respectively. Ryland and Levy (1988) administered the DC scale to college employees (Burger, 1992). They reported an alpha coefficient of .78 for the scale. Furthermore, in 1989, Reed reported a Cronba ch's alpha coefficient of .77 with a sample of union workers. Additionally, the DC scale's consistency of scale scores has been calculated and 66 determined a person's control stable over time. Unfortunately, because researchers are rarely interested in administering the same personality scale to a person more than once, information about the stability of desire over time is limited. However, in the original desirefor-control study (Burger & Cooper, 1979), a test-retest correlation of .75 was calculated, and Braukmann (1981) calculated a correlation of .70 between the two test scores (Burger, 1992). Therefore, there is some evidence that would indicate that the DC scale was reliable over time. Data Collection Procedure A travel company utilizing both methods of purchasing travel was contacted and permission granted for the study enabling access to individuals who recently purchased travel from the organizations. The company was selected based upon their prominence in the travel industry and the ability to provide an adequate database from which to draw e-mail addresses. This resulted in 1,334 participants being invited to participate in the study. Upon receiving an e-mail invitation, the participants accessed or linked to the questionnaire Website. The respondents who answered all questions were utilized for hypothesis testing. All responses were submitted to a Common Gateway Interface (CGI) made available by the University of Utah. The information was saved in numeric form and downloaded into a Microsoft Access database. Confidentiality was maintained at all times through several Javascript commands, which inhibited any information from being sent to the server unless authorized or designated to the account. The information was then transferred into a Statistical Package for Social Sciences (SPSS) version 9.0 for Windows for statistical analysis. 67 Data Analysis Data from the questionnaire were submitted to a Microsoft Access file, where the responses were downloaded to an SPSS database. Once the information was downloaded, frequency scores were calculated on all demographic, behavioral, and control questions. The descriptive statistic scores were utilized to identify possible profiles that may be revealed within the data set and select categories such as gender, education, and income. The analysis of desire-for-control questions was calculated in a fashion similar to what Burger (1992) developed. In order to determine the level of desire for control the values of questions 7, 10, 16, 19, and 20 where reversed in the DC scale, thus recoding the variable prior to analysis. Next, all the original answers to the 15 un-reversed questions and 5 recoded answers were summed together. Frequency and mean scores were calculated to determine the participants' average desire for control. Mean scores from various samples tend to identify mean scores around 100 with standard deviations of about 10. According to Burger (1992), respondents scoring above 100 are considered to possess a high desire for control and those below 100 would be considered to exhibit low desire for control. For hypothesis testing, logistical regression was used to determine if the desire for control variable was a predictor of online purchasing selection. In addition, the statistic utilized illustrated other competing variables that covaried with desire for control, thus developing a predictive set of variables based on desire for control. CHAPTER 4 RESULTS The following chapter will present a summary of the results of the analysis including a profile of the respondents, the desire-for-control related variables, and the results of the hypothesis testing. An interpretation of these results will be reported in the fifth chapter. Population Profile The e-mail addresses for the sample were verified for accuracy and those addresses in error or no longer functioning were eliminated by the travel agency. A systematic random sample (every third name) of the refined e-mail lists resulted in 3,102 names to be contacted. Electronic invitations were delivered to 1,034 current addresses. Four hundred individuals responded to the questionnaire, resulting in a 38.6% response rate. The demographic related questions of the instrument revealed that a majority of respondents were female (57.1%), had completed a college education (55.6%), and earned $30,000 or more annually (83.7%) (see Table 3). Ninety-eight percent of the respondents reported prior purchase of pleasure travel via the Internet and 86.8% of the respondents reported prior purchase of business travel via the Internet. When making a travel-purchase selection over the Internet, 77.8% of the respondents felt that price was extremely important to very important in making a purchasing decision. Respondents also indicated that the amount of control over the actual purchasing decision was extremely important to very important (78.8%). Length of time searching for travel purchases over the Internet was mixed (see Table 4), while 50.1% of travel purchasers 69 indicated that they utilized an Internet travel agency to purchase travel 15% of the time or less. Population Computer Purchasing Behaviors Seventy-four percent of respondents spent between 30 and 180 minutes surfing the Internet for general information each day. These individuals spent the greatest amount of time searching the Internet at their leisure (see Table 5) and have had access to a computer for an average of 2 years. When accessing a computer, most people utilized a computer from their home (91.3%) or work (74%). Computer use specifically related to travel information was examined in addition to general computer behaviors. Ninety-seven percent of the respondents reported researching travel over the Internet and 85% reported purchasing travel over the Internet. The respondents indicated control over their travel purchase as the most important factor influencing their decision to making a travel-purchase decision via the Internet (see Table 6). The second most frequently cited reason for making a travel-purchase decision over the Internet was price. Most respondents who purchased travel reported making their own travel arrangements (88.3%) while 10% of the respondents reported making arrangements through a travel agency. When making a travel selection, 42% of the respondents preferred to utilize an Internet search engine, such as Travelocity, whereas 12% of the respondents indicated sole dependence on the Internet travel agency Web sites, such as Morris Murdock Travel. Furthermore, 16.6% of the respondents utilized an Internet travel-agent site for travel purchases 75% of the time, as opposed to 26% of respondents, who reported never utilizing an Internet travel agent. 70 Hypothesis Testing Individuals with a higher desire for control will be more likely to purchase online travel through an Internet travel supplier and individuals with a lower desire for control will be more likely to purchase travel through an Internet travel agent when controlling for the effect of market-segmentation variables. Respondents were asked to complete the desire-for-control scale developed by Burger (1992) representing high and low desire for control. In addition, several questions were designed to probe respondents' preference for travel purchase through the Internet, when controlling for the effect of market-segmentation variables. The respondents indicated the importance of control of the purchase transaction (x = 5.87), the cost of purchase (x = 5.87), and the expediency of the process (x = 4.73) when purchasing travel (see Table 6). Mean scores were calculated for the desire-of-control scale questions, including annual income, importance of time, respondents' educational level, computer accessibility (see Table 7), length of stay (see Table 8), and the cost of the travel purchase. Some significant observations regarding the results included the respondents' access of the Internet via computers at home (91%) and the use of Internet purchase of travel lasting 1 to 5 days (39.6%) and 6 to 10 days (49.1%). Logistic regression was utilized to determine a predictive set of variables to predict the respondents' preferred method of purchasing travel over the Internet. The variables were placed into the following order set with the sum of the desire-for-control scale as the first variable entered from the set. 1. Desire for control 2. Income 3. Time to purchase 71 4. Education level of purchaser 5. Computer accessibility 6. Length of stay 7. Cost of purchase P-values were calculated to determine any predictive variable sets (see Table 9) regarding respondents' preferred method of travel purchasing, but the data revealed no such predictive set of variables. Based on these results the model resulted with no significance and the hypothesis was rejected. However, a follow-up analysis of the sum of desire-for-control scale ordered set was replaced by an importance-of-control ordered set. The p-value revealed (see Table 10) that only the new control variable was significant at the .05 level (one-tailed). 72 Table 3 Internet Travel Purchasers Income (N = 398) Annual Income Frequency Percent $1 to $20,000 9 2.3 $20,000 to $29,999 15 3.8 $30,000 to $49,999 73 18.3 $50,000 and up 260 65.3 Prefer not to state 41 10.3 73 Table 4 Importance Regarding the Length of Time Searching the Internet for Travel (N = 393) Importance Frequency Percent Extremely Unimportant 25 6.3 Very Unimportant 22 5.6 Unimportant 46 11.7 Neutral 81 20.6 Important 84 21.4 Very Important 65 16.5 Extremely Important 70 17.8 Table 5 Hours Spent Surfing the Internet Daily Internet Access Mean Hours SD N Daily Surfing 1.42 2.02 316 Work Related 1.40 1.94 354 Leisure Related 12.22 5.72 389 74 Table 6 Important Travel-Purchasing Factors Travel Factors Mean SD N Control of Purchase 5.87 1.80 380 Cost of Purchase 5.85 1.88 385 Time it takes to Purchase 4.73 1.65 381 Note, Scale of 1 to 7 Table 7 Computer Location Access (N = 400) Computer Access Frequency % At Home 365 91.0% At School 41 10.3% At Library 48 12.0% At Friends 42 10.5% At Relatives 74 18.5% At Work 296 74.0% 75 Table 8 Vacation Length of Stay (N = 400) Length in Days Frequency % 1 to 5 154 39.6% 6 to 10 196 49.1% 11 to 15 27 7.0% 16 to 20 2 0.6% 21 to 25 2 0.6% 26 and above 1 0.3% 76 Table 9 Desire for Control Predictive Set Variable S.E df Sig. Exp(B) Desire for Control .017 1 .819 1.004 Income .164 1 .639 .937 Importance of Time .149 1 .651 .935 Education .105 1 .464 .926 Computer Accessibility .131 1 .810 .969 Vacation Length .080 1 .638 .963 Importance of Price .129 1 .962 1.006 Constant 2.119 1 .475 .220 *Significant at the .05 level (one-tailed). 77 Table 10 Importance of Control Predictive Set Variable S.E df Sig. Importance of Control .189 1 .029* .862 Income .159 1 .642 .929 Importance of Time .150 1 .922 1.015 Education .098 1 .852 .982 Computer Accessibility .123 1 .925 .989 Vacation Length .040 1 .938 .997 Importance of Price .192 1 .087 1.388 Constant 1.304 1 .180 .174 *Significant at the .05 level (one-tailed). Exp(B) CHAPTER 5 DISCUSSION Purpose of Study The purpose of this study was to examine which group, those who purchase travel through an Internet travel agent or those who purchase travel through an Internet travel supplier, possessed a greater amount of desire for control. This chapter will examine and discuss the results of the study and recommend possible approaches for future research. Prior to the data collection it was anticipated that individuals who purchase travel from an Internet travel agency were different from those who purchase travel through an Internet travel supplier. An analysis of the responses did not reveal a predictive variable(s) to support the hypothesis. However, evidence from the study does support the need for further research, which should be conducted in the area of Internet travel-purchasing behaviors and an individual's level of control over the purchasing decision. Internet Travel Purchaser Profile Prior to this research, relatively little information was known regarding the profiles of individuals possessing a preference for purchasing travel from an Internet ravel supplier or Internet travel agency. Travel-related and popular news literature supported the proposition that Internet travel buyers significantly differ on a variety of variables (Anderson, 2000; Jones, 2000; Middleton, 2001). However, the results of this study indicated that purchasers 79 were similar on several personal characteristics revealing little to no significant differences between the two types' purchasing techniques. E-mail invitations were sent out to two groups of purchasers, those who historically utilized an Internet travel agent for travel purchases and those who purchased travel from an Internet travel site on their own. From these two groups most of the respondents were directly responsible for establishing their own travel itinerary (91.7%) through travel suppliers, with only 10% of the respondents using an Internet travel agent for travel purchases. Surprisingly, half of those who received invitations were believed to purchase travel only through an Internet travel agent. This is evidence of the trend over the last few years of computer usage and Internet growth. Further, it demonstrates that people are becoming increasingly more comfortable with the Internet and their own computer skills, thus providing support to the hypothesis that more people are purchasing their own individual Internet travel because of their improved computer skills resulting from user experience (Jones, 2000). Furthermore, in examining the distinctions between the two groups, of the 48 respondents who solely utilized an Internet travel agent to purchase travel, only 6 respondents scored below 100 on the desire-for-control scale, thus suggesting that desire for control is not the variable that explains the difference that might exist between the two types of purchasers. In examining demographic information about the two groups, annual income levels and education levels were evenly distributed between individuals who purchase travel through an Internet travel agent and Internet travel supplier. Furthermore, the results revealed that 50.1% of the respondents did utilize an Internet travel agent for at least some purchase. The evidence indicates that the respondents from this study seem to use both methods of 80 purchasing travel over the Internet. Hypothesis Testing Hypothesis testing utilized the enter method of logistic regression to calculate whether any significant predictive groups existed that might explain why a person purchased through a specific Internet purchasing method. Seven variables were placed into the algorithm in the following order: 1. Desire for control 2. Income 3. Time to purchase 4. Education level of purchaser 5. Computer accessibility 6. Length of stay 7. Cost of purchase From this set of variables, the statistical analysis failed to reveal any significant predictive set. Therefore, the proposed model was not effective in revealing a set of variables that could be utilized to predict if a buyer would use an Internet travel agent or Internet travel supplier. However, another question was included to better understand the importance an Internet travel purchaser placed on the amount of control they have over the purchasing decision. The results from this question were substituted for the desire for control results within the model. The logistic regression analysis of this revealed a significant relationship between the method of Internet travel purchasing and importance of control, but failed to indicate any predictive groups from the remaining marketing variables, thus providing some support that control may play a role with Internet travel purchasing, while discounting an individual's desire for control. 81 Future Research The desire-for-control scale may not have been the most appropriate instrument to have used when attempting to develop a predictive set of variables when purchasing travel over the Internet. Therefore, research should examine other measurement tools and the comparison of other control scales or possibly develop an importance-of-control scale specifically designed to examine Internet travel-purchasing behaviors. Conclusion In closing, desire for control had never been utilized to examine the area of Internet travel purchasing and was anticipated that it would assist in providing a better understanding of the rapid adoption of Internet travel-purchasing behaviors. This study pioneered the first exploration of the topic and has provided an alternative method of examining a relatively unfamiliar variable's influence on Internet purchasing decisions. It is apparent that control plays some role in Internet purchasing decisions and provides a possible reason why so many people have removed themselves from utilizing a brick and mortar travel agency for travel purchasing. It seems as though the study observed an alternative level of control that has never been examined prior to this research. The importance-of-control variable does not resemble characteristics of control as described during the literature review regarding desire for control, locus of control, Type A behavior, or self-efficacy. The study demonstrated that desire for control might not be a significant variable in determining Internet purchasing decisions. Therefore, marketers may not be required to focus on desire for control when developing portals or Web pages for travel purchasing. However, importance of control might explain how the purchaser develops an internal perception of 82 why he purchased a specific travel product or service. Further research is needed to understand the rapid adoption of Internet travel-purchasing behaviors in order to reveal a means for Internet travel agencies and Internet travel suppliers to market and attract clients for their services. APPENDIX THE INTER-RALSTON TRAVEL PREFERENCE QUESTIONNAIRE 84 The Inter-Ralston Travel Preference Questionnaire Thank you for participating in the Inter/Ralston Travel Preference Questionnaire. This Questionnaire was designed by Frederick J. Inter for the completion of his Doctorial Degree for the University of Utah's Parks, Recreation, and Tourism Department. The information gathered in this Questionnaire will be completely confidential and used for research purposes. 85 TRAVEL PURCHASING BEHAVIOR SCALE Section 1. Demographic and Other Information Please read each question carefully and respond appropriately _____ male _____ female (check one) How many years of school have you completed? (Check the one that applies.) High School 1 2 3 4 5 6 7 College 8 9 Graduate School 10 11 12 What is your estimated household Income? ○ Less than $20,000 ○ $20,000 to $29,999 ○ $30,000 to $49,999 ○ $50,000 to $74,999 ○ $75,000 to $99,999 ○ $100,000 or more ○ I prefer not to state How many travel purchases for pleasure do you make annually? Do you research pleasure travel over the Internet? Yes_____ No_____ Do you purchase pleasure travel over the Internet? Yes_____ No_____ Do you purchase travel over the Internet? Yes_____ No_____ On a typical day, how many hours do you spend surfing the Internet? At work __________ hours For leisure __________ hours How many years have you had access to a computer? __________ years How many computers are in your household? Where do you access a computer and the Internet? (Check all that apply.) At home _____ At School _____ At Library _____ At a friends _____ At a relative's _____ At work _____ 86 Other _______________ How important is price when making your pleasure travel purchasing decision (On a scale of 1 = Extremely Unimportant to 7 = Extremely Important) ○1○2○3○4○5○6○7 How important is the length of time you spend search for travel over the Internet (On a scale of 1 = Extremely Unimportant to 7 = Extremely Important) ○1○2○3○4○5○6○7 How important is the amount of control you have over the actual purchasing decision for your travel purchase (On a scale of 1 = Extremely Unimportant to 7 = Extremely Important) ○1○2○3○4○5○6○7 How many days in length is your average pleasure vacation? Do you usually make your own travel arrangements? Yes_____ No_____ If no, check who does. (Check all that apply) Spouse _____ Secretary/staff _____ Friend _____ Travel agent _____ Family member other than spouse _____ Other _______________ When purchasing travel over the Internet I prefer to use the following method of purchase (Please select one). ○ Internet Travel Supplier (such as an airline or corporate hotel website) ○ Travel Agency (such as Morris Murdock Travel) ○ Internet Travel Search Engine (such as Expedia or Travelocity) ○ Internet Travel Destination (such as Disney World or Las Vegas website) ○ I do not purchase travel over the Internet When purchasing travel via the Internet, what percentage of your pleasure travel is purchased through an Internet travel agent? (Indicate a point on the scale that most applies to you.) 0 10 5 20 15 30 25 40 35 50 45 60 55 70 65 80 75 90 85 100 95 87 Section 2. "Desire for Control" Scale1 Please read each question carefully and respond to it by expressing the extent to which you believe the statement applies to you. For all items, a response from 1 to 7 is required. Use the number that best reflects your belief when the scale is defined as follows: 1 = The statement does not apply to me at all 2 = The statement usually does not apply to me 3 = Most often, the statement does not apply to me 4 = I am unsure whether or not the statement applies to me, or it applies to me about half of the time 5 = The statement applies more often than not 6 = The statement usually applies to me 7 = The statement always applies to me ____1. I prefer a job where I have a lot of control over what I do and when I do it. ____2. I enjoy political participation because I want to have as much of a say in running government as possible. ____3. I try to avoid situations where someone else tells me what to do. ____4. I would prefer to be a leader than a follower. ____5. I enjoy being able to influence the actions of others. ____6. I am careful to check with everything on an automobile before I leave for a long trip. ____7. Others usually know what is best for me. ____8. I enjoy making my own decisions. ____9. I enjoy having control over my own destiny. ____10. I would rather someone else take over the leadership role when I'm involved in a group project. ____11. I consider myself to be generally more capable of handling situations than others are. ____12. I'd rather run my own business and make my own mistakes than listen to someone else's orders. ____13. I like to get a good idea of what a job is all about before I begin. ____14. When I see a problem, I prefer to do something about it rather than sit by and let it continue. ____15. When it comes to orders, I would rather give them than receive then. ____16. I wish I could push many of life's daily decisions off on someone else. ____17. When driving, I try to avoid putting myself in a situations where I could be hurt by 88 another person's mistake. ____18. I prefer to avoid situations where someone else has to tell me what it is I should be doing. ____19. There are many situations in which I would prefer only one choice rather than having to make a decision. ____ 20. I like to wait and see if someone else is going to solve a problem so that I don't have to be bothered with it. 1 Source: Burger (1992). Reprinted with permission of Plenum Press © 1992. REFERENCES Abelson, R., & Levi, A. (1985). Handbook of social psychology. New York, NY: Random House. Abramson, L., Seligman, M., & Teasedale, J. (1978). Learned helplessness in humans: Critique and reformulation. Journal of Abnormal Psychology, 87(1), 49-74. Amarante, K. (2002, December 30). One in five to book online by 2005. Retrieved February 7, 2003, from http://www.clia.org/news_online-reservations.cfm Anderson, K. (2000). Modern conveniences. Retrieved October 2, 2000, from http://findarticles.com/cf)0m0VOU/9_300/64996574/p2/article.jhtml?term=travel+ag ent+history Anderson, L., DeVellis, R., Boyles, B., & Feussner, J. (1989). Patients' perceptions of their clinical interactions: Development of the multidimensional desire for control scales. Health Education Research, 4(3), 383-397. Andreason, A. (1965). Attitudes and behaviors: A decision model in new research marketing. Berkeley, CA: Institute of Business and Economic Research, University of California. Angleitner, A., & Wiggins, J. (1986). Personality assessment via questionnaires: Current issues in theory and measurement. Berlin, Germany: Springer-Verlag. Antaki, C., & Brewin, C. (1982). Attributions and psychological change: Applications of attribution theories to clinical and educational practice. London, England: Academic Press. Averill, J. (1973). Personal control over aversive stimuli and its relationship to stress. Psychology Bulletin, 80(4), 286-303. Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change. Psychology Review, 84(2), 191-215. Bandura, A. (1986). Social foundations for thought and action: A social cognitive theory. Englewood Cliffs, NJ: Prentice-Hall. Bennett, M. (1993). Information technology and travel agency: A customer service perspective. Tourism Management, 14(2), 259-266. 90 Bentler, P. (1995). EQS Structural Equations program manual. Encino, CA: Multivariate Software, Inc. Bettelheim, B. (1943). Individual and mass behavior in extreme situations. Journal of Abnormal and Social Psychology, 38, 417-451. Bonn, M., Furr, L., & Susskind, A. (1999). Predicting a behavioral profile for pleasure travelers on the basis of Internet use segmentations. Journal of Travel Research, 37(3), 333-340. Bruner, R. (1998). Net results. Indianapolis, IN: Hayden. Burger, J. (1984). Desire for control, locus of control, and proneness to depression. Journal of Personality, 52(1), 76-89. Burger, J. (1985). Desire for control and achievement-related behaviors. Journal of Personality and Social Psychology, 48(6), 1520-1533. Burger, J. (1986). Desire for control and the illusion of control: The effects of familiarity and sequence of outcomes. Journal of Research in Personality, 20(1), 66-76. Burger, J. (1987). Desire for control and conformity to a norm. Journal of Personality and Social Psychology, 53(2), 355-360. Burger, J. (1989). Negative reactions to increases in perceived personal control. Journal of Personality and Social Psychology, 56(2), 246-256. Burger, J. (1992). Desire for control: Personality, social, and clinical perspectives. New York, NY: Plenum Press. Burger, J., & Cooper, H. (1979). The desirability of control. Motivation and Emotion, 3(3), 381-393. Carr, H. (1938). Part I of symposium on the law of effect. Psychological Review, 45, 191199. Cattell, R. (1971). Abilities: Their structure, growth, and action. New York, NY: Houghton Mifflin. Chervenak, Keane, & Co. (1988, January). CKC Report. The Hotel Technology Newsletter, 5, 3. Cohen, E. (1974). Toward a sociology of international tourism. Social Research, 39(1), 164182. Cohen, J. (1964). Behavior in uncertainty. London, England: Unwin. Conn, S., & Rieke, M. (1994). The 16PF fifth edition technical manual. Champaign, IL: Institute of Personality and Ability Training. 91 Connell, J. (1990). Context, self and action: A motivational analysis of self-esteem process across the lifespan. In D. Cicchetti & M. Beegly (Eds.), Continuities and discontinuities in development (pp. 153-173). New York, NY: Plenum. Connell, J., & Wellborne, J. (1991). Competence, autonomy and relatedness: A motivational analysis of self-system processes. In M. Gunnar & A. Sroufe (Eds.), Minnesota Symposium on Child Psychology (pp. 43-77). Chicago, IL: University of Chicago Press. Continental Airlines. (1998). Annual Report. Retrieved July 21, 2002, from http://www.continental.com Continental Airlines. (1999). Annual Report. Retrieved July 21, 2002, from http://www.continental.com Continental Airlines. (2000). Annual Report. Retrieved July 21, 2002, from http://www.continental.com Continental Airlines. (2001). Annual Report. Retrieved July 21, 2002, from http://www.continental.com Cooper, C. (1993). Progress in tourism, recreation, and hospitality management. New York, NY: Belhaven Press. Copeland, D., & McKenney, J. (1988). Airline reservations: Lessons from history. MIS Quarterly, 3, 353-370. Costa, P. T., Jr., & McRae, R. R. (1988). Personality in adulthood: A six-year longitudinal study of self-reports and spouse ratings on the NEO Personality Inventory. Journal of Personality and Social Psychology, 55(5), 853-863. Crossely, J., & Jamison, L. (1993). Introduction to commercial and entrepreneurial recreation. Champaign, IL: Sagamore. Dann, G. (1977). Anomie, ego-enhancement and tourism. Annals of Tourism Research, 4(1), 184-194. Del Rosso, L. (2001, October 1). Firm revises predictions for online bookings. TW Crossroads. Retrieved February 4, 2003, from http://www.twcrossroads.com/news/newsearchwrapper.asp?Article=7775 Delta Air Lines. (1997). Annual Report. Retrieved July 21, 2002, from http://www.delta.com/inside/investors/annual_report/archieve/index Delta Air Lines. (2001). Annual Report. Retrieved July 21, 2002, from http://www.delta.com/inside/investors/annual_report/archieve/index 92 Dorsey, J. (1999, May 17). Hotels and resorts. TW Crossroads. Retrieved January 26, 2002, from http://www. twcrossroads.com/news/newsearchwrapper.asp?Article=14614 Drucker, P. (1954). The practice of management. New York, NY: Harper. Drucker, P. (1994). Post capitalist society. London, England: Butterworth-Heinerman. Durbin, F. (2002, February 23). Travel technology. TW Crossroads. Retrieved January 26, 2002, from http://www. twcrossroads.com/news/newsearchwrapper.asp?Article=19610 Eagly, A., & Himmelfarb, S. (1978). Attitudes and opinions. Annual Review of Psychology, 29, 517-554. Elig, T., & Frieze, I. (1979). Measuring causal attributions for success and failure. Journal of Personality and Social Psychology, 37(4), 621-634. Erikson, E. (1974). Dimensions of new ideology. New York, NY: Norton. Eysenck, H., MacLeod, C., & Matthews, A. (1987). Cognitive functioning and anxiety. Psychological Research, 49(1), 189-195. Eysenck, H. J. (1967). The biological basis of personality. Springfield, IL: Thomas. Eysenck, H. J., & Eysenck, M. W. (1985). Personality and individual differences: A natural science approach. New York, NY: Plenum. Feather, N. (1969). Attribution of responsibility and valence of success and failure in relation to initial confidence and task performance. Journal of Personality and Social Psychology, 13(2), 129-144. Featherly, K. (2000, February 7). Say goodbye to virtual mom-and-pop travel shops. Retrieved February 4, 2003, from http://www.newsbytes.com Folkman, S. (1984). Personal control and stress and coping processes: A theoretical analysis. Journal of Personality and Social Psychology, 46(4), 839-852. Forrester Research. (1999, July 1). Yahoo!, Travelocity, and Expedia receive top ratings for online travel sites. Retrieved November 16, 2002, from http://www.findarticles.com/cf_0/m0EIN/1999_July1/55042167 Friedman, M., & Lackey, G. (1991). The psychology of human control. New York, NY: Praeger. Friedman, M., & Rosenman, R. (1974). Type A behavior and your heart. New York, NY: Knopf. Frieze, I., Bar Tal, D., & Carroll, J. (1979). Defining success in classroom settings. In J. 93 Levine & M. Wang (Eds.), New models, new extensions of attribution theory (pp. 126-162). New York, NY: Springer-Verlag. Frieze, I., Moss, M., & Olson, J. (1988). Perceptions of success in work environments. In S. Zelen (Ed.), New models, new extensions of attribution theory (pp. 126-162). New York, NY: Springer-Verlag. Gallup Organization. (1989). Unique four nation travel study reveals traveler types. London, England: American Express. Galton, F. (1886). Measure of character. Fortnightly Review, 36, 179-185. Gamble, P. (1993). Progress in tourism, recreation, and hospitality management. New York, NY: Belhaven Press. Glass, D. (1977). Behavior patterns, stress, and coronary disease. Hillsdale, NJ: Erlbaum. Glass, D., & Singer, J. (1972). Urban stress: Experiments on noise and social stressors. New York, NY: Academic Press. Godwin, N. (1987). Complete guide to travel agent automation. New York, NY: Delmar Publishers, Inc. Goldberg, L. (1993). The structure of phenotypic personality traits. American Psychologist, 48(1), 26-34. Gorsuch, R. (1983). Factor analysis. London, England: Erlbaum. Gough, H. (1987). California Personality Inventory administrator's guide. Palo Alto, CA: Consulting Psychologists. Harris, L. (1996). U.S. travel agency automation survey. Travel Weekly, 65(1), 118-134. Harter, S. (1978). Effectance, motivation reconsidered: Toward a developmental model. Human Development, 21(1), 36-64. Heider, F. (1944). Social perception and phenomenal causality. Psychological Review, 51(3), 358-384. Heider, F. (1958). The psychology of interpersonal relations. New York, NY: Wiley. Hewstone, M. (1983). Attribution theory. Oxford, England: Basil Blackwell. Hewstone, M. (1989). Causal attribution. Oxford, England: Basil Blackwell. Hogan, L. (1986). Hogan Personality Inventory manual. Minneapolis, MN: National Computer Systems. 94 Houston, B., & Snyder, C. (1988). Type A behavior pattern: Research, theory, and intervention. New York, NY: Wiley. Howard, J., & Sheth, J. (1969). Theory of buyer behavior. New York, NY: Wiley. Hull, C. (1943). Principles of behavior. New York, NY: Appleton-Century-Crofts. Jackson, M., White, G., & Schmierer, C. (1996). Tourism experiences within an attributional framework. Annals of Tourism Research, 23(3), 798-810. Jaspars, L. J., Hewstone, M., & Fincham, F. (1983). Attribution theory research: Conceptual, developmental and social dimensions. New York, NY: Academic Press. Jones, D. (2000, July 31). An on and down relationship. Retrieved November 16, 2002, from http://www.findarticles.com/cf_0/moEIN/2000_July31/64996575 Jones, E., & Davis, K. (1965). From acts to dispositions: The attributional process in person perception. In L. Berkowitz (Ed.), Advances in experimental social psychology, Vol. 2 (pp. 219-266). New York, NY: Academic Press. Jones, G. (2001, January 8). Online travel takes off, but needs loyalty. CNN Asia. Retrieved August 11, 2002, from http://asia.cnn.com/2001/tech/computing/01/08/travel.take.off.idg/ Joreskog, K. (1973). A general method for estimating a linear structural equation system. In A. Goldberg & O. Duncan (Eds.), Structural equation models in social sciences (pp. 85-112). New York, NY: Seminar Press. Kelley, H. (1967). Attribution theory in social psychology. In D. Levine (Ed.), Nebraska Symposium on Motivation (pp. 191-240). Lincoln, NE: University of Nebraska Press. Kelley, H. (1973). The process of causal attribution. American Psychologist, 28(1), 107-128. Kiesler, C. (1966). Control and number of choice alternatives. Psychology Reports, 18(2), 603-610. Klien, D., & Seligman, M. (1976). Reversal of performance deficits in learned helplessness and depression. Journal of Abnormal Psychology, 85(1), 11-26. Kline, P. (1993). The handbook of psychological testing. London, England: Routledge. Kline, P. (1994). An easy guide to factor analysis. London, England: Routledge. Kluckhohn, C. (1953). Personality in nature, society, and culture (2nd ed.). New York, NY: Alfred A. Knopf. Kotler, P., Bowen, J., & Makens, J. (1996). Marketing for hospitality and tourism. Upper Saddle River, NJ: Prentice Hall. 95 Langer, E. (1983). The psychology of control. Beverly Hills, CA: Sage. Langer, E., & Dweck, C. (1973). Personal politics: The psychology of making it. Englewood Cliffs, NJ: Prentice Hall. Langer, E., & Rodin, J. (1976). The effects of choice and enhanced personal responsibility for the aged: A field experiment in an institutional setting. Journal of Personality and Social Psychology, 34(2), 191-198. Langer, E., & Roth, J. (1975). Heads I win, tails it's chance: The illusion of control as a function of the sequence of outcomes in a purely chance task. Journal of Personality and Social Psychology 32(6), 951-955. Lassiter, E. (1998, June 30). Shifting to online sales. TW Crossroads. Retrieved January 26, 2002, from http://www.twcrossroads.com/news/newsearchwrapper.asp?ArticleID=7775 Lazar, T. (1999, January 14). The future of the travel industry. TW Crossroads. Retrieved February 4, 2003, from http://www.twcrossroads.com/news/newsearchwrapper.aspArticle?=12160 Levinson, D. J. (1986). A conception of adult development. American Psychologist, 41(1), 313. Lewis, D., & Bridger, D. (2000). The soul of the new consumer. London, England: Nicholas Brealey. Lodziak, C. (2002). The myth of consumerism. Sterling, VA: Pluto Press. Logan, H., Baron, R., Keeley, K., Law, A., & Stein, S. (1991). Desired control and felt control as mediators of stress in a dental setting. Health Psychology, 10(3), 352-359. Lowyck, E., Van Langerhove, L., & Bollaert, L. (1992). Typologies of tourist roles: Perspectives on tourism policy. London, England: Mansell. Luhrman, D. (2001, October 23). Internet poised to take a quarter of tourism sales. Retrieved January 26, 2002, from http://www.Worldtourism.org/newsroom/releases/morereleases/October2001/011023.htm Mandelbaum, R. (2002, August 3). Commissions in the hotel industry: Agents for change. Retrieved February 7, 2003, from http://www.hotelonline.com/Mews/PR2002_3rd/Aug02_TACommissions.html Mannell, R., & Kleiber, D. (1997). A social psychology of leisure. State College, PA: Venture. Marks, L. (1998). Deconstructing locus of control: Implications for practitioners. Journal of 96 Counseling and Development, 76(3), 251-260. Maslow, S. (1943). A theory of human motivation. Psychological Review, 50(2), 370-396. Mathieson, A., & Wall, G. (1982). Tourism: Economic, physical, and social impacts. Harlow, England: Longman. Matthews, G., & Deary, I. (1998). Personality traits. New York, NY: Cambridge University Publishing. Mayo, E., & Jarvis, L. (1981). The psychology of leisure travel. Boston, MA: CBI Publishing. McClelland, D. (1987). Human motivation. New York, NY: Cambridge Press. Melton, A. (1941). Encyclopedia of education research. New York, NY: Macmillan. Merriam-Webster. (2002). Webster's Third New International Dictionary. Retrieved May 3, 2002, from http://www.m-w.com/home.htm Middleton, V. (2001). Marketing in travel and tourism (3rd ed.). Boston, MA: ButterworthHeinerman. Mill, A., & Morrison, A. (1992). The tourism system: An introductory text. Englewood Cliffs, NJ: Prentice-Hall. Miller, N., & Dollard, J. (1941). Social learning and imitations. New Haven, CT: Yale University Press. Mischel, W. (1968). Personality and assessment. New York, NY: John Wiley & Sons. Modahl, M. (2000). Now or never: How companies must change today to win the battle for Internet consumers. New York, NY: Harper-Collins. Moghaddam, F., & Studer, C. (1998). Illusions of control. Westport, CT: Praeger. Moutinho, L. (1987). Consumer behavior in tourism. European Journal of Marketing, 21(10), 1-44. Nicosia, F. (1966). Consumer decision making process: Marketing and advertising implications. Englewood Cliffs, NJ: Prentice-Hall. Nozick, R. (1990). The normative theory of individual choice. New York, NY: Garland. Nunnally, J. (1967). Psychometric theory. New York, NY: McGraw-Hill. On-line travel trends. (2001, September 24). Travel Weekly. Retrieved February 4, 2003, from http://www.findarticles.com 97 Oppermann, M. (1999). Data-based marketing by travel agencies. Journal of Travel Research, 37(3), 231-237. Perlmuter, J., & Montey, R. (1979). Choice and perceived control. Hillsdale, NJ: Erlbaum. Pervin, L. (1994). A critical analysis of current trait theory. Psychological Inquiry, 5(2), 103113. Peters, G. (1996). Beyond the next wave: Imagining the next generation of customers. London, England: Pitman. Pizam, A., & Mansfield, Y. (1999). Consumer behavior in travel and tourism. New York, NY: The Haworth Hospitality Press. Plog, H. (1974). Why destination areas rise and fall in popularity. Cornell Hotel and Restaurant Quarterly, 14(4), 55-58. Reed, T. (1989). Do union organizers matter? Individual differences, comparing practices, and representation election. Industrial and Labor Relations Review, 43(1), 103-118. Reynolds, T., & Olson, J. (2001). Understanding the consumer decision making: The meansend approach to marketing and advertising strategy. Mahwah, NJ: Lawrence Erlbaum. Rodin, J. (1990). Control by any other name: Definitions, concepts, and processes. In J. Rodin, C. Schooler, & W. Shaie (Eds.), Self-directedness: Cause and effects throughout the process of life course (pp. 1-17). Hillsdale, NJ: Lawrence Erlbaum. Rotter, J. (1966). Generalized expectations for internal versus external control of reinforcement. Psychological Monographs, General and Applied, 80(1), 1-28. Sanfilippo, M. (2000, July 31). Coming full circle. Retrieved March 17, 2002, from http://www.findarticles.com/cf_0/m)VOU/9-300/64996573/p2/article.jhtml.?terrmtravel+agent+history Saville, P., Holdsworth, R., Nyfield, G., Cramp, L., & Mabey, W. (1984). The Occupational Personality Questionnaire (OPQ). London, England: Saville & Holdsworth. Schaal, D. (2001, August 27). Online booking soars in first half of year. TW Crossroads. Retrieved February 9, 2003, from http://www.twcrossroads.com/news/newsearchwrapper.asp?ArticleID=34353 Schaal, D. (2002, July 18). Travel technology. TW Crossroads. Retrieved January 26, 2002, from http://www.twcrossroads.com/news/newsearchwrapper.asp?ArticleID=27739 Schmoll, G. (1977). Tourism production. London, England: Tourism International Press. 98 Seligman, M. (1975). Helplessness: On depression and death. San Francisco, CA: Freeman Press. Seligman, M., & Maier, S. (1967). Failure to escape traumatic shock. Journal of Experimental Psychology, 74(1), 1-9. Shattuck, H. (2003, January 6). 2002: A year of changing demands. Retrieved February 7, 2003, from http://www.chron.com/sc/CDA/stroy.hts/travel/1714871 Sheldon, P. (1997). Tourism information technology. New York, NY: CAB International. Sjostrom, H., & Nilsson, R. (1972). Thalidomide and the power of the drug companies. London, England: Penguin Books. Skinner, E. (1995). Perceived control, motivation, and coping. Thousand Oaks, CA: Sage. Smith, R., Wallston, B., Forsberg, P., & King, J. (1984). Measuring desire for control of health care process. Journal of Personality and Social Psychology, 47(2), 415-426. Smith, R., Woodward, N., Wallston, K., Rye, P., & Zylstra, M. (1988). Healthcare implications of desire and expectancy for control in elderly adults. Journal of Gerontology: Psychological Sciences, 43(1), 1-7. Southwest Airlines. (1994). Annual Report. Retrieved July 20, 2002, from http://www.southwest.com/about_swa/financial/annual_reports.html Southwest Airlines. (1995). Annual Report. Retrieved July 20, 2002, from http://www.southwest.com/about_swa/financial/annual_reports.html Southwest Airlines. (1996). Annual Report. Retrieved July 20, 2002, from http://www.southwest.com/about_swa/financial/annual_reports.html Southwest Airlines. (1997). Annual Report. Retrieved July 20, 2002, from http://www.southwest.com/about_swa/financial/annual_reports.html Southwest Airlines. (1998). Annual Report. Retrieved July 20, 2002, from http://www.southwest.com/about_swa/financial/annual_reports.html Southwest Airlines. (1999). Annual Report. Retrieved July 20, 2002, from http://www.southwest.com/about_swa/financial/annual_reports.html Southwest Airlines. (2000). Annual Report. Retrieved July 20, 2002, from http://www.southwest.com/about_swa/financial/annual_reports.html Southwest Airlines. (2001). Annual Report. Retrieved July 20, 2002, from http://www.southwest.com/about_swa/financial/annual_reports.html Streufert, S., & Streufert, S. (1969). The effects of conceptual structure, failure, and success 99 on attributions of causality and interpersonal attitudes. Journal of Personality and Social Psychology, 11(2), 138-147. Stride, N. (1996, July 1). Travel agencies must be alert to change or they face extinction. New Zealand Herald, Sec. 3, p. 1. Sutherland, M., & Sylvester, A. (2000). Advertising and the mind of the consumer. Sydney, Australia: Griffin Press. Thompson, S., Cheek, P., & Graham, M. (1987). The social psychology of health. Newbury Park, CA: Sage. Thorndike, E. (1899). The associative processes in animals. Boston, MA: Ginn. Thorndike, E. (1911). Animal intelligence. New York, NY: MacMillan. Vellas, F., & Becherel, L. (1999). The international marketing of travel and tourism: A strategic approach. New York, NY: St. Martin's Press. Weiner, B. (1985). An attributional theory of achievement motivation and emotion. Psychological Review, 92(3), 548-573. Weiner, B. (1986). An attributional theory of motivation and emotion. New York, NY: Springer-Verlag. Weiner, B., Freize, I., Kukla, A., Reed, L., Rest, S., & Rosenbaum, R. (1971). Perceiving the causes of success and failure. Morristown, NJ: General Learning. Winstead, D. (2002, December 27). Travel agents must change to survive. Retrieved February 7, 2003, from http://www.gazette.net/200252/business/practical/1372001.html Wright, L., May, K., & Jackson, K. (1991). Exaggerated social control and its relationship to the Type A pattern as measured by the structured interview. Journal of Research in Personality, 25(1), 135-136. Wright, L., & Newman, R. (1994). A more bias proof measure for the Type A subcomponent of exaggerated interpersonal control. Educational and Psychological Measurement, 54(1), 146-155. Wright, L., Von Bussmann, K., Friedman, A., & Khoury, M. (1990). Exaggerated social control and its relationship to Type A behavior pattern. Journal of Research in Personality, 24(2), 258-269. Yesawich, P. (1996). Travel agents: Dinosaurs or divas? Lodging Hospitality, 52(2), 18. Zelen, S. (1988). New models, new extensions of attribution theory: The Third Attribution Personality Theory Conference, CSPP-LA, 1988. New York, NY: Springer-Verlag. 100 Zuckerman, M., Kuhlman, D., Thornquist, M., & Kiers, H. (1991). Five (or three) robust questionnaire scale factors of personality without culture. Personality and Individual Differences, 12(3), 929-941. Zummuto, R. (1982). Assessing organizational effectiveness. Albany, NY: State University of New York Press.