. 31 .51 .c: .2 1 wk... :3: n :2...» 3?... 2...}: 3 : 531. .J. ~ .9 . . . £53.... fittest! 3.1.3? .3. ‘ . ..z;\v\.: .. .. (.119: .Z .. A. .1 v I: lily-J: I» r x.‘ i. .C .. ..: . :1... i. MICHIGA Ill/IIIIll/IiiIii/iiiillifiilii/Yliillilli'lilii/illifill «a 3 1293 00793 3776 ll/ ——-— LIBRARY Michigan State University In -v ‘— This is to certify that the dissertation entitled The Influence of Shopper Type and Mood on Impulse Purchasing Behavior presented by Dong Whuy Lee has been accepted towards fulfillment of the requirements for Ph.D. degree in Business Administration / // /"' V . ._/ é;//‘l’ [A l A V or professor Date W9 MSU is an Affirmative Action/Equal Opportunity Insriturion 0-12771 PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. ”BATE DUE DATE DUE DATE DUE 2 .2 a 3 Fm 57766! J , q water-$31 F985' 5; 3903 4 FEgQ a? g 005 DEL o 3 2012 MSU Is An Affirmative Action/Equal Opportunity Institution THE INFLUENCE OF SHOPPER TYPE AND MOOD ON IMPULSE PURCHASING BEHAVIOR bY Dong Whuy Lee A DISSERTATION Submitted to Michigan State University in partial fulfillmemt of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Marketing and Transportation Administration 1990 W‘Q4W7 ABSTRACT THE INFLUENCE OF SHOPPER TYPE AND MOOD ON IMPULSE PURCHASING BEHAVIOR by Dong Whuy Lee Research for more than the last 40 years has shown impulse purchasing to be widespread among consumers and across numerous product categories. Earlier studies had focused on the nonpersonal aspects of this consumer behavior, such as the rate or the product categories. More recently, studies have shifted their attention to the psychological content of consumers under the impulse purchasing situation. In investigating the psychological aspects of impulse purchasing behavior, this study considered two influential factors: shopper type (recreational vs. economic) and mood (positive vs. negative). Both have shown a close relationship with impulse purchasing behavior, but the issue of impulse purchasing has not been studied including both factors together. A conceptual model relating impulse purchasing, shopper type and mood, was developed and two theories from psychology were used to explain it. Five hypotheses were developed based on the conceptual model. A field survey was conducted at a mall using questionnaire to measure entering mood, shopper type, impulse purchasing rate, and exiting mood. Data were collected from 555 subjects and analyzed by ANOVA, Duncan procedure of multiple range test, and t-tests. In comparing shopper type and mood, shopper type was found to be more influential than mood on impulse purchasing behavior. Recreational shoppers who entered the mall in a positive mood and economic shoppers who entered the mall in a negative mood, revealed relatively higher impulse purchasing’ rates. In comparing exiting mood states, among the shoppers who entered the mall in a positive mood, only recreational shoppers who exited in a positive mood showed the higher impulse purchasing rates than those who exited in a negative mood. In addition to the results of the hypotheses, the interaction effect between shopper type and mood was found, even though it is marginally significant. Copyright by DONG WHUY LEE 1990 To my parents, Kyung . and Jooky ACKNOWLEDGMENTS I want to thank Gilbert Harrell, Chairman of the dissertation committee, M. Bixby Cooper, dissertation committee member, and Thomas Page,Jr. , dissertation committee member for their considerable assistance. vi TABLE OF CONTENTS LIST OF TABLES ........ ...... ............... ..... ... LIST OF FIGURES ..... ....... .... ..... ...... ......... I. INTRODUCTION ................................ II. III. ISSUE AND OBJECTIVES ..... ............. ..... CONTRIBUTIONS .............................. LITERATURE REVIEW ... ........ ... ............... IMPULSE PURCHASING BEHAVIOR ............... . .. 1. Studies on Non-psychological Aspects .. 2. Studies on Psychological Aspects ....... SHOPPER TYPE - RECREATIONAL OR ECONOMIC ........ MOOD ....... .............. ...... . ...... ...... .. A, Studies on Mood in Psychology .............. v‘ 1) Mood and Recall ...................... .. 2) Mood and Judgment ..... .. ............... 3) Mood and Behavior ...................... Q} Studies on Mood in Marketing ............... 1) Studies on Mood and Shopping Behavior with Situational Factors .............. . 2) Studies on Mood and Shopping Behavior _i without Situational Factors ......... .. CONCEPTUAL MODEL AND HYPOTHESES ............... . THEORIES UNDERLYING THE CONCEPTUAL MODEL .... .. 1. Situation - Mood - Behavior ....... ..... . .. 2. Mood 1 - Purchase Type - Mood 2 ......... .. HYPOTHESES ..... ............................. .. HYPOTHESIS I ............................ .. HYPOTHESIS II ... ...... . ................. .. HYPOTHESIS III ............ ..... . ........ .. HYPOTHESIS IV ..................... ..... . .. HYPOTHESIS V ............................ .. Page ix IV. METHOD ........... MEASURES ......... 1. SHOPPER TYPE .............................. 2. MOOD ... 3. IMPULSE PURCHASING RATE ... .............. . SURVEY ................ ...... ................. 1. QUESTIONNAIRE 2. DESCRIPTION OF SURVEY ..................... 1) PLACE 2) TI“ ......OOOOOOOOUOCII......-...... 3) PROCEDURE .....QOO......‘IOCOOOOOOOOOO. SAMPLE ............ ......... V. RESULTS .......................... ....... .. HYPOTHESIS I RESULTS HYPOTHESIS II RESULTS HYPOTHESIS III RESULTS HYPOTHESIS IV RESULTS HYPOTHESIS V RESULTS VI. DISCUSSION ..................... .......... DISCUSSION OF RESULTS LIMITATIONS OF THE STUDY FUTURE RESEARCH AREAS coo ..... ooo-coocccoouonooo-c cccccc o-ooooo-oooooaoooo- ooooooo oooo-ooouooo-oooo oooooooooo coco-0.0.0.0... ...... ...-00.000.00.00... 0-... ooooooooooooooooooooooooo 00-0 ....... o ........ on ooooooooo OOOOOOOOOIOOOOI REFERENCES .... ....................................... APPENDIX A .....IOOCOOCOOOQOOOIOC ...... ......O... Consent Form APPENDIX B ......I... IIIIIIIII O. ...... ...... Questionnaire viii 65 65 65 67 7O Number LIST OF TABLES 1. ANOVA Results of Hypothesis I ........ ........... .. 2. Cell Means .......................... ............. . 3. Results of Hypothesis II/Positive Mood ...... ...... 4. Results of Hypothesis II/Negative Mood ............ 5. Results of Hypothesis III/Recreational Shopper .... 6. Results of Hypothesis III/Economic Shopper ........ 7. Results of Hypothesis IV/Recreational Shopper ..... 8. Results of Hypothesis IV/Economic Shopper ......... 9. Results of Hypothesis V/Recreational Shopper ...... 10. Results of Hypothesis V/Economic Shopper . ......... ix LIST OF FIGURES Number Page 1. Conceptual Model Describing the Relationship among Shopper Type, Mood, and Purchase Type ............. 53 2. Impulse Purchases as a Subset of Unplanned Purchases .......................................... 72 3. Interaction between Shopper Type and Mood .......... 80 4. Tests of Hypothesis II ............................. 82 5. Multiple Range Test/Duncan Procedure ............... 85 6. Tests of Hypothesis III ............................ 87 7. Tests of Hypothesis IV .................... ..... .... 91 8. Tests of Hypothesis V .............................. 94 9. Summary of the Results ............... .............. 96 10. Interaction Effects of Shopper Type and Mood on Impulse Purchasing Behavior ............. ........102 INTRODUCTION Impulse buying is a pervasive and distinctive type of consumer behavior and also the focus of many studies. Research for more than the last 40 years has shown impulse purchasing to be widespread among consumers and across numerous product categories (Clover 1950; Applebaum 1951; West 1951) . Bellenger, Robertson, and Hirschman (1978) found that between 27 and 62 percent of purchases at department stores fell into the impulse category and most product categories were affected by impulse buying. According to another study, nine out of ten shoppers occasionally buy on impulse (Welles 1986). Several reasons have been given for this escalating trend. Marketing innovations such as credit cards, cash machines, 24- hour retailing, home shopping networks, and telemarketing make it easier than ever for consumers to purchase products on impulse (Rook 1987). Also, factors such as work and leisure time pressures, increased geographic mobility, a rise in the number of two-career households, and greater discretionary income suggest that consumers may take less time and effort in purchase planning (Williams and Dardis 1972) . Furthermore, the growth of self-service merchandising, heavier levels of advertising, greater emphasis on in-store displays, and other promotional tools all make it more convenient for shoppers to 1 2 use the store itself as a catalog for making brand decisions (Cobb and Hoyer 1986). Consequently, the prevalent nature of impulse purchasing, as documented in the above studies, clearly warrants further attention and more comprehensive research. ISSUE AND OBJECTIVES Past studies on impulse purchasing could be classified into two basic groups. The first group of studies focused on measuring the rate of impulse purchasing or on finding out which product categories were bought by impulse (Clover 1950; West 1951; Cox 1964; Kollat and Willett 1967; Prasad 1975; Bellenger, Robertson and Hirschman 1978; Cobb and Hoyer 1986). However, considering the fact that almost anything can be purchased on impulse and that it is the individuals, not the products, who experience the impulse to consume as Rook and Hock (1985) pointed out, this category of studies did not seem to deal with the core of the issue. More recently, a second group of studies shifted its attention to the psychological content of consumers under the impulse purchasing situation (Weinberg and Gottwald 1982; Rook and Hock 1985; Rook 1987). That is, this group of studies tried to reveal the behavioral aspects of impulse purchasing. 3 However, these studies are still at the stage of descriptively summarizing observations and episodes of impulse purchasing behavior. So, testable hypotheses and empirical findings have not been presented yet. In addition, studies in this area do not include the theoretical or conceptual framework on which they are based. This brief review of past researdh on impulse purchasing indicates a need for research on impulse purchasing behavior which addresses the psychological aspects of consumers in an impulse purchasing situation with a theoretical framework and the empirically testable hypotheses. In examining, impulse purchasing behavior, this paper will consider two important factors - shopper type and mood. Both have been shown to be influential on purchasing behavior. It has been suggested that much can be learned about retail patronage behavior from looking at consumers' general shopping orientation (Darden and Ashton 1975; Bellenger, Robertson, and Greenberg 1977). Accordingly, considerable attention has been given to developing classifications of consumers' characteristic and behaviors. In recent years, an even more simple shopper typology, that of the recreational vs. economic shopper, has proven to be meaningfully related to shopping behavior in a variety of situations (Bellenger, Robertson, and Greenberg 1977). The characteristics of recreational shoppers 4 have been identified in several studies (Gillette 1970; Hansen and Deutcher 1977; Bellenger and Korgaonkar 1980; Gutman and Mills 1982; Barnes 1984; Jarboe and McDaniel 1987). Bellenger and Korgaonkar(1980) and Jarboe and McDaniel(1987), in particular, revealed that recreational shoppers haveaigreater tendency to buy what they want regardless of need and to spend more time in nonplanned shopping activities. This implies that the recreational shoppers are more likely to buy impulsively. Mood is another important factor influencing purchasing behavior. In the field of psychology, the obvious link between mood and behavior has been the subject of countless studies (Isen and Levin 1972; Isen and Simmonds 1978; Batson, Coke, Chard, Simth, and Taliaferro 1979; Isen, Clark, and Schwartz 1976; Clark and Isen 1982). The other important area of research relevant to mood is environmental psychology. The Mehrabian-Russell theory, a leading environmental psychology approach, offers a very sound theoretical basis for the relationship between situation, mood, and behavior (Mehrabian and Russell 1974). In marketing, studies have implicitly or explicitly considered mood as a mediating variable (Tauber 1972; Kotler 1974; Belk 1974, 1975; Markin, Illis, and Narayana 1976; Russell and Mehrabian 1976; Donovan and Rossiter 1982; Gardner 1985) by 5 examining the influence of the shopping atmosphere on consumer behavior. These findings support the importance of situational influence (n1 consumption behavior through the change in affective states. More specifically, other studies found or implied a close relationship between mood states and impulse purchase (Kotler 1974; Donovan and Rossiter 1982; Weinberg and Gottwald 1982; Belk 1984). These studies emphasized the importance of feeling states induced by situational and cognitive factors as the determinants of the extent to which the individual spends beyond his or her original plans. Consequently, this study is designed to investigate impulse purchasing behavior through its relationship with two important factors - shopper type and mood. For this purpose, a conceptual model will be developed and tested to answer the following three basic research questions. 1. Which factor is more influential on impulse purchasing behavior; shopper type or mood? 2. Do recreational shoppers buy more impulsively than economic shoppers under the same mood state? 3. Do shoppers under a positive mood state buy more impulsively than those under a negative mood state within the same shopper type? CONTRIBUTIONS Thus far in marketing, impulse purchasing, shopper type, and mood have been studied separately as three unrelated areas. However, there has been a consistent and strong implication that there is an overlapping point between these three areas. Therefore, this study will integrate three traditionally independent areas in marketing on the basis of the implications below. Shopper type studies have consistently shown implications that recreational shoppers buy more impulsively than economic shoppers. Even if these studies have never focused on impulse purchasing behavior as the main issue of their research, they frequently mentioned this stronger impulsiveness of recreational shoppers. Studies on mood also have shown a close relationship between impulse purchasing behavior and mood. These studies have found that people in a positive mood treat future financial consequences lightly, take less time to reach a purchase decision, and engage less in rechecking information. Secondly, as mentioned previously, earlier studies on impulse purchasing focused on the rate or the categories of product bought on impulse. That is, these studies focused on 7 nonpersonal aspects of impulse purchasing. However, this point has been criticized since it is the individual who makes the purchase impulsively. Accordingly, another contribution of this study is that the focus is on the personal aspect of impulse purchasing behavior - shopper type and mood. Thirdly, so far studies on impulse purchasing have not considered any factors which are relevant to or influential on impulse purchasing behavior. As mentioned, earlier studies focused on nonpersonal aspects of this purchasing behavior and relatively recent studies shifted their attention toward the psychological content of consumers, but they are descriptive and exploratory in nature. Therefore, another contribution of this study is that two influential factors on impulse purchasing behavior are included in the study: shopper type and mood. Finally, as pointed out, research to date on impulse purchasing has not presented any theoretical or conceptual basis for the study of this issue. This study will demonstrate a conceptual model emphasizing the interrelated nature of impulse purchasing, shopper type, and mood with support from theories grounded in the field of psychology. From all this, five hypotheses will be developed and tested empirically. LITERATURE REVIEW In order to investigate impulse purchasing behavior in relation to shopper type and mood, the relevant past studies for each area will be reviewed briefly in this chapter. In reviewing, emphasis will be placed on descriptions of the research in terms of the usefulness of the findings for this particular study. IMPULSE PURCHASING BEHAVIOR There have not been many studies on impulse purchasing, even though research on this issue dates back more than 40 years. However, studies on impulse purchasing behavior are basically classified into two groups - studies focusing on non- psychological aspects of impulse purchasing behavior and those emphasizing psychological aspects of this type of consumption behavior. 1. Studies on Non—psychological Aspects This category of studies deals with impulse purchasing behavior from the nonpsychological perspective. That is, the focus of this type of research is on nonpersonal aspects of 8 9 impulse purchasing behavior such as measuring the rate of impulse purchase (Clover 1950; West 1951; Kollat and Willett 1967; Prasad 1975; Bellenger, Robertson, and Hirschman 1978) or focusing on product categories purchased by impulse (Cox 1964; Prasad 1975; Bellenger, Robertson, and Hirschman 1978; Deshpande and Krishnan 1979). Most of earlier studies belong to this group. Clover(1950) studied the relationship between the length of store hours and impulse purchasing using 154 retail stores representing 19 different types. Using survey data, he found store hours had a significant influence on the type of purchase. The average estimated percentage of their sales due to impulse buying was 21.8 per cent. West(1951) empirically measured the rate of impulse purchase for many different product categories. The results of the survey indicated that 36.8 per cent of all purchases made in four types of outlets were impulse purchases. Kollat and Willett(1967) found that, in terms of relative frequency, the average customer purchased 50.5 per cent of the products on an unplanned basis. For this study, they presented an intention typology which explained the level of specification of intention before entering the store. The major intentions were classified into five categories - product and brand, product only, product class only, need recognized, and need not recognized. Prasad (1975) compared the rate of impulse purchasing between discount and 10 department stores. The result was that the incidence rate of unplanned buying was consistently higher(62.4%) in the discount stores than in the department stores(39.3%). Bellenger, Robertson, and Hirschman (1978) conducted a survey with a sample of 1600 department store customers to investigate the extent of impulse purchasing. Their findings indicated 38.7 per cent of the purchases were on impulse. Another group of studies has focused more on the product categories bought on impulse. To study the relationship between shelf space and product sales, Cox (1964) classified test products into staple and impulse products without presenting an explicit basis for this classification. In the end, he failed to support the hypothesis that impulse items respond more to variations in shelf space than do staples. Prasad (1975) presented rankings of merchandise categories according to the rate of incidence of unplanned buying at department and discount stores. The "Women and Girls' Wear" category and the "Children and Infants' Wear" category had the highest and the lowest rates, respectively, at the department stores. At the discount stores, the "Men and Boys' Wear" category and the "Toys" category showed the highest and the lowest rates, respectively. Bellenger, Robertson, and Hirschman (1978) found that the percentage of impulse purchases showed considerable variation by merchandise line with a low of 27 per cent in women's lingerie and a high of 11 62 per cent in costume jewelry. Deshpande and Krishnan (1979) found that impulse purchases are associated with low cost items and they appear not to be associated with income or with the usage of credit cards. Besides the studies on the rate or the product categories of impulse purchasing, there are several other studies on impulse purchasing behavior which do not deal with the issue from a psychological point of view. Applebaum (1951) summarized his experience in grocery stores in terms of type of customers and their general buying behavior patterns. Referring to the issue of impulse purchasing, he pointed out the problems in measuring impulse purchasing such as the imperfection. of shopping lists (written or mental). Stern (1962) presented four types of impulse buying - pure impulse buying, reminder impulse buying, suggestion impulse buying, and planned impulse buying. This seems to be fairly comprehensive in explaining impulse purchasing behavior. He also listed factors influencing impulse purchasing based on his observations, such as low price, mass distribution, ease of storage, etc.. Kollat and Willett (1967) tried to discover what customer characteristics are associated with differential susceptibility to unplanned purchasing using demographic variables and general food shopping behavior variables. However, the results revealed that these variables do not 12 influence the percentage of unplanned purchases. Prasad (1975) investigated to what extent differential levels of unplanned buying may be explained by customer differences in socioeconomic-demographic variables. Results indicated that the selected socioeconomic-demographic characteristics did not explain the difference in shoppers' susceptibility to unplanned buying. However, based on the study by Bellenger, Robertson, and Hirschman (1978), age, dollar amount of purchase, and merchandise line did yield significant differences between impulse and nonimpulse purchases. They found that impulse purchasing tended to be greater for the under-35 and over-65 age groups and that customers making purchases in the $35-to-$50 range made a larger percentage of impulse purchases than those in the $5-to-$10 range. As reviewed briefly, these studies focus on the rate of impulse buying and/or the product categories bought on impulse. Most earlier studies on the issue of impulse purchasing belong to this category and have been criticized for their limitations. Besides the most blatant limitation of lacking an appropriate theory base, the focus of study was not on the behavioral aspects of consumers. That is, considering the fact that the key player in buying is the individual, studies need to focus more on the consumers' psychological aspects or, at least, variables explaining this type of purchasing behavior, and the behavioral nature of the l3 impulses to consume as pointed to by Rock and Hoch (1985). In addition to this key drawback, there are other shortcomings in the aforementioned group of studies as criticized frequently by more recent studies. A product orientation has dominated much of the discussion about impulsive consumption. However, considering that almost anything can be purchased impulsively (Rock and Hoch 1985; Cobb and Hoyer 1986; Rock 1987), placing sole emphasis upon product type provides a limited perspective in studying this consumption behavior and tends to divert attention from the internal motivation and its expression that is crucial to impulse purchase. Another problem of this group of past studies is the inconsistency of their findings about the rate of impulse purchase. At least, a couple of explanations for this problem can be discussed. First, the operationalization of measuring the rate of impulse purchasing varied (Cobb and Hoyer 1986). The use of different definitions, and, therefore, different means of measurement makes it difficult to compare findings across studies. Second, the types of store used varied across studies. For instance, Clover (1950) used variety stores, Applebaum (1951), Cox (1964), and Kollat and Willett (1967) obtained data from the grocery stores, and Bellenger, Robertson, and Hirschman (1978) and Deshpande and Krishnan (1979) conducted their studies at department stores. 14 Therefore, comparing the rate of impulse purchase based on the data obtained from these different types of stores has not provided conclusive evidence. 2. Studies on Psychological Aspects Recently, studies on impulse purchasing have shifted their attention toward psychological aspects of impulse purchasing behavior. This stream of research has tried to investigate the psychological content of consumers under impulse buying situations, such as psychological elements of impulse purchase (Rook and Hoch 1985), distinctive behavioral features of impulse buying (Rook 1987), and variables differentiating planner, partial planner, and impulse purchaser (Cobb and Hoyer 1986). Weinberg and Gottwald (1982) characterized impulse buying as making purchases with high emotional activation, low cognitive control, and largely reactive behavior. They measured degree of activation, cognitive processes, and/reactive processes using galvanic skin reaction, interviews, and observations in a laboratory setting. The results showed that impulse buyers assess themselves as being more emotionalized than nonbuyers, that impulse buyers differ significantly in their facial expressions from nonbuyers, and that information processing 15 plays a part in the buying decision, but its influence is smaller than that of emotional engagement. Rock and Hoch (1985) sought to develop a general model which could account for both the common process and content elements in consumers' impulse episodes from in-depth interviews. This study summarized five crucial elements that distinguish impulsive from nonimpulsive consumer behavior. These five elements are: a sudden and spontaneous desire to act, a state of psychological disequilibrium, the psychological conflict and struggle, a reduced cognitive evaluation, and ignorance of the consequences. Additionally, demographic profiles were examined to see the relationship between personal characteristics and impulsivity. Females were found to enjoy shopping more than males and to be more impulsive in their shopping behavior. Younger people tended to be more impulsive in their shopping behavior than older people. Cobb and Hoyer (1986) assessed the extent of planning which occurs prior to entering the store in terms of both product category intent and specific brand intent for two common grocery products. Based on these measures, they identified three purchaser types - planner, partial planner, and impulse purchaser. As the independent ‘variables, five groups of predictors - general shopping behavior, personality, demographics, task variables, and shopping lifestyles - were 16 examined to assess their viability in making differentiations among the three groups. Among the findings, task variables proved the most useful in discriminating among purchaser types across product categories. of the demographic variables, sex proved significant for both product categories they used. Recently, Rock (1987) addressed the issue of impulse purchasing by providing a discussion of selected social science interpretations of impulsive human behavior, a review of previous research, and a report of the findings from an exploratory study of consumers' self-reported impulse buying episodes. From a questionnaire asking three open-ended questions about impulse buying experiences, he found. the following distinctive features of impulse buying - spontaneous urges to buy, power and compulsion, excitement and stimulation, synchronicity(a perfect and unexpected match between need and product), product animation, hedonic elements, conflict, and disregard for consequences. Studies reviewed in this section showed the shift in focus toward the psychological content of consumers under impulse buying situations or, at least, toward the search for explanatory variables for this type of buying behavior. This shift indicates critical development in the study of impulse purchasing. Therefore, this recent stream of research seems to have begun to focus on the core of the issue. However, 17 there are still at least two critical and basic limitations which should be discussed and resolved. Most importantly, none of these studies proposed any theoretical or conceptual framework on which a study is supposed to be based. Given the fact that one of the important contributions of a study is to refine a theory and/or add new knowledge to a theory and that a theory serves as the framework of studies on an issue, this lack of theoretical framework is the most critical shortcoming which needs to be resolved. Another limitation of these studies is that their conclusions are not based on the results of empirical tests using hypotheses. Rather, the conclusions were mainly obtained from collecting and summarizing descriptions of consumers' episodes. Accordingly, they are descriptive and phenomenological. This limitation goes hand in hand with the lack of conceptual framework mentioned previously. SHOPPER TYPE - RECREATIONAL OR ECONOMIC The shopper typologies proposed in the literature share a common goal of categorizing consumers into a limited number of groups, or ‘types. Great diversity across studies is evident in how, or on what basis, the various shopper types are discerned. Moreover, the focus of the studies has ranged 18 from specific products, to groups of products, to the retail marketplace in general. Accordingly, a brief review of the various studies is useful in appraising current knowledge of shopper types. The first taxonomy of shoppers was offered by Stone (1954), whose interest was in understanding the social relationship binding urban residents to the community. Using depth interviews with a sample of 124 female department store shoppers, Stone was able to differentiate four fairly distinct shopper types based on their varying orientations toward the activity of shopping: the economic consumer, the personalizing consumer, the ethical consumer, and the apathetic consumer. Economic shoppers were characterized by a careful approach to shopping, giving heightened attention to merchandise assortment, price, and quality. Personalizing shoppers appeared to seek a personal relationship with the retail personnel, while ethical shoppers were willing to sacrifice lower prices and wider selections of goods in order to behave consistently with moralistic beliefs. Finally, the apathetic shopper did shopping largely out of necessity, with the shopping activity holding no intrinsic interest and constituting, at best, a burden. Stephenson and Willett (1969) proposed a conceptual taxonomy of shoppers based on the relationships between major aspects 19 of consumers' shopping strategies and the key characteristics of the transactions. According to their framework, four types of shoppers are store loyal shoppers, compulsive-recreational shoppers, convenience shoppers, and price—bargain conscious shoppers. Darden and Reynolds (1971) investigated general consumer shopping orientations in relation to product usage. Their study provides additional evidence of Stone's (1954) findings. Although, they did not attempt to classify consumers into discrete types, Darden and Reynolds nevertheless found support for the shopper orientations noted earlier by Stone. Using grocery shoppers, Darden and Ashton (1975) factor analyzed consumers' ratings of preference for supermarket attributes and were able to discern seven distinct ' configurations. The specific shopper types identified were quality-oriented shoppers, fastidious shoppers, convenience shoppers, demanding shoppers, trading stamp collectors, stamp avoiders, and apathetic shoppers. Moschis (1976) studied shopping orientations among cosmetic buyers using structured AIO (activity/interest/opinion) type rating measures and developed a configuration of six distinct shopper types - store-loyal shoppers, brand-loyal shoppers, "specials" shoppers, psychosocializing shoppers, name- conscious shoppers, and problem-solving shoppers. Moschis's typology seems to mix shopping behavior strategies and underlying motivational—perceptual orientations. Williams, 20 Painter, and Nichols (1978) based their grocery shopper typology on the perceived attributes of preferred grocery stores, rather than the preference ratings for various store attributes as Darden and Ashton (1975) did using grocery shoppers. Analyzing these data with a clustering routine, they found four shopper types: the low-price shopper, the convenience shoppers, the involved shoppers and the apathetic shoppers. Recently, Westbrook and Black (1985) attempted to develop a motivation-based shopper typology. Considering subjective satisfaction as a direct indicator of motive strength, they hypothesized seven major dimensions of shopping motivation and confirmed them. The seven dimensions of shopping motivation identified were: (1) anticipated utility of prospective purchases; (2) enactment of an economic shopping role; (3) negotiation to obtain price concessions from the seller; (4) optimization of merchandise choice in terms of matching shoppers' needs and desires; (5) affiliation with reference groups; (6) exercise of power and authority in marketplace exchanges; and (7) sensory stimulation from the marketplace itself. Application of this taxonomic analysis procedure allowed Westbrook and Black to identify six groups of shoppers, differing appreciably in the.gratifications received from shopping activity - (1) consumers involved in virtually all aspects of the process of shopping identified, (2) 21 consumers with high level of motivation to optimize merchandise choice and the desire to exercise power and authority, (3) consumers who are shopping-process apathetic, (4) consumers with choice optimization and role enactment motivations, (5) consumers with desires for the gratification of both economic and noneconomic motivations, and, finally, (6) consumers appearing to derive only moderate levels of gratification from each of the seven postulated shopping motives. The significance of this taxonomy is its foundation on the motivations which presumably account for consumers' shopping behavior. Emerging from these taxonomic efforts reviewed above is a diverse array of shopper types. Such diversity is perhaps not surprising in view of the wide variation in empirical approaches and research contexts across studies. Shopper taxonomies have been proposed for individual product classes, broad product assortments, shopping centers, and shopping as a general activity. Moreover, taxonomic efforts have been based on a variety of different forms of shopper response; store attribute importance ratings, store image characteristics, endorsements of shopping activity/interest/opinion statements, and global attitudinal expressions about shopping. Accordingly, consistency across studies can hardly be found and, therefore, it is difficult to draw any significant generalizations from these findings. 22 In recent years, attention has been focused on a more simple shopper typology which has proven to be meaningfully related to shopping behavior in a variety of situations. That is the recreational/economic shopper classification (Bellenger, Robertson, and Greenberg 1977; Bellenger and Korgaonkar 1980; Williams, Slama, and Rogers 1985). In studying motives for retail patronage, Bellenger, Robertson, and Greenberg (1977) suggested two shopper types labeled as (l) the convenience (economic) shopper and (2) the recreational shopper as the result of their study using a sample of female shoppers. The notion underlying this classification is that some consumers view shopping as strictly a means to an end, something to be endured, while others actually enjoy shopping and treat it as a form of recreation. The title "economic shopper" may be misleading in that economic shoppers are not necessarily interested in obtaining the best values; they simply dislike shopping. Those who enjoy shopping as a leisure activity are called recreational shoppers. As an extension of the study by Bellenger, Robertson, and Greenberg (1977), Bellenger and Korgaonkar (1980) designed a study to contribute additional insights into the characteristics and behaviors of shoppers with specific shopping orientations. This research also used the orientation of recreational shopping as opposed to convenience or economic shopping. 23 Another study which is also an extension of Bellenger, Robertson, and Greenberg (1977) was published by Williams, Slama, and Rogers (1985) to compare two shopper types - recreational shopper and economic shopper - in terms of a number of strategically important shopping characteristics. Those characteristics examined included the tendency to give and receive personal advice regarding purchases, price sensitivity and deal proneness, attitudes toward advertising, attitudes toward generic, store, and national brands, attitudes toward different types of retail outlets, orientation toward fashion, attitudes about credit, the tendency to seek variety, and, finally, the tendency to be brand loyal. _§ehavioral characteristics found by these studies using a simple classification of shoppers - recreational and economic - offer important implications in understanding different behaviors of different shoppers. Especially, characteristics of recreational shoppers give valuable and direct insights to understanding impulse purchasing behavior. Bellenger and Korgaonkar (1980) hypothesized that recreational shoppers engage more in nonplanned purchases and are more likely to continue to shop after making a purchase. This hypothesis was strongly confirmed. Furthermore, they described recreational shoppers as those who are less likely to have an idea of what 24 they are going to buy when they go shopping. In conclusion, they found that impulse purchases account for a high percentage of sales, that recreational shoppers are more likely to go on shopping trips without a preplanned purchase in mind,and that a general profile of recreational shoppers suggests a greater amount of impulse purchasing. Williams, Slama, and Rogers (1985) also presented similar findings. In studying behavioral characteristics of the recreational shopper, they found that recreational shoppers, as compared with economic shoppers, have a greater tendency to buy what they want regardless of need, and to spend more time in non- planned shopping. There are also several similar findings in the studies which classified shoppers into more than two categories. Even though these studies classified shoppers into more complicated types, shoppers who could be regarded as recreational shoppers in a dichotomized scheme tended to exhibit more unplanned purchasing behaviors. Tauber (1972) supported the tendency of impulse purchasing behavior by recreational shoppers based on the motives for shopping he classified, He argued that the mobility of shoppers increases exposure to new shopping alternatives and enhances opportunities for impulse shopping. He also believed impulse shopping may be prompted by the motives which are 25 unrelated to actual buying such as role playing, physical activity, sensory stimulation, status , authority, and communication with others having a similar interest. In studying demographic profiles and behavioral characteristics of Sunday shoppers, Barnes (1984) found that recreational shoppers account for 40 percent of Sunday shoppers. That is, forty percent of Sunday shoppers go to the mall for no specific reason. This study also described recreational shoppers as those who go shopping in order to acquire information and for "something to do," those who do not have any particular purchase in mind, those who tend to make impulse purchases, and those who continue to browse after making a purchase. In studying a profile of browsers in regional shopping malls, Jarboe and McDaniel (1987) mentioned a simple recreational aspect of browsing. They see browsing as recreational window shopping motivated only slightly by the desire to make a purchase. They also describe browsers as potential impulse buyers who are more likely to engage in unplanned shopping. One of the hypotheses that browsers are more likely to return home with items they have not previously planned to purchase was supported empirically in this study. Another similar study on browsing done by Bloch, Ridgway, and Sherrell (1989) also mentioned a recreational aspect of browsing activity and similar profiles of browsers. They stated that, although 26 browsers may patronize a retailer without specific purchase plans, they may in fact make unplanned purchases because of in-store promotion and exposure to new products. This again shows the high possibility of unplanned purchasing by recreational browsers. So far, many different shopper typologies have been reviewed in this section. Concerning the complexity and inconsistency of earlier typologies, the more recent and simple typology focusing on a two-factor classification system of economic versus recreational shoppers seems to be more useful in that it is simpler and it tends to account for more of the other shopping orientations discussed in earlier studies. This review of previous studies on shopper types also gives evidence that impulse purchasing behavior is closely related to recreational shopping orientation. As compared to the economic shoppers, the recreational shoppers are less likely to have an idea of what they are going to buy before going shopping, are more likely to engage in unplanned purchasing, are more likely to continue shopping after making a purchase, and spend more time on unplanned shopping. 27 MOOD Mood has been regarded as an important variable influencing behavior. In psychology, there have been many studies supporting the close relationship between mood and general human behavior. Recently, in marketing, more attention has been given to the relationship between mood and consumption behavior to investigate the influence of feeling states of consumers on shopping behavior. In this section, a brief review is presented of the studies on the relationship between mood and behavior from psychology and from marketing. 1. Studies on Mood in Psychology The literature in social psychology contains many studies in which terms such as "feelings," "emotions," "moods'" and, more generally, "affect" are used. Moods are generally defined as the feeling states that are subjectively perceived by individuals, which are transient (Gardner 1985). Moods are particularized to specific times and situations (Peterson and Sauber 1983) and contrasted with those that are relatively stable and permanent (Westbrook 1980). Moods are distinguished from emotions, which are usually more intense, are closely tied to specific behavior, and are likely to disrupt ongoing behavior (Clark and Isen 1982). 28 Feeling states or moods are induced by pleasant or unpleasant experiences - pleasing music, noise, a beautiful scene, something positive or negative happening to a person — or by recall of positive or negative experiences from memory. The cause of a person's feeling state does not necessarily become the target of the behavior affected by it. A given feeling state may affect a great variety of judgments and behaviors. In other words, feelings have neither specific behavioral impulses nor specific targets associated with them. Rather, whatever induces a feeling state tends to gently redirect ongoing thinking and behavior and/or to affect what behavior or thoughts will occur next, but within the existing context (Clark and Isen 1982). In summary, feeling states are general and pervasive, having no inherent targets, and they usually do not interrupt ongoing behavior. They are relatively transitory, they can occur frequently, often in the normal course of everyday life, and they consist of thinking about positive or negative material. In contrast, emotions are more intense and do involve arousal as well as a cognitive component that is usually thought to provide the interpretation and the positive or negative valence of the emotion (Schachter and Singer 1962). Further, the intensity of emotions, combined with the fact that emotions are closely tied to specifiable behavior, means that 29 emotions are likely to disrupt ongoing behavior and to result in behavior directed toward a different goal (Clark and Isen 1982). However, differentiating clearly between emotion and mood is not easy since, in many cases, the feeling state and emotion are elicited together. That is, mood and emotion have basically the same nature with different levels of intensity. Research in psychology indicates that mood states exert an important influence on recall, judgment, and behavior. Therefore, the following sections review findings about the influence of mood states on recall, judgment, and behavior, respectively. 1) Mood and Recall The relationship between mood and recall has been explained largely in network models of memory. In these models, human memory can be demonstrated in terms of an associative network of semantic concepts and schemata that are used to describe events. An event is represented in memory by a cluster of descriptive propositions. These are recorded in memory by establishing new associative connections among instances of the concepts used in describing the events. The basic unit of thought is the proposition; the basic process of thought is activation of a proposition and its concepts. The contents 30 of consciousness are the sensations, concepts, and propositions whose current activation level exceeds some threshold. Activation presumably spreads from one concept to another, or from one proposition to another, by associative linkages between them. Activation of a node can be accomplished either by presentation of the corresponding stimulus pattern or by prior activation of an associated thought (Bower 1981). In other words, each distinct mood has a specific node or unit in memory that can enter into associations with coinciding events. Activation of this mood unit aids recall of events associated with it: it also primes emotional schemata for use in free association, fantasies, and perceptual categorization. More specifically, mood-state-dependent effect seems to be useful in explaining the relationship between mood and recall. Mood-state-dependent effect means that events learned in one mood state can be remembered better when one is put back into the same state one was in during the original experience. That is, memories acquired in one mood state are accessible mainly in that state but are "dissociated" or not available for recall in an alternate state. It is as though the two mood states constitute different libraries into which a person places memory records, amd a given memory record can be recalled only by returning to that library, or mood state, in which the event was first stored (Bower 1981). 31 There are many findings supporting this effect of mood-state- dependency. Moods act as powerful releasers for memories stored at earlier occasions of those moods, and unaided recall is enhanced when mood at the time of retrieval matches mood at the time of encoding, and when the encoding mood can serve as a retrieval cue (Bower, Monteiro, and Gilligan 1978). People in a pleasant mood recall a greater percentage of their recorded pleasant experiences than of their unpleasant experiences, whereas people in an unpleasant mood recall a higher percentage of their unpleasant rather than their pleasant memories (Bower 1981). Material stored in a person's memory linked with a positive feeling will be more likely to come to mind (Isen, Shalker, Clark, and Karp 1978) and will come to mind faster (Teasdale and Fogarty 1979) when a person is in a positive feeling state than when he or she is not. Moods cue similarly toned material in memory and cause that material to be more likely to come to mind (Clark, Milberg, and Ross 1983). Mood at the time of exposure affects what information is recalled by facilitating the retrieval of mood- congruent items (Gardner 1985). Retrieval moods facilitate overall retrieval of mood—congruent material from memory (Clark and Waddell 1983). 32 2) Mood and Judgment In general, mood states have been found to bias evaluations and judgments in mood congruent directions. There seem to be two processes which result in this tendency. A mood state increases the availability of mood-congruent thoughts or information in memory and, then, this mood-congruent information is used as the basis for current judgment or evaluation. That is, people who are in a pleasant or an unpleasant mood increase the availability of similarly valenced events in memory. This, in turn, could lead people to overestimate the prevalence of such events in their lives and bias their judgments. There are many studies showing consistent evidence that mood states bias evaluations and judgments in mood-congruent directions. Isen, Shalker, Clark, and Karp (1978) induced a positive feeling state in some randomly selected people in a shopping mall by giving them a free gift. People who had received the free gift, in contrast to a control group, later reported, on an apparently unrelated consumer survey, that their cars and television sets performed better and had better service records. Bower (1981) also found that mood affects the way people elaborate on or draw inferences from interpersonal events and that their expectations and predictions are positive or negative depending on their mood. 33 This study also showed evidence that mood biased the categories used in interpreting events, especially when the events are ambiguous. When subjects were feeling good, the recalled incidents were judged as more pleasant and when they were feeling bad, the incidents were evaluated as more unpleasant. Clark and Isen (1982), in studying the relationship between feeling states and social behavior, also mentioned that such feeling states, negative as well as positive, influence what people think about and the judgments they make, and thus have important effects on social behavior. Furthermore, they found that feeling states influence the prediction of the future. That is, people in a pleasant mood predict the future more positively and those in a unpleasant mood evaluate it more negatively. Forest, Clark, Mills, and Isen (1979) found that people, in whom a positive feeling state has been induced, rate slides of ambiguous scenes as more pleasant than do people who are not in a positive feeling state. Schwarz and Clare (1983) explained the various influences of feeling states on judgment and evaluation. They explained that a happy or a sad event in life may increase the availability of similarly valenced events in memory and this, in turn, could lead people to overestimate the prevalence of such events in their lives and bias their judgments. In 34 addition, people may use their mood at the time of judgment as information in evaluating current events. 3) Mood and Behavior As reviewed above, mood influences recall and judgment toward a mood-congruent direction. Accordingly, these influences of mood on recall and judgment have important effects on behavior. People who are in positive moods have been found to behave in ways that can maintain these positive feeling states. For example, being in a positive feeling state has been shown to cause people to reward themselves more generously, to choose to look at positive rather than negative self—relevant information, to help others more, to report greater liking for others and more positive conceptions of people, to increase willingness to strike up a conversation or to approach strangers for information, and to be more receptive to persuasive communication (Clark and Isen 1982). However, the behavioral effects of negative moods may be more complex than the effects of positive moods. That is, it seems that negative feeling states do not as consistently produce antisocial behavior or reduce prosocial behavior. In addition, sometimes negative feeling states have been shown to increase positive behaviors, just as positive feeling 35 states do. There have been various findings which can support the complexity and inconsistency of the influences of negative feeling states on behavior as those mentioned above. Although, negative feeling states have sometimes been shown to increase antisocial behavior and aggression (Baron and Bell 1976), and have sometimes been shown to decrease prosocial behavior and helping (Underwood, Froming, and Moor 1977), there are many studies that report the failure to find effects for negative feelings that are opposite to the effects found for positive moods. For example, while Mischel and Moore (1973) found that people in positive feeling states did selectively choose to look at positive self-relevant information, people in negative feeling states did not Selectively choose to look at negative self-relevant information. While Isen (1970) found that positive feelings increased helping, she did not find that negative feelings decreased helping. In addition, negative feeling states such as guilt or embarrassment, incompetence, anger, and sadness haVe been shown to increase self-reward or to be associated with increased helping or compliance with a request (Clark and Isen 1982) . The relationship between feeling states and behavior, especially the complexity and inconsistency of the influence of l"negative feeling states on behavior, can be understood 36 through automatic and controlled processing proposed by Posner and Snyder (1975). Since these processes are based on the spreading activation theory of memory, a brief description on this theory will be mentioned first. Spreading activation theory, which is popular among cognitive psychologists, regards memory as a network which includes a large and permanent collection of nodes (a node is a point of interaction in a network). In this network, materials can be stored either semantically or episodically. That is, in addition to the possibility that feelings are linked to nodes representing individual behaviors, objects, situations, or episodes that occurred in the past, it is also possible that eXperiences are linked to each other in memory in accord with how they make the person feel. In short, this network is a Collection of nodes representing objects, experiences, behaviors, and episodes and nodes representing feeling tones. 1:'112I:‘thermore, the way they are associated is that similar feeling tones are closely related with episodic memories relevant to those feeling tones. In this network, there are two processes - automatic processes and controlled processes - which can explain the relationship be‘tVWeen behavior and feeling states. According to Posner and SnYder (1975) , automatic processes occur (1) without conscious awat‘eness, (2) without intention, and (3) without producing 37 interference with other ongoing mental activities. According to this process, when a person is experiencing a feeling state, thoughts or events associated with or responsible for that feeling state automatically cue similarly toned materials objects, experiences, behaviors, and/or feeling in memory and this activation (episodes, tones stored in the past) spreads out in the network. Thus one may be more likely to think about episodes, objects, experiences, and/or behaviors which are similar to the current feeling state. As a result, this increased accessibility of material related to a person's current feeling tone may then affect his or her impression of the world and behavior. When, for example, a positive feeling state manipulation, such as the giving of a free gift, takes place, material linked to Such acts (including a positive feeling tone and other material associated with that node) should be activated as a r eSult of automatically spreading activation and should either Come to mind, or be brought closer to coming to mind. Therefore, one may be more likely to think about how kind otfliers are, about how pleasant the day has been, and so on. Once the threshold for activation of positive material is reached, the accessibility of positive material in memory is incJreased and the person should be in a positive feeling state. 38 The same kind of analysis of the operation of automatic processes can be made for a negative feeling state. That is, if one is in a negative feeling state, possible negative behaviors stored in memory along with negative feelings may be activated. However, it should be pointed out that, overall, the effect of automatic processes may be less influential for negative moods and negative material than for positive states and positive material. In other words, the process that has been described thus far is consistent with the majority of effects positive feeling states have been shown to have, and similarly, but to a lesser extent, it accounts for the data on the effects of negative feeling states on behavior. However, this difference of influences of feeling states on behavior between positive and negative can be explained by anether process called controlled process. As suggested by Posner and Snyder (1975) , it is likely that controlled process interacts with automatic activation processes to determine behavior. Accordingly, it is believed that many of the effects of feeling states on behavior cannot be accounted for wi‘tliout suggesting that, in addition to automatic processes, feeling states give rise to certain distinctive controlled pr<><=esses or strategies. Controlled processing is a process 1n"<>J.ving a mechanism of conscious awareness and intention, one that takes time, requires effort, and is of limited 39 capacity. It ranges in complexity from complicated, planned strategies to the relatively simple set to perceive or react; but regardless of the degree of complexity of these conscious processes, their distinguishing characteristics are that they have a limited capacity. Considering these two processes, it may be that a positive feeling state gives rise to a conscious strategy to maintain that state. Mbre specifically, people in positive feeling states may direct their attention to material in memory that will maintain that mood through a controlled process. Thus people in a positive mood might think about behaviors that have produced positive feelings in the past and might be more likely to perform those behaviors in order to maintain their moods. People in negative feeling states may also use controlled strategies, but in that case, in order to change their moods. In other words, they may think of and perform behaviors associated with positive feelings specifically to relieve their negative feeling state. These differences imply that the behavioral consequences of any given mood or affect- generating situation will depend on whether and how effectively automatic or controlled processes are activated and on specific aspects of the situation related to the factors that influence this. Therefore, conditions that can attenuate the effectiveness of controlled processes may play a crucial role in the kind of behavior that will result when 40 a person is in the negative feeling state. So far controlled and automatic processing have been discussed as if they were separable. However, this separation is artificial. They jointly determine many of the effects that feeling states have on behavior. Again, the two types of processes are more likely to contribute to the same effects in the case of positive feeling states and positive behavior. And it is in the case of negative feeling states that automatic and controlled processes work in opposite directions. In other words, it seems reasonable to think that most of the time, when people are in a negative feeling state (automatic process), they would like to get out of that state, and will search for activities (controlled process) that might make them feel positive. This description raises the question of howr a person's consequent feeling state is determined after going through both automatic and controlled processes when these processes try to lead to opposite directions. It seem that the resulting feeling state. depends on ‘the strength. of two processes. That is, factors that can affect the willingness ofi a person to put effort (controlled process) into terminating a negative state (automatic process) might be a crucial determinant of whether the effects of automatic or controlled processes predominate. Thus it can be predicted 41 that the effects of automatic processes might predominate over the effects of controlled processes when one has failed in strategic efforts to alleviate his or her negative feeling state. Likewise, a successful controlled process might lead to the dominance of the effects of a controlled process over those of an automatic process. Conclusively, the influence of negative mood states on the behavior is more complicated and inconsistent than that of positive feeling states and, therefore, its prediction is difficult. 2. Studies on Mood in Marketing In marketing, recently, more attention has been given to the study of influence of mood on consumer behavior. Studies on this issue in marketing can be classified into two groups. One group of studies has dealt with mood as the mediator between situational factors and shopping behavior. Within this group, some studies explicitly have dealt with mood as a mediator between atmospherics and purchasing behavior (Kotler 1974; Lutz and.Kakkar 1975; Donovan.and.Rossiter 1982; Havlena and Holbrook 1986) , and others have studied its mediating role only implicitly (Belk 1974; Belk 1975; Baker 1975). The other group of studies has focused on the 42 relationship between mood and behavior without considering atmospherics or situational factors (Axelrod 1963; Zajonc and Markus 1982; Belk 1984; Gardner and Vandersteel 1984; Isen 1984; Gardner 1985). 1) Studies on Mood and Shopping Behavior with Situational Factors Studies which emphasize the relationship between situational variables and shopping behavior through the mediating role of mood have investigated this relationship using one of the leading environmental psychological approaches, namely the Mehrabian-Russell Model. The fundamental proposition of Merabian and Russell's theory is that the impact of the situation on behavior is mediated by emotional responses, so that any set of conditions initially generates an emotional (affective, connotative, and feeling) reaction, which in turn leads to a behavioral response. Further, the universe of all possible emotional responses may be represented by one or a combination of three basic dimensions - pleasure, arousal and dominance. As a way of studying the relationship among situational variables, mood, and.shopping behavior, Lutz and.Kakkar (1975) evaluated the Mehrabian-Russell theory. This study was 43 designed as a partial replication of Belk's (1974) study. The results supported the Mehrabian-Russell framework and stated that one of the primary advantages of the Mehrabian-Russell framework is that it allows quantification of situational descriptions along the three emotional response dimensions - pleasure, arousal, and dominance, thus facilitating the use of powerful multivariate statistical procedures in assessing situational influence. Further, Donovan and Rossiter (1982) tested the Mehrabian- Russell environmental psychology model in retail settings. For this study, they used only two emotional dimensions - pleasure and arousal - out of the original three emotional dimensions developed by the Mehrabian-Russell model. The results suggested that store atmosphere, engendered by the usual myriad of in-store ‘variables, is represented psychologically by consumers in terms of two major emotional states - pleasure and arousal - and that these two emotional states are significant mediators of intended shopping behaviors within the store. Additionally, this study listed many meaningful conclusions and plausible suggestions about the two emotional dimensions they used. Simple affect, or store-induced pleasure, is a very powerful determinant of shopping behaviors within the store, including spending behavior. The influence of emotional affect is often overlooked in retail store studies where cognitive influences 44 are emphasized. Arousal, or store—induced feelings of alertness and excitement, can increase time spent in the store and also willingness to interact with sales personnel. Finally, as predicted by the Mehrabian-Russell model, arousal works positively only in store environments that are already pleasant, and arousal may have no influence (or even a negative influence) in unpleasant store environments. Havlena and Holbrook (1986) assessed the comparative reliabilities and validities of two competing typologies of emotion - Mehrabian-Russell's and Plutchik's - when they are applied to the representation of experiences associated with consumption activities. Specifically, independent samples of judges rated real experiential descriptions on either Mehrabian-Russell's dimensions - pleasure, arousal and dominance - or Pdutchik's emotional categories. Further, separate samples of judges evaluated artificial descriptions on the same competing framework. They concluded that, in general, the three: dimensions of jpleasure, arousal, and dominance captured more information about the emotional character of consumption experiences than did their measurement via the eight basic emotional categories recommended by Plutchik. Therefore, within the context of consumer behavior, the Mehrabian-Russell framework is more useful than Plutchik's scheme for positioning consumption experiences in an emotion space and for developing experience- 45 specific emotional profiles. Kotler (1973) also studied the relationship between atmospherics and purchasing behavior with affective state as the mediator. This study, which is not relevant to the Mehrabian-Russell theory, presented a conceptual model showing the causal chain connecting atmosphere and purchase probability. This model describes the relationship using four steps - sensory qualities of atmospherics, buyer's perception of these qualities, modification of buyer's information and affective state by these sensory qualities, and impact of this modification on purchase probability. This model, basically, depicts the flow from situational variables to purchasing behavior through the mediating role of modified affective state. Also, in explaining this model, Kotler stated that atmosphere may serve as an affect-creating medium. That is, the colors, sounds, and textures of the space surrounding the purchase object may directly arouse visceral reactions that contribute favorably to purchase probability. There are also several studies investigating the relationship between situational factors and purchasing behavior implicitly considering the mediating' role of mood. They see the influences of situational variables on buying behavior through buyers' perceptions of the situations, which seem to include buyers' moods as a part. These studies emphasized the 46 importance of situational influences on consumer behavior and supported the strong relationship between situational variables and consumer behavior. Belk ( 1974) found that the amount of situational influence on consumer food preferences was apparent. In this study, situational main effects and interactions provided nearly half of the explained variance in meat and snack preferences. Another of Belk's (1975) studies on situational variables also stressed the importance of situational influence on consumer behavior. He argued that the growing recognition of limitations in the ability of individual consumer characteristics to explain variation in buyer behavior could be resolved by examining situational influences on consumer behavior. Barker (1975) emphasized the need for psychological theories of environment and the need for correct understanding of environments as being dynamic and purposeful in terms of persons in investigating the relationship between environments and consumer behavior. In addition to emphasizing the importance of situational influence on consumer behavior, Wicker (1975) made a list of suggestions in studying these influences, such as identifying the boundary and characteristics of the consumer situation, developing the descriptive dimensions of the situation, and investigating the processes by which features of situations come to influence behavior . 47 2) Studies on Mood and Shopping Behavior without Situational Factors There are several studies in marketing which deal with the influence of mood states on consumer behavior without considering situational variables. Works in marketing have often overlooked the effects of consumers' moods on their behavior. Recently, researchers have suggested that feeling states may play an important role in consumer information processing and in other aspects of consumer behavior. However, studies in this area have not.been intensive and.have not developed a sound theoretical basis, yet. Instead, research is currently at the stage of understanding the importance of the influence of mood on consumer behavior, reviewing and summarizing findings from psychology and trying to extend them to a consumer behavior context. For example, Gardner (1985) reviewed findings of influence of mood states on consumer behavior in general and concluded that mood states have direct and indirect effects on behavior, evaluation and recall. Again, this study mainly reviewed and summarized the findings from.psychology such as that positive Ihggdg, appear to enhance the likelihood that. a host of loehaviors may be performed, mood states seem to bias eevaluations and judgments in mood congruent-directions, and 48 unaided recall is enhanced when mood at the time of retrieval matches mood at the time of encoding and when the encoding mood can serve as a retrieval cue. Based on the review, Gardner proposed a conceptual model of the role of mood states in consumer behavior, which, specifically, depicts the influences of service encounters, point-of-purchase stimuli, and communications at retail stores on recall, evaluation, and behavior through the mediating role of consumers' mood states. However, the usefulness of this descriptive model is in question since this model is not examined empirically. Gardner and Vandersteel (1984) discussed the effects of buyers' moods on consumer behavior. In addition to a brief review of findings from psychology, they pointed out three circumstances under which the effects of mood on consumer behavior may be important. They are 1) a respondent's mood may be a nuisance factor inducing bias in a survey or experiment, 2) consumers' moods upon exposure to marketing stimuli may be determined by factors beyond the marketers' control, and 3) marketing stimuli may be used to influence consumers' moods. Also, as one of the methodological problems, they criticized the commonly used experimenter- controlled mood inducement method in that it is difficult to determine the exact nature of the induced mood through this manipulation method. Belk (1984) also examined applications of mood inducement in buyer behavior. In this study, he 49 emphasized the need to consider factors affecting buyer mood as well as some of the controllable antecedents, psychological processes, and consequences that are involved. There are some other studies regarding the influence of mood on the components of consumers' psychological processes, which underlie the resulting behavior. Axelrod (1963) studied the relationship between moods and attitudes toward products and found that the evaluation of products were systematically related to mood changes. In a study on the influence of positive affect on decision making and cognitive organization using induced positive affect, Isen ( 1984) supported the findings from psychology that positive affect can serve as a retrieval cue for positive material in memory and that the very strategies that subjects use in solving problems may be influenced by the presence of positive feelings. So far, studies on the relationship between mood and consumer behavior have been reviewed from both psychology and marketing literature. As stated previously, in psychology, mood states have been found to have a significant influence on recalling :materials from memory and evaluating objects. These influences of mood, states on recall and evaluation, are consistently toward mood-congruent directions. However, the influence of moods on behavior is more complicated. A positive mood leads behavior to the same direction (positive 50 behavior), but a negative mood results in behavior either negative or positive depending on the strength between a negative mood (automatic process) and an effort to change it (controlled process) as reviewed earlier. In marketing, studies have shown evidence that situational factors have influence on consumption behavior through the mediating role of moods. Based on these findings, the relationship between moods and, specifically, impulse purchasing can be examined. That is, consumers in a shopping situation are influenced by such factors in the atmosphere as colors, music, and decorations, which intend to make shoppers more pleasant and stimulated. These moods influenced by a pleasant atmosphere may make it easier for shoppers to recall mood-congruent pleasant items in memory, evaluate products more positively, and eventually imake shoppers spend money on unplanned items. In support of this relationship, there are studies specifically showing evidence of this impulse purchasing behavior through the influence of mood states as summarized below. Impulse buying decisions are accompanied by strong emotions and low cognitive control (Weinberg and Gottwald 1982). Mood seems to be a key ingredient in determining whether we feel like rewarding ourselves with a purchase and sharing by giving to others (Belk 1984). Whereas cognitive factors may largely 51 account for store selection and for most of the planned purchases within the store, emotional responses induced by the environment within the store are primary determinants of the extent to which the individual spends beyond his or her original expectations (Donovan and Rossiter 1982). Various components of the atmosphere may trigger sensations in the buyers that create or heighten an appetite for certain goods, services, or experiences (Kotler 1973). The strategies that people use in solving problems may be influenced by the presence of positive feelings. 'That is, those in the positive affect condition take significantly less time to reach a purchase decision, use fewer types of information, and engage in less rechecking of information (Isen 1984). Positive moods increase attraction, helping, expectations of future positive events, and the perceived pleasantness of scenes (Clark, Milberg, and Ross 1983). Impulse purchasing is buying which is not planned by the customer before entering a store, but ‘which results from a stimulus created by a sales promotional device in the store (Applebaum 1951) . CONCEPTUAL MODEL AND HYPOTHESES Based on the review of literature about shopper type, mood, and impulse purchase, respectively, a conceptual model has been developed, which depicts the influences of shopper type and mood on impulse purchasing behavior (Figure 1). This model consists of basically four constructs - Shopper Type, Mood 1, Purchase Type, Mood 2. For Shopper Type, a two-category classification is used since attention has been focused on this simple shopper typology in recent years and it has proven to be meaningfully related to shopping behavior in a variety of situations (Williams, Slama, and Rogers 1985). This is the typology of recreational and economic shopper originally developed by Bellenger, Robertson, and Greenberg (1977). Recreational shoppers are those who enjoy shopping as a leisure activity and economic shoppers are those who simply dislike shopping (Bellenger and korgaonkar 1980; Williams, Slama, and Rogers 1985). Mood is described as the feeling state that is subjectively perceived by the individual, particularized to specific time and situation, and is transient. For this study, mood is classified into two simple and basic categories - pleasure and displeasure. .Accordingly, pleasure includes feelings such as happy, joyful, satisfied, pleasant, etc. and displeasure 52 53 SHOPPER TYPE 0 RECREATIONAL o ECONOMIC E SITUATION MOODI SHOPPING ENVIRONMENT Y PURCHASE TYPE 0 IMPULSE o NONIMPULSE . SHOPPING SITUATION Y MOOD 2 CONCEPTUAL MODEL DESCRIBING THE RELATIONSHIP AMONG SHOPPER TYPE, MOOD. AND PURCHASE TYPE Figure! 54 represents feelings such as angry, sad, unpleasant, depressed, etc.. Mood 1 is the feeling state a shopper has when entering a shopping environment. It is likely to be influenced by any event in a shopper's everyday life. Mood 2 is the shopper's feeling state when exiting a shopping environment. It is considered to be influenced.by all the surroundings and by the purchases the shopper makes within the shopping environment. This Mood 2 may be different from Mood 1 due to the influences of situational factors and/or the purchase within the shopping environment. Or it may be the same as Mood 1 if a shopper is not influenced by situational factors or the purchase and/or a shopper's Mood 1 is too strong to be changed by any outside influences. Finally, type of purchase would.be either impulse purchase or :nonimpulse purchase. In this model, the type of purchase is «considered to be influenced by shopper type and Mood 1. At the same time, type of purchase is supposed to have an influence on Mood 2. The purchase itself can be classified into two groups - impulse purchase and nonimpulse purchase. An impulse purchase is the purchase of any item which is not planned before entering the shopping situation. That is, this impulse purchasing behavior is supposed to be influenced by the situational variables under the shopping situation and 55 Mood 1. A nonimpulse purchase is the purchase which is originally planned before entering the shopping situation. This nonimpulse purchase is regarded as the shopping behavior which is not influenced by the situational factors during shopping. THEORIES UNDERLYING THE CONCEPTUAL MODEL 1. Situation - Mood --Behavior This relationship among situation, mood, and behavior can be explained using a leading environmental psychology approach, the Mehrabian - Russell model. As was discussed in the literature review chapter, the key idea of this model is that situational factors initially generate an emotional reaction, ‘which leads to a behavioral response. This theory explains .Situation-Mood 1-Purchase and Shopping Situation-Mood 2-the behavior after leaving a shopping environment. That is, the .Situation before shopping results in.Mood 1 and Mood 1 has an influence on Purchase Type. By the same token, the Shopping Situation generates Mood 2, which would be influential on the behavior that followed after shopping. 56 2. Mood 1 - Purchase Type - Mood 2 This relationship among Mood 1, Purchase Type, and Mood 2 can be explained using two processes proposed by Posner and Snyder (1975), showing the effects of feeling states on behavior; automatic process and controlled process. The two processes can explain the relationship between Mood 1, Purchase Type, and Mood 2 of this conceptual model. Mood 1 can be regarded as the feeling state through automatic process. Making a purchase can be considered as the controlled process. According to the theory of the two processes, when Mood 1 is positive, making a purchase becomes the controlled process trying to maintain this positive feeling state. When Mood 1 is negative, making a purchase becomes the controlled process trying to change the negative moods into positive feeling states. The resulting feeling state (Mood 2) after these two processes will be positive when Hood 1 is positive. However, when Mood 1 is negative, the resulting Mood 2 can be either negative or positive depending on the strength of the effects of automatic and controlled processes as discussed earlier. That is, if the effect of automatic processes (Mood 1) predominates over that of controlled processes (making a purchase), the resulting Mood 2 will remain negative. But, if the effect of the controlled process (making a purchase), which tries to change the 57 negative mood into a positive one, predominates over the effect of the automatic process (Mood 1), then the Mood 2 will change to positive. HYPOTHESES Based on the model described above, five hypotheses have been developed and tested. HYPOTHESIS I: Shopper type is more influential on impulse purchasing behavior than mood. The objective of this hypothesis is to compare the strength of influence on impulse purchasing behavior between shopper type and.mood, .As discussed in the literature review chapter, shopper type is classified into recreational and economic shoppers and tells the degree Of preference for shopping. Shopper type is relatively permanent and stable. In other words, shopper type is not easily changeable and, therefore, is maintained for a relatively long period of time. 1However, mood .is relatively 'transitory and. unstable since it is perceived by the individual and it is particularized to a 8Pacific time and situation. Furthermore, mood is not intense, not tied to a specific behavior, and not likely to 58 disrupt ongoing behavior. Instead, mood tends to gently redirect ongoing thinking and behavior. Past studies have shown more evidence of the relationship between shopper type and impulse purchasing behavior than of the relationship between mood and impulse purchasing behavior. Many consistent findings on the relationship between shopper type and.impulse purchasing can be found in.past studies” The review of previous studies gives evidence that impulse purchasing behavior is closely related to recreational shopping’ orientation. Specificallyy as compared. to 'the economic shoppers, recreational shoppers: (1) are less likely to have an idea of what they are going to buy before going shopping, (2) are more likely to engage in unplanned purchasing, (3) are more likely to continue shopping after making a purchase, and (4) spend more time on unplanned shopping. Studies on the relationship between mood and impulse purchasing behavior are not intensive, yet. Instead, research is currently at the stage of understanding the importance of the influence of mood on consumer behavior, reviewing and summarizing findings from psychology and trying to extend them to consumer behavior situations. However, there are several findings supporting the relationship between mood and impulse purchasing behavior. Mood seems to be a key ingredient for 59 self-rewarding with a purchase (Belk 1984) . Whereas cognitive factors may account for the planned purchases, the mood induced. by the environment is a jprimary’ determinant Of unplanned purchasing (Donovan and Rossiter 1982). The atmosphere may trigger an appetite for certain goods, services or experiences through the change in mood state (Kotler 1973). Impulse purchasing results from a stimulus created by a sales promotional device in the store (Applebaum 1951). Based on evidence from.past studies, the influence of shopper type on impulse purchasing behavior seems greater than that of mood. In other words, the difference in impulse purchasing rates between the two shopper types (recreational vs. economic) is greater than that between the two mood states (pleasant vs. unpleasant). HYPOTHESIS II: Recreational shoppers buy more impulsively than economic shoppers under the same mood state. Many studies have been done on the issue of shopper types and their behavioral characteristics. Recently, research has been focused. on. a :more simple dichotomized. typology of shoppers. These studies have not directly focused on the issue of impulse purchasing behavior. Instead, as implications. of‘ their findings, these studies frequently 60 mention behavioral characteristics of consumers relevant to impulse purchasing behavior. Therefore, in HYPOTHESIS II, the relationship between shopper type -recreational vs. economic- and impulse purchasing behavior is to be directly tested. According to the findings from past studies, recreational shoppers tend to make more purchases on impulse than economic shoppers, as mentioned in the section on. HYPOTHESIS I. Therefore, in HYPOTHESIS II, recreational shoppers are hypothesized to make more purchases on impulse than economic shoppers under the same mood state. HYPOTHESIS III: Shoppers under a positive mood state buy more impulsively than those under a negative mood state within the same shopper type. This hypothesis is to examine the influences of a positive mood state and a negative mood state on impulse purchasing behavior within the same shopper type. Studies on the behavioral influences of different mood states can rarely be fOund in marketing literature. waever, the findings from studies in psychology can be the basis for supporting HYPOTHESIS III. People in a positive mood state have been shown to increase attraction, helping, expectation of future positive events and treat future financial 61 consequences lightly (Clark, Milberg, and Ross 1983), which would result in more unplanned purchasing behavior. In addition, the following evidence supports that shoppers under a positive mood state buy more impulsively than those under a negative feeling state. Positive moods appear to enhance the likelihood that a host of behaviors may be performed (Gardner 1985). People who are in the positive feeling state take less time to reach a purchasing decision and engage in less rechecking of information (Isen 1984). Pleasant materials are easier to recall than unpleasant materials, represented in greater quantities than unpleasant materials in memory and are more interconnected as well (Clark and Isen 1982). So, shoppers under a positive mood state are hypothesized to buy'more impulsively than shoppers under a negative mood state within the same shopper type. HYPOTHESIS IV: Among shoppers who start in a positive mood (Mood 1, Positive), those who exit in a positive mood (Mood 2, Positive) buy more impulsively than those who exit in a negative mood (Mood 2, Negative) within the same shopper type. 62 This hypothesis is to compare levels of impulse purchasing between positive mood and negative mood in Mood 2 for those who are in a positive mood in Mood 1. Shoppers in a positive mood in Mood 1 would try to maintain this positive mood state according to the theory. That is, automatic and controlled processes influence behavior toward the same direction - a positive direction - to maintain this mood state. That is, shoppers in this mood state are more likely to make impulse purchases in an effort to maintain their initial positive feeling state, and, accordingly, their exiting mood.state (Mood 2) is more likely to remain positive. Therefore, among shoppers who are in a positive mood in Mood 1, shoppers who are in a positive mood in Mood 2 are hypothesized to buy more impulsively than those in a negative mood state in Mood 2. HYPOTHESIS V: Among shoppers who start in a negative mood (Mood 1, Negative) , those who exit in a positive mood (Mood 2, Positive) buy more impulsively than those who exit in a negative mood (Mood 2, Negative) within the same shopper type. This hypothesis is to compare levels of impulse purchasing 63 between positive and negative mood states in Mood 2 for those who are in a negative mood in Mood 1. Two possibilities can be discussed for the change in mood from negative in Mood 1 to positive in Mood 2. First, it is supposed that shoppers will try to change their initial negative feelings (Mood 1, Negative) into positive ones. These active and conscious efforts through controlled process can be regarded as one possibility. In doing so, situational factors, such as promotional displays, sales people, etc., in the shopping environment, which are designed to make shoppers attracted to stores and/or products, also help these efforts of changing shoppers' initial negative feeling states. The other possibility is that the purchases themselves could change shoppers' initial negative moods (Mood 1) into positive ones (Mood 2). In other words, regardless of the influences from situational factors, by making purchases shoppers could feel better and, therefore, change initial negative mood states into positive ones. This implies that shoppers who are in a positive mood when they exit (Mood 2) are more likely to make purchases (including purchases on impulse) than those who exit in a negative mood. Both possibilities discussed above lead to the HYPOTHESIS V that among the shoppers who are in a negative mood when they 64 start shopping, shoppers who exit in a new positive mood are more likely to have made purchases impulsively than those who exit in the same negative mood within the same shopper type. METHOD A field survey was employed to test the five hypotheses developed based on the conceptual model. For the survey, a questionnaire was created using measures for shopper type, mood, and impulse purchasing rate, respectively. The survey was conducted.in.a closed.mall setting for ten days, including weekdays and one weekend. Data from 555 subjects were collected. In this chapter, details of measures, survey, and sample are discussed. MEASURES 1. Shopper Type This study' used. a five-item semantic differential scale developed by Williams, Slama, and Rogers in 1985. This scale is a modification of the original scale developed by Bellenger and Korgaonkar(1980). The scale by Bellenger and Korgaonkar involved the use of a five-point semantic differential scale to measure respondents' enjoyment of shopping. Responses ranged from "Enjoy very much" to "Strongly dislike." Recreational shoppers were classified as those who enjoy 65 66 shopping or enjoy shopping very much. Those who were neutral toward, disliked or strongly disliked shOpping were classified as economic shoppers. In order to overcome the weaknesses of the original single item scale, Williams, Slama, and Rogers (1985) developed a multiple item summative scale. For this scale, ten questions were originally generated. The questions referred to the respondents' enjoyment of shopping activities and their desire to spend time on shopping. Through an iterative process of eliminating questions with low item to total correlation, those questions yielding the highest Cronbach's alpha were found and designated as the final scale as shown below. I generally enjoy shopping for gifts for other people. I generally enjoy shopping for clothes. I like to shop for groceries. I enjoy going to stores to shop even when I am looking for nothing in particular. I enjoy shopping in general. This study obtained a Cronbach's alpha of .75 for this five- item scale, indicating acceptable internal consistency for the relatively short scale. The questions were measured on a five point Likert scale 67 ranging from "Strongly agree" to "Strongly disagree." Respondents in the current study were classified as recreational shoppers if they Obtained a score on the summative scale indicating that, on the average, they "Agree" or "Strongly agree" with the statements referring to their enjoyment of shopping. That is, the score for each question ranged from 0 to 4. "Strongly agree" was coded as 0, and "Strongly disagree" was coded as 4. Accordingly, total scores from the 5 questions ranged from 0 to 20. Lower scores indicated enjoyment of shopping, with those scoring from 10 to 20 classified as economic shoppers, and those scoring less than 10 classified as recreational shoppers. 2. Mood Mehrabian and Russell(1974) presented a valuable theory that the impact of the situation on behavior is mediated by emotional responses, so that any set of conditions initially generates an emotional reaction, which in turn leads to a behavioral response. Furthermore, they proposed a three- dimensional scale to measure these emotional states, known as PAD - Pleasure, Arousal, and Dominance. This measure consisted of 18 items - 6 items for each dimension. The current study uses this PAD scale to measure the shoppers' mood states. 68 In developing measures for mood using these three dimensional variables, Mehrabian and Russell created a series of several hundred hypothetical situations, and presented them to subjects in three separate studies. In doing so, they were able to obtain responses on several semantic differential scales designed to measure the three basic emotions mentioned above. By factor- analyzing the data, it was found that the three factors corresponding to pleasure, arousal, and dominance could be used to construct a simple 18-item instrument for self-report.measurement of emotional responses to any conceivable situation. The Mehrabian-Russell dimensions of pleasure, arousal, and dominance represented an attempt to define emotional states. It should be noted that, although the first two dimensions have been well established in the literature(Smith and Ellsworth 1985) , the third dimension of dominance has not been well supported (Russell 1978; Russell and Pratt 1980; Donovan and Rossiter 1982). That is, evidence for the suitability of the pleasure and arousal dimensions appears very convincing over a broad spectrum of situations, but evidence for the dominance dimension is more tenuous. Furthermore, Russell, who developed the measurement, also agreed that the six items of the dominance dimension required a cognitive interpretation by the subject, and therefore, would not be purely applicable 69 in situations calling for only affective responses. Arousal, the second dimension, is activity-oriented. It is a measure of how "wide awake" the organism is, and of how ready it is "to act." Most studies have not regarded emotion and mood as different, and, furthermore, differentiating clearly between emotion and mood has not been easy since they are usually elicited together. However, arousal has been thought to be more relevant to emotion than to mood, when the differentiation between emotion and mood is made. Secondly, some studies have shown that arousal and.affect have a weak relationship. For example, Donovan and Rossiter(1982) found that the arousal regression coefficient for affect is near zero, and proposed that affect is related more to pleasure than arousal. Accordinglyy in measuring mood states for this study, only the pleasure dimension was used to classify mood states into positive and negative. For this scale, Cronbach's alpha of .82 and .86 were Obtained.for'Mood.1 and Mood 2, respectively. The scale below consists of 6 items representing the pleasure dimension, and has a seven-point Likert scale for each item. 70 Melancholic : : : : : : Contented Satisfied : : : : : : Unsatisfied Bored : : : : : : Relaxed Happy : : : : : : Unhappy Annoyed : : : : : : Pleased Hopeful : : : : : : Despairing Scores for each item range from -3 to +3. Therefore, the sum of the scores of all 6 items ranges from ~18 to +18. In differentiating a positive mood from a negative mood, scores ranging from -18 to 0 were regarded as a negative mood and scores ranging from 1 to +18 were classified as a positive mood. In this study, mood was measured two times: when shoppers entered the mall (Mood 1), and when shoppers exited the mall (Mood 2). To measure Mood 2 (mood when shoppers exited the mall), the same 6 items were used with the order changed randomly from the initial order used to measure Mood 1. 3. Impulse Purchasing Rate To measure the impulse purchasing rate, each subject was asked to list items and the quantities of each item purchased in the 71 following three categories - "Items Planned for Today," "Items Planned Before But Not for Today," and "Items Not Planned At All." So far, most past studies have classified level of planning into two groups; "Items Planned for Today" and "Items Unplanned for Today." "Items Unplanned for Today" has been regarded as those items bought on impulse. However, considering the fact that some items unplanned for "today" could be those planned before and recalled during today's shopping, all the items classified into "Items Unplanned for Today" are not necessarily items bought on impulse. With this understanding, this study created another category, "Items Planned Before But Not for Today," which should be treated as nonimpulse items. According to these three categories of planning, items which belonged to the "Items Planned for Today" category were classified as planned purchases, and items belonging to either of the other two categories ‘were regarded. as unplanned purchases. Finally, among the items regarded as the unplanned purchases, only items which belonged to "Items Not Planned At All" category were classified as impulse purchases. This classification among planned purchases, unplanned purchases, and impulse purchases is shown in Figure 2. In other words, this study tried to overcome the problem of using unplanned purchases as impulse purchases for the purpose 72 Unplanned Purchases Planned Purchases Impulse Purchases Impulse Purchses as a Subset of Unplanned Purchases Figure 2 of operationalization in all the past studies and, instead, regarded impulse purchases as a subset of unplanned purchases. In other words, impulse purchases are those purchases which are not planned for a specific shopping trip, and which have never been planned before. Based on this classification, the ratio of impulse purchasing was measured by the ratio of the number of items belonging to the "Items Not Planned At All" category, over the total number of items bought. Therefore, the impulse purchasing ratio ranged from 0 to 1. Here,"0" meant none of the purchases was made on impulse, and "1" meant all the items purchased were 73 bought impulsively. SURVEY 1. Questionnaire .A four—page questionnaire was developed using measures for shopper type, mood, and impulse purchasing rate mentioned in the previous section (Appendix B). The questionnaire consisted of two parts. The first part was answered when shoppers entered the mall and the second part was completed when they exited. The first part consisted of measures for shopper type, mood, and items they planned to buy. The measure for shopper type was a 5—item scale focusing on the shoppers' enjoyment of shopping, and their intention to spend time on shopping as described in the earlier section. In terms of the mood scale, even though only 6 items measuring the pleasure dimension were relevant to this study among the total 18 PAD items, this questionnaire contained all 18 items of PAD scale for the purpose of further study in the future. The last section of the first part of the questionnaire measured the items, and the quantity of each item, shoppers planned to buy before entering the shopping mall. 74 The first part of the questionnaire was completed by the shoppers when they entered the mall. Also, information on shopper type, mood upon starting shopping (Mood 1), and items and quantities planned, were collected using this first part of the questionnaire. The second part of the questionnaire consisted of measures for mood, demographic information, unplanned purchasing rate, and impulse purchasing rate. The same PAD scale with randomly rearranged order was used to measure the mood when shoppers exited the mall (Mood 2). Demographic information such as age, income, marital status, and sex were collected. The last section of the second part measured what shoppers purchased. In this section, shoppers were asked to list the items and the number of items actually bought and, also, to categorize them according to the levels of planning as described in detail in the measurement section. The second part of the questionnaire was to be completed when shoppers exited after shopping. Accordingly, shoppers' responses were collected at the time of their exiting mood (Mood 2) . Demographic information, the rates of unplanned purchasing, and those of impulse purchasing were collected at that time as well. 75 2. Description of Survey 1) Place The survey was conducted at the Meridian Mall located in Okemos, IMichigan. The mall is a typical closed. mall, accommodating many retail stores. The survey was conducted inside a major entrance, which had a large parking lot in front of it. Two students were employed to ask shoppers if they would participate in the survey. Tables and chairs were provided for shoppers to use while completing the questionnaires. As the shoppers entered the mall, they were approached by the students, and asked to participate in the survey with.the explanation that the survey was about.consumer behavior. After completing the first part, they were asked to go about their usual shopping, and then to return afterward, to complete part two. Because the shoppers parked in front Of the entrance where the survey was being conducted, 85 per cent of the shoppers who participated in part one did come back to complete the second part. 2) Time The survey was administered for ten days, weekdays and one weekend. It began in the early afternoon and lasted until nine in the evening-the mall's closing time. Therefore, data were collected during six to seven hours, everyday except 76 Sunday. Because there was a promotional event on that Sunday, the survey was not allowed on that Sunday by the mall office. There was no promotional event at all during the rest of the ten day period, which might have influenced the results of this survey. Thus, all ten days were just ordinary days for this mall. 3) Procedure As the shoppers entered the mall through the entrance, they were approached and asked to participate in the survey. At this time, they'were also instructed.that the survey’was about consumer behavior, and that it had two parts. Approximately one out of ten shoppers approached.participated in.the survey. As the shoppers sat at the tables, they were instructed to answer the first two pages of the questionnaire and then to come back and finish the remaining two pages, which they did not see. In order to find the correct questionnaire when shoppers returned to complete the second part, shoppers were asked to write their initials on the first page. After finishing the first two pages, the shoppers handed in their questionnaires, and started shopping. When the shoppers came back, they answered the remaining two pages without additional instructions. Shoppers were not given any form of compensation for their participation. They left the mall upon completion of the questionnaires. 77 SAMPLE During the ten days in which the survey was conducted, a total of 555 completed questionnaires were collected. Out of 555 subjects, 236 were male, and 319 were female. Age distributions were: 20 and below, 132; 21-30, 184; 31-45, 173; above 45, 66. Income proportions(in $1000's) were: below 7, 192; 7-25, 168; 25-45, 130; above 45, 65. As for marital status, 322 were singles, 54 were married without child, and 179 were married with children. In terms of subjects, adult shoppers entering the mall were approached indiscriminately. Accordingly, data were collected from the various types of shoppers and, therefore, the sample was presumably representative of all types of shoppers using a mall. RESULTS The data were analyzed using ANOVA, Duncan procedure of multiple range test, and t-tests. ANOVA was performed for Hypothesis I in order to analyze the effects of shopper type and mood. For Hypohtesis II and III, Duncan procedure of multiple range test and t-tests were employed. For Hypothesis IV and V, t-tests were used to compare group means. In this chapter, details of the results for each hypothesis are described. HYPOTHES IS I RESULTS Hypothesis.I was: Shopper type is more influential on impulse purchasing behavior than mood. For this hypothesis, the effects of the two variables, shopper type and mood, on impulse purchasing behavior were compared. Since there were two variables and each variable had two levels, a two-factor ANOVA was used for the test of this hypothesis. The results provided the information on main effects of shopper type and mood, respectively, and interaction effects between the two variables as shown in Table 1. The results reveal that the P value for main effects of 78 79 Effects SS DF MS F P Value Shopper Type 0.340 1 0.340 2.946 0.087 Mood 0.051 1 0.051 0.439 0.508 (Shopper'Iype) x 0.400 1 0.400 3.474 0.063 (Adood) Table 1: ANOVA Results of Hypothesis 1 shopper type is 2.946 with its p-value of 0.087. Therefore, it is found that there is a "weak" main effect of shopper type on impulse purchasing rate. The interpretation of the results is that the two different shopper types - recreational and economic - have different influences on impulse purchasing behavior, even though the difference is not strong. For the main effects of mood state, results show an F value of 0.439 and a p-value of 0.508. It is interpreted that there is no significant main effect of mood state on impulse purchasing rate. This findings lead to the conclusion that the two mood states do not have different influences on impulse purchasing behavior . In terms of the interaction effects between shopper type and mood, the results show an F value of 3.474 and a p-value of 0.063, repectively. Therefore, a marginally significant 80 Shopper Type R E 0.23 . . + 0 15 0 21 (334) (139) (473) Mood 0.20 0.28 0.23 _ (48) (34) (82) 0.23 0.18 0.21 *7 (382) (173) (555) ( ) : Cell Size Table 2 : Cell Means IPR 0'30 _ E (0.28 ) R ( 0.23) 0.25 — 0.20 ‘ R ( 0.20 ) 0.15 -L \ T E(0.15) l l + _ MOOD Fig. 3 : Interaction between Shopper Type and Mood 81 interaction effect between two variables appears to exit. The cell means and the graphical illustration of interaction effect between shopper type and mood are shown Table 2 and Figure 3, respctively. As shown on Figure 3, since there is an interaction effect between shopper type and mood state, the two lines, representing recreational shoppers and economic shoppers, are not parallel. That is, the mean impulse purchasing rate of recreational shoppers (0.23) is higher than that of economic shoppers (0.15) in a positive mood, but it (0.20) is lower than that of economic shoppers (0.28) in the negative mood condition. Findings from the results demonstrate that shopper type has a weak influence on impulse purchasing rate and mood does not exert a strong influence on impulse purchasing rate. Therefore, it is concluded that shopper type is more influential on impulse purchasing behavior than mood as hypothesized. HYPOTHESIS II RESULTS Hypothesis II was: Recreational shoppers buy more impulsively than economic shoppers under the same mood state. 82 In this hypothesis, impulse purchasing rates of recreational shoppers were compared with those of economic shoppers under each mood state. Figure 4 shows tests for Hypothesis II. For the positive mood, among the shoppers who entered the mall under a positive mood state (Mood 1, positive), recreational shoppers were compared with economic shoppers in terms of For Positive Mood Mood 1 Shopper Type test 1 Ikmbkgmnwdwomi Nkndl smnmefnnn test 2 'RmmtffhmommMSni Ifigne4 83 impulse purchasing rate. The same comparison was made for the negative mood state. Two t-tests (one for the positive mood and. one for’ the :negative :mood) were ‘used. to test. this hypothesis. First, the results for the positive mood state are shown in Table 3. For this state, the results show that the impulse purchasing rates of recreational shoppers are significantly greater than those of economic shOppers. The t value is 2.35 and the p-value is 0.0095. This difference is statistically significant. Shopper Type X n I value DF p value Rec. 0.23 334 2.35 471 0.0095 Eco. 0.15 139 Table 3: Results of Hypothesis II / Positive Mood Consequently, the results show that recreational shoppers who enter the mall under a positive mood state buy more impulsively than economic shoppers who enter the mall under the same mood state as hypothesized. Secondly, the results for the negative mood state are shown in Table 4. For the negative mood, the results show no 84 Shopper Type X n t value DF p value Rec. 0.20 48 -0.94 80 0.1745 Eco. 0.28 34 Table 4: Results Of Hypothesis II / Negative Mood difference in impulse purchasing rates between recreational shoppers and economic shoppers. The t value is -0.94, and the p-value is 0.1745. Therefore, for the negative mood state, Hypothesis II is not supported. In addition to the t-tests, Duncan procedure of multiple range test at alpha 0.1 was performed to compare the group means for Hypothesis II. Figure 5 shows the results of this test. First, the test of group 1 and group 2 is for a positive mood and the test of group 3 and group 4 is for a negative mood. The results of the multiple range test show group 1 and group 2 are significantly different in their means. And, the test shows no difference between group 3 and group 4. That is, the results show the higher impulse purchasing rates of recreational shoppers under a positive mood state, and no difference in impulse purchasing rates between the two shopper 85 Shopper Type R E + Group 1 Group 2 0.23 0.15 Mood _ Group4 Group 3 0.20 0.28 G G G G R R R R P P P P Mean Group 1 2 3 4 0.2339 GRPl ** 0.1540 GRPZ 0.2753 GRP 3 * 0.1996 GRP4 (*) denotes pairs of groups significantly different at the 0.100 level (**) denotes pairs Of groups significantly difierent at the 0.050 level Fig. 5 : Multiple Range Test / Duncan Procedure types in a negative mood. Therefore, the results of both t- tests and multiple range test, are the same. In conclusion, it is found that, under a positive mood state, recreational shoppers buy more impulsively than economic shoppers as hypothesized. However, under a negative mood 86 state, there is no statistical difference between recreational shoppers and economic shoppers, in terms of impulse purchasing rate. Based on the findings of Hypothesis II, it can be seen that the results of comparing the two shopper types under a positive mood and under a negative mood are different. This difference can be explained by the interaction effects between shopper type and mood state found in Hypothesis I using ANOVA. That is, the mean impulse purchasing rate of recreational shoppers is the higher in a positive mood, and that of economic shoppers is the higher in a negative mood due to the interaction effects, which are shown as the nonparallel lines on Figure 3. HYPOTHESIS I II RESULTS Hypothesis III was: shoppers under a positive mood state buy more impulsively than those under a negative mood state within the same shopper type. This hypothesis compares impulse purchasing rates between two mood states within each shopper type. For recreational shoppers, impulse purchasing rates between positive Mood 1 and negative Mood 1 were compared, regardless of Mood 2. The same comparison was also made for economic shoppers, as shown in Figure 6. lior Recreational Shoppers—l Shopper Type ii @ a IPRl Q m test 1 Em Economic Shoppers] ll Shopper Type Mood 1 IPR3 test 2 IPR4 Tests of Hypothesis 111 Figure 6 In other words, a comparison between IPR 1 and IPR 2 was made for recreational shoppers and another comparison between IPR 3 and IPR 4 was made for economic shoppers. Accordingly, two t-tests were performed, one for each shopper type. First, the results for recreational shoppers are shown on Table 5. For recreational shoppers, the t value was 0.62 and the p- 88 Moodl X n t value DF p value Pos. 0.23 334 0.62 380 0.268 Neg. 0.20 48 Table 5: Results of Hypothesis III / Recreational Shopper value was 0.268. Therefore, there is no statistical difference in impulse purchasing rates between the two mood states for recreational shoppers. For economic shoppers, the results of the t-test show that the t value is -2.18, and the p-value is 0.0155 (Table 6). This means that shoppers under a negative mood state buy more MOOdl I? n t value DF p value Pos. 0.15 139 -2.18 171 0.0155 Neg. 0.28 34 Table 6: Results of Hypothesis III / Economic Shopper impulsively than those under a positive mood state for economic shoppers, which is, interestingly, the opposite from 89 what was hypothesized. The results of Duncan procedure of multiple range test used for' Hypothesis II, ‘were also applied for the tests of Hypothesis III. As shown on Figure 5, the results of multiple range test indicate the significant difference between the two mood states for economic shoppers (group 2 and group 3), and no difference between the two mood states for recreational shoppers (group 1 and group 4). The results for Hypothesis III, using multiple range test, are also the same as those of t-tests for the same hypothesis. This hypothesis was based on the findings and implications on the relationship between mood and behavior in psychology. The key idea of the relationship is that people under a positive mood buy more impulsively than those under a negative mood. However, the results of this study indicate the Opposite of those findings and implications on the issue of impulse purchasing behavior in relation to mood state. In conclusion, for Hypothesis III, it is found that there is no statistical difference in impulse purchasing rates between two mood states for recreational shoppers, and that, for economic shoppers, shoppers under a negative mood state show more impulsiveness than those who are under a positive mood state. 90 Based on the results of the two comparisons of Hypothesis III, one for recreational shoppers and the other for economic shoppers, it is found that the results of comparing the two mood states are not consistent between the two shopper types. This inconsistency of the two results, again, can be explained by the interaction effects between shopper type and mood, as discussed previously and shown on Figure 3. HYPOTHES IS IV RESULTS Hypothesis IV was: among the shoppers who start in a positive mood (Mood 1, Positive), those who exit in a positive mood (Mood 2, Positive) buy more impulsively than those who exit in a negative mood (Mood 2, Negative) within the same shopper type. This hypothesis compares impulse purchasing rates between two mood states when shoppers exit the shopping environment (Mood 2) after having entered initially with a positive mood state (Mood 1, Positive) for each shopper type. For recreational shoppers who entered the shopping environment in a positive mood, a comparison was made between the shoppers who exited the mall in a positive mood (Mood 2, Positive) and those who exited in a negative mood (Mood 2, Negative). Also, the same comparison was made for economic shoppers who entered [Br Recreational Shoppgl Shopper Type Mood l Mood 2 03 0 '6 E test 1 09 E I For Economic Shoppers Shopper Type M_ood 2 IPR3 test 2 IPR4 Tests of Hypothesis IV Figure 7 the mall in a positive mood. This comparison is shown in Figure 7. For recreational shoppers, the t-test between IPR l and IPR 2 was performed and, similarly, a comparison between IPR 3 and IPR 4 was made for economic shoppers. The results for recreational shoppers are shown in Table 7. 92 Mood 2 I" n I value DF p value Pos. 0.25 281 1.73 332 0.042 Neg. 0.16 53 Table 7: Results of Hypothesis IV / Recreational Shopper The results from the t-test show a t value of 1.73 and a p- value of 0.042. Therefore, it is concluded that, of those recreational shoppers who enter the mall in a positive mood, those who exit in a positive mood buy more impulsively than those who exit in a negative mood, as hypothesized. Next, the results from the t-test for economic shoppers are shown in Table 8. For economic shoppers, it turns out that there is no statistical difference between the two exiting mood states, in terms of impulse purchasing rate, with a t value 0.89 and a p-value 0.1875. That is, among economic shoppers who entered the shopping environment in a positive mood, those shoppers who exited the mall in a positive mood did not show higher impulse purchasing rates than those in a negative mood . When analyzing the two results of the two shopper types together, it is concluded that, among the recreational 93 Mood 2 f n I value DF p value POS. 0.17 109 0.89 137 0.1875 Neg. 0.11 30 Table 8: Results of Hypothesis IV / Economic Shopper shoppers who entered the mall in a positive mood, those who exited in a positive mood buy more impulsively than those who exited in a negative mood, and, among economic shoppers who entered the mall in a positive mood, there was no difference in the rates between the two exiting mood states. HYPOTHESIS V RESULTS Hypothesis V was: among shoppers who start in a negative mood (Mood 1, Negative), those who exit in a positive mood (Mood 2, Positive) buy more impulsively than those who exit in a negative mood (Mood 2, Negative) within the same shopper type. This hypothesis is similar to Hypothesis IV, except for the negative entering mood (Mood 1, Negative). That is, Hypothesis V compares impulse purchasing rates between the two mood states when shoppers exit the shopping environment (Mood 2), among those who entered in a negative mood state (Mood l, I For Recreational ShoppchI Shopper Type Mood 1 Mood 2 O3 9 test 1 e B 09 E I For Economic Shoppgl Shopper Type Mood 1 Mood 2 O O e a test 2 09 ‘3 52 Tests of HypothesisV Figure 8 Negative) within each shopper type. This comparison is shown on Figure 8. A comparison between IPR 1 and IPR 2 was made for recreational shoppers, and a comparison was performed between IPR 3 and IPR 4 for economic shoppers. For this hypothesis, two t-tests were performed: one for recreational shoppers and one for economic shoppers. For 95 recreational shoppers, the test results showed a t value of - 0.02 and a p-value of 0.493 (Table 9). Mood 2 f n I value DF p value Pos. 0.20 18 -0.02 46 0.493 Neg. 0.20 30 Table 9: Results of Hypothesis V / Recreational Shopper The test results showed the t value to be very close to zero, and the p-value to be large. From this, it is concluded that, among the recreational shoppers who entered the mall in a negative mood state, there is no statistical difference in impulse purchasing rates between shoppers who exited in a positive mood and those who exited in a negative mood. Mood 2 X- n I value DF p value Pos. 0.25 12 -0.28 32 0.3915 Neg. 0.29 22 Table 10: Results of Hypothesis V / Economic Shopper 96 The second test of economic shoppers revealed similar results, as shown in Table 10. For economic shoppers, the results were a t value of -0.28 and a p-value of 0.391. Again, there was no statistical difference between the two groups. In other words, among economic shoppers who entered the shopping environment in a negative mood state, there was no significant difference in impulse purchasing rates between those who exited the mall in a positive mood, and those who exited in a negative mood. Thus, results from the two tests for Hypothesis V show no statistical difference in impulse purchasing rates between the two exiting mood states (Positive and Negative Mood 2), among shoppers whose entering mood was negative (Negative Mood 1). The results for all 5 hypotheses are shown in Figure 9. HYPOTHESIS I Shopper > HYPOTHESIS II Mood 1 Shopper Type For Positive Mood 3 t = 2 35 \/ p = 0.0095 Q t = 094 For Negative Mood e u p = 0.1745 HYPOTHESIS III For Recreational Shoppers For Economic Shoppers HYPOTHESIS IV For Recreational Shoppers For Economic Shoppers HYPOTHESIS V For Recreational Shoppers For Economic ShOppcrs 97 Shopper Type Mood 1 é a t = 0.62 p = 0.268 t = -2.18 /\ p = 0.0155 Shopper Type M 1 Mood 2 t = 0.173 1:089 ® + ' ' p = 0.1875 Shower Type M 1 Mood 2 = 002 ® — ' ' p = 0.493 t ----0.28 CE) — ' ' p = 0.3915 Summary of the Results Figure 9 DISCUSSION This chapter consists of 3 sections. The first part deals with further discussion, implications, and speculation based on the results of the five hypotheses. The second section is devoted to the limitations of this study. Finally, in the third part, potential future research areas are considered. DISCUSSION OF RESULTS The average impulse purchasing rate found in this study is 21 per cent. However, the rates found in the past studies range from.3.8 per cent (Clover 1950; using furniture store) to 62.4 per cent (Prasad 1975; using discount store). 'These rates are not easily comparable, since stuies used different types of stores to collect data. Grocery stores, in general, have shown relatively higher rates. Because this is the first study using a closed mall setting, the rate found is not comparable to any previous studies. However, it appears that the impulse purchasing rates of shoppers at the mall are relatively low. The results of Hypothesis I indicated a weak main effect of shopper type and a weak interaction effect of shopper type and mood on impulse purchasing rate. However, no significant main 98 99 effect of mood state was found. In Hypothesis I, the shopper type was hypothesized to be more influential on impulse purchasing behavior than on mood. In fact, it was not easy to give a priori directionality for this comparison since none of the past studies has dealt with these two variables at the same time. As mentioned earlier, the directionality of the comparison that shopper type is more influential than mood is given simply because shopper type is more permanent and stable in nature than.mood. However, from the results of Hypothesis I, an influence of shopper'type on impulse purchasing rate was found and, furthermore, the interaction effect of shopper type and mood was discovered, as shown in Figure 3. The, results of ZHypothesis II revealed. that recreational shoppers bought more impulsively than economic shoppers, for those entering the mall in a positive mood state, and that there was no difference between the two shopper types in a negative mood state. There was, however, a difference in the two results, one for the positive mood and one for the negative mood” .AS'mentioned earlier, this difference is due to the interaction effects of shopper type and mood on impulse purchasing rates as shown in Figure 3, ANOVA results. Secondly, the.higher impulse purchasing rate for recreational 100 shoppers for a positive mood state can be explained by the automatic process and controlled process, mentioned in the theory section earlier. The theory of automatic process and controlled process states that people in a positive mood state (automatic process) tend to maintain and even to enhance their positive mood state through the behaviors related to positive mood (controlled process). That is, recreational shoppers who entered the mall in a positive mood state tried to maintain and enhance their positive mood. Consequently, they were more likely to make impulse purchases. The results of Hypothesis III revealed that economic shoppers who entered the mall in a negative mood bought more impulsively than those in a positive mood, and that there was no difference between the two mood states for recreational shoppers. By comparing the two results, each for each shopper type, it was, again, found that they were not consistent. This difference is, also, considered as the result of interaction effects between shopper type and mood. Meanwhile, the higher impulse purchasing rate under a negative mood state for economic shoppers can be explained, again, by the theory of automatic process and controlled process. For a negative mood state, the theory states that people who are 101 in a negative mood state try to overcome and change their negative mood into a positive one by performing behaviors associated with positive feelings. That is, the higher impulse purchasing rate of economic shoppers under a negative mood state can be regarded as the result of a strong effort to overcome and change their negative feelings to positive ones. For economic shoppers in general, staying in the shopping environment is not enjoyable to them, by definition. In short, economic shoppers in a negative mood and in a shopping environment.are assumed to be in the worst mood state for them (automatic process) and, accordingly, the strength of their conscious efforts to convert their negative feelings into positive ones (controlled process) are strong. Consequently, they make more purchases on impulse. So far, the results of the first 3 hypotheses were discussed separately. However, a conclusive discussion from the results of all 3 hypotheses together can be made using a simple 2 by 2 table as shown in Figure 10. This table can be used to show the relationship among shopper type, mood, and impulse purchasing behavior. This table has two columns (recreational and economic shoppers) and two rows (positive and negative mood states). The signs inside the table show the comparisons of impulse purchasing rates in four cells using Duncan procedure of 102 multiple range test and the t-tests. The positive and negative signs for mood indicate positive mood and negative mood, respectively. For the discussion of the results, (+) and (-) signs were added to recreational and economic shoppers, respectively. Considering the way these two shopper types were classified, (+) means that the influence of a ShopperType R(+) E(-) Cell1(+,+) Ce112(+,-) ._ a _ (+) X11 =0~23 X12 =0.15 S=0.36 S=O.27 Mood —— I l /\b— (-> X21 = 0.20 _— 3:22 = 0.28 3:037 3:035 Cell4(-,+) Cell3(-.-) a : significant at 0.05 b : significant at 0.10 Interaction Effects of Shopper Type and Mood on Impulse Purchasing Behavior Figure 10 103 shopping environment on recreational shoppers, is more likely to be positive, and (-) implies that the influence of the same environment on economic shoppers is more likely to be a negative one. The results from Hypothesis II indicate the following relationships; cell 1 is greater than cell 2, and cell 3 is equal to cell 4. The results from Hypothesis III show that cell 1 is equal to cell 4, and cell 3 is greater than cell 2. Again, cell 1 and cell 3 are explained by the theory of automatic process and controlled process. Shoppers belonging to cell 1 are those who are in a positive mood and in a positive environment (+,+). This is based on the theory that they tend to maintain this strong (+,+) positive feeling by making purchases on impulse. Consequently, cell 1 shows higher impulse purchasing rates. Meanwhile, shoppers categorized in cell 3 are those in a negative mood and in a negative environment (-,-) . They possess strong negative feelings and, therefore, behave in a ‘way that makes them feel better. Consequently, in an effort to change their feelings, they are more likely to make purchases on impulse and show higher impulse purchasing rates. 104 The finding for cell 3 is important in terms of the interaction.effect of shopper type and mood state, in relation to impulse purchasing. It is, in fact, the exact opposite of findings and implications from past studies, which have dealt with shopper type and impulse purchasing, or mood and impulse purchasing, separately. In past studies, recreational shoppers have consistently been regarded as those who buy more impulsively, and, in terms of mood state, shoppers under a positive mood state have been thought to be more impulsive under a shopping situation. In other words, this finding, which is the exact opposite of past implications, was obtained for the first time, by considering both shopper type and mood together, in studying impulse purchasing behavior. The unexpected high impulse purchasing rate of this cell was further analyzed, using a frequency distribution of the rates of 34 shoppers who belong to the cell. Frequency shows 18 zeros, 4 ones, and the remaining 12 between 0.25 and 0.75. Accordingly, the unexpected high rate of this cell does not seem to be due to a few extremely high outliers. Cell 2 is explained by a "balancing out effect" between positive and negative forces in the cell. For the shoppers belonging to cell 2, their initial positive mood (+) is balanced out by the negative environment (-) for economic shoppers. Accordingly, there seems no particular reason for them to perform any specific behavior to maintain or change 105 their feelings. Consequently, the impulse purchasing rates for this cell are lower than cell 1 and cell 3. Finally, cell 4 shows no statistical difference from the other three cells. This cell is supposed to be the same as cell 2 in terms of the balancing out effect, in that there are two opposite forces working within the cell - a negative mood, and a positive environment for recreational shoppers. However, the results of this cell do not show a strong "balancing out effect" like cell 2, and, accordingly, the impulse purchasing rates of this cell are not lower than those of cell 1 or cell 3. In summary, this 2 by 2 table based on findings from Hypothesis I, II, and III, leads to the conclusion that (l) recreational shoppers(+) , who are in a positive mood(+) , (cell 1) , show high impulse purchasing rates in an effort to maintain their positive feeling, and (2) that economic shoppers(-), who are in a negative mood state(-), (cell 3), demonstrate high impulse purchasing rates in an effort to change their negative mood state into a positive one, and (3) 'that economic shoppers(-), in a positive mood state(+), (cell 2), reveal lower impulse purchasing rates than cell 1 and.cell .3, presumably, due to a balancing out effect, and (4) that recreational shoppers(+) , in a negative mood(-) , (cell 4), show no difference from other cells in impulse purchasing 106 rates. The results for Hypothesis IV showed a higher impulse purchasing rate for recreational shoppers who exited the mall in a positive mood (Hood 2, Positive) than for those who exited the mall in a negative mood (Hood 2 , Negative). However, for economic shoppers, there was no statistical difference between the two exiting mood states. The higher rates for recreational shoppers who exited the mall in a positive mood state (Hood 2, Positive) rather than those exiting in a negative mood (Mood 2, Negative) can be interpreted in the following way: shoppers in this category are recreational shoppers, who both start and finish shopping in a positive mood state. They are likely to make more purchases on impulse because they are in a positive mood throughout the period, unlike those who exit in a negative mood. Here, it seems appropriate to think about possible reasons why some recreational shoppers exited the mall in a negative mood even though they started shopping in a positive mood. In fact, data showed 16 per cent of recreational shoppers who entered the mall in a positive mood, exited the mall in a negative mood. The first reason can be found from one of the common phenomena people feel after making a purchase 107 impulsively. It has been discovered in past studies that shoppers sometimes feel guilty after making an impulse purchase (Rook and Hock 1985; Rock 1987). This may be one reason why some shoppers had a negative feeling when they exited the mall. Another reason might be the financial constraint shoppers experienced during shopping, which failed to counteract their strong desire to make an impulse purchase. Additionally, there could have been other unknown and individual reasons leading shoppers to feel negative upon leaving the mall. The second test for Hypothesis IV dealt with the same comparison between the two exiting mood states, but for economic shoppers. Based on the results, it can be concluded that there is no statistical difference for the comparison between the two exiting mood states for the shoppers in this category. However, it seems appropriate to discuss the plausible mechanism of how shoppers exited the mall in the two different mood states. For the shoppers who exited the mall under the positive mood, it seems that their positive starting mood (+) and the mood enhanced by impulse purchasing (+) outweighed their being economic shoppers (-). According to the data, 78 per cent of economic shoppers who entered the mall in a positive mood belongs to this category. The same reasoning can be made for the shoppers who exited under the negative mood state. For this group of shoppers, it seems :- ——__--—.—- 108 that their being economic shoppers (-) outweighed the positive influences from the positive starting mood (+) and making impulse purchases (+) . The remaining 22 per cent of the economic shoppers in a positive entering mood.was categorized into this group. In short, the two different exiting mood states of this group appear to have beeen determined by the strengths of two forces (+,-) working in opposite directions, even if they make the same level of impulse purchasing. Hypothesis V is similar to Hypothesis IV, in that it compares the two exiting mood states. But Hypothesis V tests those shoppers whose starting mood was negative (Mood 1, Negative). Results show no statistical difference in impulse purchasing rates for either recreational shoppers or economic shoppers. Interpretation of the results of the two tests, for recreational shoppers and for economic shoppers, can be made through the same reasoning made for Hypothesis IV. For the recreational shoppers who exited in a positive mood, it seems that the jpositive influences of being recreational shoppers(+), and making impulse purchases(+), outweighed the negative influence of their starting in a negative mood(-). Data reveal that, among the recreational shoppers in a negative entering mood, 38 per cent exited in a changed positive mood. Meanwhile, for the group of shoppers who exited in a negative mood state, it seems that the negative 109 influences from the starting negative mood(-) outweighed the positive influences from making impulse purchases(+) and their being recreational shoppers(+). Sixty two per cent of the recreational shoppers in a negative entering mood, stayed in the same negative mood when they exited. Again, even if these two groups make the same level of impulse purchasing statistically, the exiting mood state seems to be the result of the relative strengths of positive and negative forces. For economic shoppers, the group of shoppers who exited in a positive mood were those who had made impulse purchases and successfully converted their initial negative mood into a positive one. Actually, 35 per cent of the economic shoppers in a negative entering mood, successfully, changed their mood and exited in a positive mood. For this group of economic shoppers, it seems that the positive influence from making impulse purchases(+) outweighed the negative influences of being economic shoppers(-) and initial negative mood state(— ). Meanwhile, the group of shoppers who exited in a negative mood state were those who tried to change their negative feeling into a positive one by making impulse purchases, but, being unsuccessful, ended up with a negative mood state. In other words, the negative forces from being economic shoppers and their initial negative mood state were stronger than the positive influence originating from making impulse purchases. This category of shoppers accounts for the remaining 65 per 110 cent of the economic shoppers in a negative entering mood. LIMITATIONS OF THE STUDY The major limitation of this study is that the survey was conducted in a closed mall setting. Accordingly, the findings are limited to impulse purchasing behaviors at the shopping mall. Relating to this limitation is the fact that there were more recreational shoppers (382) than economic shoppers (173), and more positive mood shoppers (473) than negative (82), in the data collected. This phenomena seems typical for a closed mall setting. The proportions of shopper type and mood state may have been different if the data were collected at different types of stores, such as grocery stores, and discount stores, since shopping malls have relatively more activities and facilities in addition to the shops for just purchasing items. Consequently, it seems appropriate to limit the generalizations of the findings of this study to within the boundary of closed shopping malls. FUTURE RESEARCH AREAS There are several possible future areas of research related to this study. First, considering the limitations of this 111 study, it is worth replicating this study, using different types of stores, especially grocery stores, and comparing the results. This would greatly improve the generalizability of findings on the issue of impulse purchasing behavior. Secondly, this study tried.to differentiate impulse purchasing from unplanned purchasing in an operational manner. This is a first attempt to try to differentiate between these two very similar concepts. However, it appears that more studies should be done on refining the concept, and the measurement, of impulse purchasing behavior. Thirdly, this study considered two influential factors (shopper type and mood) in studying impulse purchasing behavior. There are, however, other variables which might have influence on impulse purchasing behavior such as the attitude of a sales person, the time of shopping, etc. Therefore, other influential variables could be included in a future study, to understand this type of consumer behavior more comprehensively. Furthermore, considering the lack of established conceptual models in this area, more studies should attempt to develop a sound conceptual model in relation to the influential variables with the theories justifying the model. For this purpose, more relevant theories and models developed in psychology should be tested within the context of marketing for their applications. 112 Finally, this study is an attempt to learn more about the affective aspects of consumer behavior. Considering at least two facts that relatively little attention has been given, in the past, to these particular aspects of consumer behavior and that people have both cognitive and affective aspects, more studies on the affective aspects of consumer behavior would produce a better and balanced understanding of overall consumer behavior from the longrun point of view. REFERENCES Ainslee, Grorge (1975), "Special Reward: A Behavioral theory of Impulsiveness and Impulse Control," Psyghglggigg; Bulletin, 82(4), 463-496. Applebaum, William (1951), "Studying Customer Behavior in Retail Stores," Journal of Marketing, 16, Oct., 172-178. Assael, Henry (1987), Consumer Behavigr and Markgting Acrion, 3rd., Kent Publishing Co., Boston MA. Axelrod, Joel N. (1963), "Induced Moods and Attitudes Toward Products," Journal of Adverrising Research, 19-24. Baker, Roger G. (1975), "Commentaries on Belk, "Situational Variables and Consumer Behavior," Journal of Consumer Research, Vol 2, Dec., 165-167. Barnes, Nora Ganim (1984), "New’Shopper Profiles: Implications of Sunday Sales," Journal of Small Business Management, Juyl, 32-39. Batson, C. Daniel, Jay S. Coke, Fred Chard, Debra Smith, and Antonia Taliaferro, (1979), "Generality of the "Glow of Good Will,": Effects of Mood on Helping and Information Acquisition," Social Psychology Quarterly, 42, 176-179. Belk, Russell W. (1974), "An Explanatory Assessment of Situational Effects in Buyer Behavior," Journal of Marketing Research, May, Vol XI, 156-163. (1975), "Situational Variables and Consumer Behavior," Journal of Co sumer Behavior, Dec. Vol.2, 157-164. (1984), "Applications of Mood Inducement in Buyer' Behavior: Comments," Advances in Consumer Behgying Vol.11. Bellenger, Danny N. and Pradeep K. Korgaonkar (1980), "Profiling the Recreational Shopper," Jou al 0 Ret ' ' 56(3), 77-92. I . Dan H. Robertson, and Barnett A. Greenberg J2_rnal__2f (1977), "Shopping Center Patronage Motives," u Retailing, Vol.53, Summer, 29-38. . and Elizabeth C. Hirschman (1978), "Impulse Buying Varies by Product," gggrngl sELAdxerfising_Be§earch. 15-18- 113 114 Bolch, P.H., N.M. Ridgway, and D.L. Sherrell (1989), "Extending the Concept of Shopping: An Investigation of Browsing Activity," Journal of Academy of Marketing Science, Vol.17, No.1, Winter. Bower, Gordon H. (1981), "Mood and Memory," Ameriggn Psychologist, Feb. Vol.36, No.2, 129-148. , Kenneth P. Monteiro, and Stephen G. Gilligan (1978), "Emotional Mood as a Context for Learning and Recall," Journal of Verbnl Learning and Verbal Behavior, Vol.17. (Siark, Margaret S. and Alice M. Isen (1982), "Toward ‘nderstanding the Relationship between Feeling States and Social Behavior," in Co n't've Social Ps cholo , eds. Albert Hastorf and Alice Isen, N.Y. Elserier (North Holland), 73—108. . Sandra Milberg, and John Ross (1983), "Arousal Cues Arousal-Related Material in Memory: Implication for Understanding Effects of Mood (n1 Memory," Journal of Verbal Learnin and Verbal Behavior, Vol.22. . and Barbara A. Waddell (1983), "Effects of Moods on Thoughts about Helping, Attraction, and Information Acquisition," Social Psychology Quarterly, 46(1), 31-35. Clover, Vernon T. (1950), "Relative Importance of Impulse Buying In Retail Stores," Journal of Marketing, July, 66-70. Cobb, Cathy J. and Wayne D. Hoyer (1986), "Planned vs. Impulse Purchase Behavior," Journal of Retailing, 67—81. Cox, keith (1964), "The Responsiveness of Food Sales to Shelf Space Changes in Supermarkets," Journal of Marketin Research, 1, May, 63- 67. /Darden, William R., and Dub Ashton (1975), "Psychographic ‘Profiles of patronage preference groups," Journal of Retailing, Vol.50, Winter, 99-112. . and F.D. Reynolds (1971), "Shopping Orientations and Product Usage Roles," Journal of Marketing Researcn, 8, 505-508. Deshpande, Rohit and S. Krishnan (1979), "Consumer Impulse Purchase and Credit Card Usage: An Empirical Examination Using the Log Linear Model, " in J. C. Olson (ed. ) Advances in Consumer Researgn, Ann Arbor: Association for Consumer Research, 7, 792—795. 115 Dickman and Meyer (1988), "Impulsivity and Speed-Accuracy Tradeoffs in Information Processing," Journal of Personality and Socigl Psychology, Vol.54(2), 274-290. Donovan, Robert J. and John R. Rossiter (1982), "Store Atmosphere: An Environmental Psychology Approach," Journal of Retailing, Vol.58, No.1, Spring. Engel, J.F., D.T. kollat, and R.D. Blackwell (1969), "Personality Measures and Market Segmentation," Business Horizons, June, 12, 61-70. Forest, D., M.S. Clark, J. Mills, and A.M. Isen (1979), "Helping as a Function of Feeling State and Nature of Helping Behavior," Mot'vation and Emotion, 3, 161-169. Gardner, Meryl P. (1985), "Mood States and Consumer Behavior: A Critical Review," Journal of Consumer Research, Vol.12. and Marion Vandersteel (1984), "The Consumer's Mood: An Important Situational Variable," Advances in Consumer Research, Vol.11. Gillette, Peter L. (1970), "A Profile of Urban In—Home Shoppers," Journal of Marketin , Vol.34, July, 40-45. Gutman, Jonathan and. Michael Mills (1982), "Fashion Life Style, Self Concept, Shopping Orientation and Store Patronage: An Integrative Analysis," Journal of Retailin , 58, Summer, 64-86. Hansen, Robert A. and Terry Deutcher (1978), "An Empirical Investigation of Attribute Importance in Retail Store Selection," Journal of Retailin , 53, Winter, 59-72. Havlena, William J. and Morris B. Holbrook. (1986), "The varieties of Consumption Experience: Comparing Two Typologies of Emotion in Consumer Behavior," Journal of Consumer Research, Vol.13, Dec., 394-404. Hilgard (1962), "Impulsive versus Realistic Thinking: An Examination of the Distinction between Primary and Secondary Process in Thought," Psychological Bulletin, 59(6), 477-488. Hirschman, Elizabeth C. and Morris B. Holbrook (1982), "Hedonic Consumption: Emerging Concepts, Methods, and propositions," Journal of Marketin , Summer, 92-101. Holbrook, Morris B. and Elizabeth C. Hirschman (1982), "The Experiential Aspects of Consumption: Consumer Fantasies, Feelings, and Fun," Journal or Consumer Research, Sept. 132— 116 140. Isen, Alice M. (1984), "The Influence of Positive Affect on Decision Making and Cognitive Organization," Advances in Consumer Research, Vol.11. , M. Clark, and M. F. Schwartz (1976), "Duration of the Effect of Good Mood on Helping: "Footprints on the Sands of Time," Jou n so al' t Social Ps cho o 34, 385-393. , P.F. Levin (1972), "The Effect of Feeling Good on Helping: Cookies and Kindness," Journal of Personality and Social Psychology, 21, 384-388. , Thomas E. Shalker, M.S. Clark, and L. Karp (1978), "Affect, Accessibility of Material in Memory and Behavior: A Cognitive Loop?" Journal of Personality and Social Psychology, 36, 1—12. , S.F. Simmonds (1978), "The Effect of Feeling Good on a Helping Task that is Incompatible with Good Mood," Social Psychology Quarterly, 41, 345-349. Jarboe, G.R. and C.D. McDaniel (1987), "A Profile of Browsers in Regional Shopping Malls," Journal of the Academy of Marketing Science, Spring, Vol.15, No.1, 46-53. Kassarjian, Harold H. (1971), "Personality and Consumer Behavior: A Review," Journal of Marketing Research, Vol.8, Nov., 409-418. Kollat, David. T. and 'Ronald. P. Willett (1967), "Consumer Impulse Purchasing Behavior," Journal of Marketing Research, Feb., 21-31. (1969), "Is Impulse Purchasing Really A Useful Concept in Marketing Decision?" Journal of Marketing, Jan. 79-83. Kotler, Philip (1974), "Atmospherics as a Marketing Tool, Journal of Retailin ng, Winter, Vol.49, No.4. Lutz, Richard J. and pradeep Kakkar(1975), "The Pshchological Situation as a Determinant of Consumer Behavior," Advances in Consumer Resaargh, Vol.1. Markin, R. J., C. M. Illis, and C.L. Narayana (1976), "Social- Psychological Significance of Store Space," Journal of Retailing, Spring, 52, 43- 54. (Moschis, George P. (1976), "Shopping Orientations and Consumer 117 Uses of Information," Journal of Retailing, Summer, 52, 61- 70. Peterson, Robert and Matthew Sauber (1983), "A Mood Scale for Survey Research," in 1983 AMA Educators' Proceedings, eds. Patrick Murphy et al., Chicago, IL: AMA, 409-414. Pollay, Richard W. (1968), "Customer Impulse Purchasing Behavior: A Reexamination," Journal of Marketin Research, Aug., 323-325. Prasad, V. Kanti (1975), "Unplanned Buying in Two Retail Settings," Journal of Reta'l'n , Fall, 51, 3-12. Reynolds, Fred D. and William D. Wells (1977), Consumer Behavior, lst ed., McGraw-Hill Book Co., N.Y. Rook, Dennis W. (1987), "The Buying Impulse," Journal of Consumar Resaarch, Sep., 189-199. and Stephen J. Hock (1985), "Consuming Impulses," in Advances in Consumer Research, Vol.12, eds. Holbrook and Hirshman, Provo, UT: Association for Consumer Research, 23-27. Russell, James A. and Albert Mehrabian (1976), "Environmental Variables in Consumer'Research," Journal of Consumer Research, 3, 62-63. Schachter, S. and J. Singer (1962), "Cognitive, Social and Psychological Determinants of Emotional State," Psychological Review, 65, 379-399. Schwarz, Norbert and Gerald L. Clore (1983), "Mood, Misattribution and Judgement of Well-Being: Informative and Directive Functions of Affective States," Journal of Personality and Social Psychology, Vol.45, No.3, 513-523. (Stephenson, Ronald P. and Ronald P. Willett (1969), "Analysis of Consumers' Retail Patronage Strategies," in P.R. McDonald (ed.), Marketing Involvement in Society and the Economy, Chicago, AMA, 316-322. Stern, Hawkins (1962), "The Significance of Impulse Buying Today," Journal of Marketin , April, 59-62. ~stone, Gregory P. (1954), "City Shoppers and. Urban ”Identification; Observation on the Social Psychology of City Life," Ame 'ca Journal of Sociolo , 60, July, 36-45. Tauber, Edward M. (1972), "Why Do People Shop?" Journal of 118 Marketing, Oct. 46-49. Teasdale, J.D. and S.J. Fogarty (1979), "Differential Effects of Induced Mood on Retrieval of Pleasant and Unpleasant Events from Episodic Memory," Journal of Abnormal Psychology, 88, 248-257. Weinberg, Peter and Wolfgang Gottwald (1982), "Impulsive Consumer Buying as a Results of Emotions," Journal of Business Research. Mar-, 43-57- Wells, G. (1986), "We're in the Habit of Impulse Buying," USA Today, (May 21). Wells, William D. and Leonard A. Lo Sciuto (1966), "Direct Observation of Purchasing Behavior," Journal of Marketing Research, Aug., 227-233. West, C. John (1951), "Results of Two Years of Study into Impulse Buying," Journal of Marketing, Jan., 362-363. /\ Lybstbrook, Robert A. (1980), "Intrapersonal Affective Influences on Consumer Satisfaction with Products," Journal of Consumer Research, 7(June), 49-54. C 2 and w.c. Black (1985), "A Motivation-Based Shopper Typology," Journal of Retailin , 61(Spring), 78-103. Wicker (1975), "Commentaries on Belk, "Situational Variables and Consumer Behavior," Journal of Consumer Research, Vol.2, Dec., 165-167. Willett, Ronald P. and David. T. Kollat (1968), "Customer Impulse Purchasing Behavior: Some Research Notes and a Reply," Journal of Markering Research, Aug., 326-330. Williams, J. and R. Dardis (1972), "Shopping Behavior for Soft Goods and Marketing Strategies," Journal of Retailin , 48(Fall), 32-41. Williams, R.H., J.J. Painter and H.R. Nichols (1978), "A Policy-Oriented Twpology of Grocery Shoppers," Journal of Retailing, 54(Spring), 27-34. Qfiilliams, Terrell, Mark Slama and John Rogers (1985), Behavioral Characteristics of the Recreational Shopper and Implication for Retail Management," Journal of the Academy of Marketing Science, Summer, Vol.13, No.3, 307-316. Zajonc, Robert B. (1980), "Feeling and Thinking: Preferences Need. No Inferences," Ame ican Ps cholo ist, Feb., Vol.35, 119 No.2, 151-175. and Hazel Markus (1982), "Affective and Cognitive Factors in Preferences," Journal of Consumer Researcn, Vol.9, Sept., 123-131. APPENDICES APPENDIX A CONSENT FORM With regard to my participation in survey: 1. I understand that this survey is about consumers' shopping behavior and I am expected to answer Part 1 now and finish Part 2 when I exit this mall. I understand that the total amount of time needed is about 3 minutes for Part 1 and another 3 minutes for Part 2. I understand that the participation is voluntary and I may discontinue at any time without penalty or loss of benefit. I understand that all results will be treated with strict confidence and the subjects will remain anonymous in any report of research findings. I understand that I may ask any questions regarding this survey during and after survey to Dong Lee (355-1202). Please go on to next page and start answering questions if you agree to participate in this survey. 120 APPENDIX B Questionnaire CONSUMER SHOPPING BEHAVIOR STUDY DEPARTMENT OF MARKETING AND TRANSPORTATION ADMINISTRATION COLLEGE OF BUSINESS ADMINISTRATION MICHIGAN STATE UNIVERSITY Thank you for participating in this survey. Your answers will be extremely useful in a study of shoppers we are conducting at Michigan State University. The responses you provide will be treated confidentially. There are two parts to the survey - Part 1 is to be answered now and Part 2 should be answered when you finish shopping. Both parts will only take a short amount of time to complete. Select the response that most closely matches your thoughts by placing an ”J” in the appropriate space. Please write your initials below so we can match both parts of the survey together. Thank you. Initials GENERAL SHOPPING ENJOYNENI Please place an ”I” in the appropriate space below. I generally enjoy shopping for gifts for other people. Strongly Strongly Agree : : : : Disagree I generally enjoy shopping for clothes. Strongly Strongly Agree : : : : Disagree I like to shop for groceries. Strongly Strongly Agree : : : : Disagree I enjoy going to stores to shop even when I am looking for nothing in particular. Strongly Strongly Agree : : : : Disagree I enjoy shopping in general. Strongly Strongly Agree : z : : Disagree YOUR MOOD AT TH MOMENT PLEASE INDICATE HOW YOU FEEL NON. Wide-awake _:___:__:_:_:_:._ Sleepy Melancholic __:_:_:_____:__:_:_ Contented Aroused __:____:____:_:_:__:____ Unaroused Influential : : : : : : Influenced 122 Awed _:+:_:__:__:_:__ Important Satisfied _____:_____:_____:_____:_____:_____:_____ Unsatisfied Autonomous _:_:_:__:___:_:_ Guided Bored _:_:__:_:_:_:_ Relaxed Happy _______ Unhappy Dominant _:___:__:__:_:___:__ Submissive Calm __:___:_:_:_:__:i__._ Excited Annoyed _____:_____:_____:_____:_____:_____:_____ Pleased Hopeful _______ Despairing Cared-for _:_:_:_:_:_:___ In control Relaxed ___:____:__:_:_:_:_ Stimulated Dull _:___:___:_:_:_:__ Jittery Frenzied _:_:__:_:__:_:_ Sluggish Controlled _____:_____:_____:_____:_____:_____:_____ Controlling Please list al_l_ the items and the number of each item you plan to buy on this shopping trip including any food items such as drinks or a meal. For example, if you plan to buy one pair of pants, one soft drink, and two greeting cards, you would list the following; EXAMPLE Item Qty. Item Qty. Item Qty. pants 1 drink 1 greeting cards 2 NON PLEASE LIST ALL THE ITEMS AND THE NUMBER OF EACH ITEM YOU PLANNED 0N TODAY’S SHOPPING TRIP. Item Qty. Qty. Qty. T A CD 0U A TO IL23 THANK YOU FOR COMING BACK. YOU M AT 0 PLEASE INDICATE HOW YOU FEEL N93. Influential Influenced Dominant Submissive Relaxed Stimulated Hide-awake Sleepy Autonomous Guided Annoyed Pleased Frenzied Sluggish Melancholic Contented Bored Relaxed Dull Jittery Calm Excited Hopeful Despairing Controlled Controlling Satisfied Unsatisfied Cared-for In control Aroused Unaroused Happy Unhappy Awed Important YOUR AGE? 20 and below _____, 21-25 _____, 26-30 _____. 31-35 _, 36-40 _, 41-45 _. 46-55 _____, 56-65 _____, above 65_____ YOUR INDIVIDUAL INCOME? below 7 _____, 7-15 _____. (in $1,000's) 15-25 _____, 25-35 _____. 35-45 ,_____, Above 45______ MARITAL STATUS? Single _____, Married/No Children _____ Married/Children_____ SEX? Male Female W- 124 Please list all the items you actually bought today. For each item, please specify the quantity you bought in the most appropriate category of planning. If you bought more than one of the same item and their nature of planning are different, please indicate the most appropriate categories by writing the quantity in each category. EXAMPLE For example. if you bought: a pair of pants which was planned on this shopping trip. a meal which has not been planned at all. and three greeting cards. two of which were planned on this shopping and one has been planned before. but not necessarily on this shopping trip today, youwill tllintheboxesasshownhelow. Items Planned Items Before But Items Not Flamed {or Not for Planned Today Today At All ESL pant-5 / dTnn er _L_ fires-ii” m1 Ca M5 74 / NOW PLEASE LIST ALL THEITEMS YOU BOUGHT AND WRITE THE QUANTITY OF EACH ITEM IN THE MOST APPROPRIATE CATEGORY OR CATEGORIES. . Items Planned Items Before But Items Not Flamed for Not for Planned Today Today At All Item