BELONGINGNESS AGAINST IMPULSE BUYING

BELONGINGNESS AGAINST IMPULSE BUYING

BELONGINGNESS, A CURE FOR NEGATIVE MOOD AND A

BUFFER AGAINST IMPULSIVE PURCHASE

INGINERU-VANCSA ANCA

BELONGINGNESS AGAINST IMPULSE BUYING

Belongingness, a Cure for Negative Mood and a Buffer against

Impulsive Purchase

INGINERU-VANCSA ANCA

Department of Marketing

Master Thesis

Date: August

2009

Address: str. 1 Decembrie 1918, bl.15, sc. I, ap. 4, Sfantu Gheorghe, jud.

Covasna, 520080, Romania

Phone number: 0040 745 041 225

E-mail:

anca.ingineru@yahoo.com

Student number: s1840983

Supervisors: Dr. Debra Trampe

Management Summary

Keywords — impulse buying, mood, belongingness, shopping

The current study is a research in the field of impulse buying and presents an investigation on the role of belongingness on impulse buying. It is researched the case in which shopping, as a cure for bad mood, may be substituted by being with others. So, being with others could act as a buffer against impulsive buying.

The study seeks to discover if the presence of a friend, but not the one of a distant co-worker, nor of a positive event, substitutes impulse buying when the participant is in negative mood.

The study was conducted in Romania, with full-time employed participants. The data was collected in July 2009, using paper- and electronic-based questionnaires. The total number of respondents was of 78 persons.

Preface

This thesis was completed under the supervision of Professor Dr. Debra Trampe at the University of Groningen, the Netherlands.

For the completion of this Master programme I would like to first and foremost thank my parents who supported me at every step of my life and helped me develop. They have given me all their support during the past year also, enabling my graduation and passing into a new phase of my life.

I would also like to thank the group of professors who guided me and offered great support throughout the master programme. I express my sincere gratitude towards Prof. Dr. Debra Trampe who helped me out when I thought I will not be able to write my thesis and graduate. She has guided me throughout the process of researching and writing this thesis. I would also like to thank Ms. Eline de Vries MSc, the second supervisor of this thesis, who has offered her support for my graduation.

Last, but not least, I would like to thank my friends and family who helped me in gathering the data needed for this research. I am grateful for their time spent for the completion of this research.

Table of content

1. Introduction ………...………...7

1.1. Problem statement ………...….7

1.2. Structure of the thesis ………...………...8

2. Literature Review ………...………...9

2.1. What Is Impulse Buying? ………9

2.2. When Do People Buy Impulsively? ………...11

2.3. Impulse Buying Tendency (IBT) ………...13

2.4. Influences on Impulse Buying ………...14

2.4.1. The Influence of Products ………...14

2.4.2. The Influence of Situational Variables ………..16

2.4.3. The Influence of Store Environment ……….17

2.4.4. The Influence of Mood ………17

2.4.5. The Influence of Culture ……….20

2.4.6. Social Influences ………..21

2.4.7. Shopping with Others ……….22

2.5. The Need to Belong ……….23

3. Research design ………..28

3.1. Research Method ………28

3.1.1. Participants and Design ………..28

3.1.2. Procedure ……….29 3.1.3. Independent Variables ………33 3.1.4. Dependent Variables ………...33 3.2. Plan of Analysis ………..34

4. Data Analysis ………..35

4.3. Descriptive Analysis of Variables ……….38

4.4. Distributions ………40

4.5. Correlations ………40

4.5.2. Correlation between Impulsive Purchase and Regret

at Scenario Level ………42

4.6. Main Analyses ……….42

4.6.1. Kruskal-Wallis for Scenario and Impulsive Urges ………..42

4.6.2. Kruskal-Wallis for Scenario and Impulsive Purchase …….44

4.6.3. Chi-Square Test for Scenario and Mood ………..44

4.6.4. Kruskal-Wallis for Scenario and Regret ………...47

4.6.5. Kruskal-Wallis for IBT and Impulsive Urges ………..47

4.6.6. Kruskal-Wallis for IBT and Impulsive Purchase ………….48

4.6.7. Kruskal-Wallis for IBT and Mood ………48

4.6.8. Kruskal-Wallis for IBT and Regret ………...48

4.6.9. The Influence of Manipulation and IBT on the Dependent Variables ………...48

5. Discussion ………...51

5.1. General Discussion ……….51

5.2. Limitations and Future Research ……….56

5.3. Implications ……….57

References ………...58

1. Introduction

Extensive research on impulse buying began in the early 1950s and sought to investigate those purchase decisions that are made after the consumer enters a retail environment (Rook, 1987). Since then, many researches have been conducted, linking impulse buying to many issues, such as mood, regrets, environment etc.

This thesis presents a research on the role of belongingness on impulse buying. It is researched the case in which shopping, as a cure for bad mood, may be substituted by being with others. So, being with others could act as a buffer against impulsive buying. The research is aimed at clothes and accessories products, without being conducted for a specific company or retailer.

1.1. Problem Statement

Many spend huge amounts of money on buying impulsively, sometimes due to the negative mood they have. This may represent a problem for many and a cure for bad mood without spending would represent a great advance. However, the people in negative mood need to have the desire to re-establish their positive mood and in the same time, need to accept and wish another way of attaining this without shopping.

The current research aims at studying if the presence of a friend, but not the one of a distant co-worker, nor the occurrence of a positive event, substitutes impulse buying when the subject is in negative mood. Further, three hypotheses were developed which will be or not supported by data as a result of the research conducted.

Hypothesis 1: The presence of a friend (a person with whom the subject has a close

relationship) substitutes the need for shopping on impulse when the subject is in a negative mood.

Hypothesis 2: The presence of a distant co-worker (a person with whom the subject

has not much in common, or has not developed a close relationship) does not substitute the need for shopping on impulse when the subject is in a negative mood.

Hypothesis 3: The occurrence of a positive event does not affect the impulse buying

Following this research, significant knowledge will be gained in the field of impulse buying. Would the hypotheses be confirmed or not, a contribution to the current literature in this domain would be added and further researches can continue the study in this direction. In the case the hypotheses are confirmed, the new vision on negative mood and its cures would strongly influence the shopping behavior of many, especially those who can afford less buying impulsively. Would the hypotheses not be confirmed, further studies could use the results of this thesis and improve the research techniques used.

1.2. Structure of the Thesis

This paper is divided into 4 chapters – Literature Review, Research Design, Data Analysis and Discussion. The Literature Review provides a review of the existing literature in the field of impulse buying. First, some definitions of impulse buying as well as different views are presented. Then, reasons for why people buy impulsively are discussed. As all people are different, all have different characteristics; this is also true for the impulse buying tendency (IBT) of each and every person. Previous studies have identified many types of influences on impulse buying and these are also reviewed. The most important part is the need to belong and its link to impulse buying, which is to be research in this study. The model proposed by the current study is also presented at the end of this part.

2. Literature Review

2.1. What Is Impulse Buying?

The use of impulse buying varies a lot among researches. Given this, the difficulty of comparing the findings of different research papers and of using the most proper definition in future studies arises. Further on, some of the definitions created and used by researchers over time will be described and it will be tried to improve on the definitions used until now.

Rook (1987) states that impulse buying occurs when a consumer experiences a sudden, often powerful and persistent urge to buy something immediately. Cobb and Hoyer (1986) used the classification scheme shown in Table 1, which demonstrates that an impulse purchase occurs when there was neither intent to buy a specific brand, not even from the category prior to entering the store. So, when people decide to buy an item(s) in the store, the purchase behaviour is referred to as impulse or unplanned purchasing.

Intent to buy the category

Yes No

Yes Planner -

Intent to buy the

brand No Partial planner Impulse purchaser

Table 1 - Classification scheme to demonstrate impulse purchasing, Cobb and Hoyer (1986)

In a later study, Rook and Fisher (1995) define buying impulsiveness as a consumer’s tendency to buy spontaneously, unreflectively and immediately. Based on the definition of Rook and Fisher (1995), Beatty & Ferrell (1998) define impulse buying as: a sudden and immediate purchase with no pre-shopping intentions either to buy the specific product category or to fulfill a specific buying task. They also highlight the fact that it does not include the purchase of a simple reminder item, which is an item that is simply out-of-stock at home.

The definitions presented until now for impulse buying tried to define as thoroughly and completely as possible this concept. If we are focusing on research and theory, maybe even in practice sometimes, such definitions may be very useful. However, Bellenger, Robertson and Hirschman (1978) have a more retailer-oriented vision. They believe that, in order to help retailers determine the level and direction of in-store promotion, impulse purchasing can best be defined in terms of whether the consumer made the decision to purchase prior to or after entering the store. So, they say that a retailer wanting to use the conceptual approach to impulse buying, may do so by asking: “Did you plan to purchase this item prior to entering the store?”.

Having a more attentive look at the definitions found in time by different scientists, and taking into consideration the shortcomings of each operationalization of impulse buying, the following definition is proposed to be used in this research:

Impulse buying is the action where one buys an item under the following conditions:

- decision is sudden, unexpected, unreflective, immediate

- the item is not planned, not out-of-stock, not a reminder item, not habitual, the purchase was not postponed until now

- the person has a powerful and persistent urge to buy something immediately - the greatest influence is the store environment (e.g. displays, promotion) - there is no recognized need for the item

- the purchase is not pursued due to hedonic shopping motivations such as gratification (which may be considered as planned)

2.2. When Do People Buy Impulsively?

Several reasons were identified for impulsive purchase to occur and researchers had different approaches for discovering reasons that could trigger consumers to buy impulsively.

Tauber (1972) identifies a number of motives for why people shop, such as: personal motives (role-playing, diversion, self-gratification, learning about new trends, physical activity, sensory stimulation), social motives (social experiences outside the home, communication with others having a similar interest, peer group attraction, status and authority, pleasure of bargaining). He also argues that impulse shopping may be prompted by one of the motives identified above with no planned purchase intended and that mobility increases exposure to new shopping alternatives and enhances opportunities for impulse shopping.

1. Conflicting goals and standards undermine control, such as when the goal of feeling better immediately conflicts with the goal of saving money; 2. Failure to keep track of (monitor) one’s own behavior renders control difficult; 3. Self-control depends on a resource that operates like strength or energy, and depletion of this resource makes self-control less effective. Trait differences in self-control predict many behaviors. Moreover, Baumeister (2002) states that if any of these fails, self-control can be undermined.

Dittmar et al.’s (1996) model (Figure 1) state that an important reason that individuals buy on impulse is to compensate for perceived self-discrepancies, that is discrepancy between his or her ideal and actual selves. They say that variables such as an individual’s materialism, degree of self-discrepancies, gender, and compulsive shopping tendencies determine, in part, the types of goods impulsively bought, and the reasons for the purchase.

2.3. Impulse Buying Tendency (IBT)

Rook and Fisher (1995) define buying impulsiveness as a consumer’s tendency to buy spontaneously, unreflectively, immediately, and kinetically. They state that highly impulsive buyers are more likely to experience spontaneous buying stimuli; their shopping lists are more “open” and receptive to sudden, unexpected buying ideas. Also, their thinking is likely to be relatively unreflective, prompted by physical proximity to a desired product, dominated by emotional attraction to it. As a result, impulsive buyers are more likely to act on urge and to respond affirmatively and immediately to their buying impulses. They state that in extreme cases, impulsive behavior is almost entirely stimulus driven; a buying impulse translates directly into an immediate purchasing response. Moreover, impulsive buyers are likely to experience buying impulses more frequently and strongly than other consumers. Furthermore, Verplaken and Herarbadi (2001) found that IBT is related to personality-based individual difference measures, including the Big Five. IBT was positively related to extraversion. The cognitive facet was inversely related to conscientiousness, personal need for structure and need to evaluate. The affective facet was related to lack of autonomy and action orientation. Their studies suggest that impulse buying tendency is anchored in fundamental personality characteristics.

To sum up, the findings of Rook and Fisher (1995), Verplaken and Herarbadi (2001) and Baumeister (2002) strongly suggest that people differ in the extent to which they are prone to impulse buying. This impulsivity is a fundamental personality characteristic (Verplaken and Herarbadi, 2001) and those that have high impulsive buying tendencies or low self-control, are more likely to buy on impulse than those who have high self-control and do not give in to the urges felt to purchase impulsively.

2.4. Influences on Impulse Buying

In addition to individual differences, other factors can also affect impulse buying. The following section presents some other influencing factors.

2.4.1. The Influence of Products

Unplanned purchasing is not limited to any one type of product or retail setting (Cobb and Hoyer, 1986; Bayley and Nancarrow, 1998). However, it seems more likely to occur on an overall basis in supermarkets and discount stores than in specialty and department stores. This is probably due to the nature of the products which lend themselves to impulse purchasing (Cobb and Hoyer, 1986). Dittmar et al. (1996) found that some classes of goods are bought relatively more frequently on impulse than others, jewelry and sports equipment having much higher impulse purchasing (relative to planned purchasing) than goods such as body care items and footwear.

Having found from previous research that consumers involved with a particular product category receive greater pleasure from these products and that they are more likely to browse retail stores that carry products with which they are involved and engage in higher levels of ongoing search, the study of Jones et al. (2003) further studied IBT combined with the view of involvement. They found that product-specific impulse buying tendency results from generalized impulse buying tendency and product involvement.

consumers are likely to make impulse purchases of products in general. Consequently, product-specific impulse buying tendency can be viewed as a manifestation of the general impulse buying tendency trait. Jones et al. (2003) think that individuals who, in general, have a tendency to buy products on impulse are more likely to possess a greater tendency to buy goods of a particular category on impulse.

Furthermore, the study of Jones et al. (2003) demonstrated that product-specific impulse buying tendency is more strongly related to product-specific impulse buying than the generalized impulse buying trait. This means, that product-specific impulse buying tendency is associated with impulse buying behavior in the same product category and not associated with impulse buying behavior in an unrelated product category (e.g. IBT for clothing was not significantly associated with impulse buying behavior for music). These results were supported across two different product classes (clothing and music such as CDs and tapes). Figure 2 extends the general trait– behavior model of impulse buying tendency by introducing product-specific impulse buying tendency and product involvement.

Figure 2 – Modeling the Product Specific Nature of Impulse Buying, Jones et al. (2003)

relationship-oriented reasons, while men had a more functional, instrumental and activity-related focus. A study of impulsive shopping among mature students suggested that clothes, jewellery, and cosmetics were bought by women, and high-tech, electronic and sports equipment by men. The study of Dittmar et al. (1995) has shown that high impulse buys differ from low impulse goods along a self-expressive vs. functional continuum. This pattern was especially accentuated for women who impulsively buy goods that project emotional and appearance concerns, while men tend to impulsively buy goods that are more instrumental.

2.4.2. The Influence of Situational Variables

Beatty and Ferrell (1998) found that time available influences the length of time the shopper will stay in the environment, and appears to especially influence his/her discretionary use of time in regards to browsing. Further, available resources (money) tend to affect individuals’ moods at the time, with lack of time and money producing negative affect (or frustration). Additionally, perception of available money generated positive feelings and a positive influence on actual impulse purchasing.

2.4.3. The Influence of Store Environment

Mattila and Wirtz’s (2008)study suggests that highly stimulating and pleasant store environments lead to enhanced impulse buying. Self-reported impulse buying was maximized when the store environment was perceived as over-stimulating (i.e. higher than desired in terms of excitement and stimulation). Familiarity with the store had a positive impact on impulse buying, thus suggesting that frequent buyers might be more prone to unplanned purchases. A highly unfamiliar environment, on the other hand, might make people more cautious of their actions, and therefore, inducing higher levels of self-control than familiar stores.

Dijskterhuis et al. (2005) also support the view that store environment has a strong influence on impulse buying, arguing that consumer behavior is often strongly influenced by subtle environmental cues. They disagree with the traditional view that argue consumer choice is based on conscious information processing, and say that many choices are made unconsciously and are strongly affected by the environment.

2.4.4. The Influence of Mood

Beatty and Ferrell (1998) found that positive affect is affected by one’s shopping enjoyment, in-store browsing, and money available, and produces more felt urges to buy impulsively. On the other hand, negative affect did not influence impulse buying urges. Thus, the strong effect of positive affect on urge suggests that positive moods produce greater approach behavior. However, negative affect did have a significant and negative loading on impulse purchase in their specified model. Thus, the authors say that perhaps negative affect tends to reduce an individual’s tendency to actually act on their urge.

findings suggest that people in some negative mood states may try to resist negative thinking and do so whenever possible. So, the effects of mood states may be greatest in those consumer-behavior situations where stimuli are ambiguous, consumers are somewhat aroused, induction and action are temporally contiguous, perceived benefits of being precise are low, and moods are positive. Gardner (1985) states that the effects of negative mood on behavior in the service sector may depend on the nature of the negative mood (e.g. guilt versus anger), the reasons consumers give for the mood, and the costs/benefits of the behavior to be performed.

Figure 3 – A Conceptual Model of the Role of Mood States in Consumer Behahiour, Gardner (1985)

dispositional negative affect were found strongly related to impulse buying tendency. Verplanken et al. (2005) found this finding surprising, because impulsive buying is usually associated and even defined in terms of pleasure and excitement, and, at least on the shopping floor, does not involve negative feelings. People who experience emotional distress may give up a degree of self-control and engage in activities that they expect to give immediate pleasure or satisfaction (Baumeister, 2002). Mick and DeMoss (1990) observed that people sometimes provide themselves with “self-gifts” as a form of indulgence in order to elevate a bad mood.

The perspective that impulse buying is “fun” and tends to elevate mood suggests that impulse buying might be primarily associated with positive affect. As already mentioned, Verplanken et al. (2005) proposed that negative rather than positive affect is the driving force behind chronic impulse buying. Silvera et al. (2008) state that to the extent that the primary function of impulse buying is as a mechanism for escaping negative psychological states, the presence of negative affect should be associated with chronic impulse buying tendencies. Their study provided unilateral support for Verplanken et al.’s (2005) proposition that chronic impulse buying tendencies are associated with affect, and particularly with negative affect.

The results of Dittmar et al. (1996) highlight the importance of mood in impulse buying. They found that mood is the most important of the psychological buying considerations in impulse purchasing, and that, for both men and women, those high in both materialism and self-discrepancies, are most concerned to improve their mood when buying on impulse. Elliott (1994) has shown that wanting to improve one’s mood is a strong motivation for compulsive buying, and mood may provide the dynamic element that heightens attention to self-discrepancies and triggers self-repair activities such as impulse buying.

may be seen as self-gifts in order to elevate bad mood, or as pleasure and excitement on the shopping floor.

2.4.5. The Influence of Culture

Kacen and Lee (2002) found from a multi-country survey of consumers in Australia, United States, Hong Kong, Singapore, and Malaysia, that both regional level factors (individualism–collectivism) and individual cultural difference factors (independent – interdependent self-concept) influence impulsive purchasing behavior. Compared to Caucasians, Asians engage in less impulse buying behavior due to trait buying impulsiveness. In addition, independence was not shown to impact Asians’ impulsive buying behavior, although it did contribute to impulsive buying among Caucasians. For Caucasians, the more independent their self-concept, the more impulsive buying behavior they are likely to engage in. Although there is no significant difference in trait buying impulsiveness between Caucasians and Asians there is a difference in their impulsive buying behavior. In addition, getting older reduces impulsive buying among our Asian sample but does not have a significant effect on our Caucasian sample. Affective feelings played a greater role in the impulsive buying behavior of Asians compared to Caucasians. For Asians, heightened arousal promoted impulsive buying.

2.4.6. Social Influences

Previous research showed that impulsive buying behaviour is influenced by many factors, such as: consumer moods (Beatty and Ferrell, 1998; Rook, 1987; Gardner, 1985), trait buying impulsiveness (Rook & Fisher, 1995), self-identity (Dittmar et al., 1995) and pleasant environments (Mattila and Wirtz’s, 2008; Dijskterhuis et al., 2005); yet, the role of social influences remains unclear.

relationship between the buying impulsiveness trait and the act of buying something on impulse is moderated by consumers’ normative evaluations of making an impulsive purchase. They argue that to have an impulse, it is not necessarily to act on it, as various factors may intervene between the impulsion and the action and that even highly impulsive buyers do not give in to every spontaneous buying demand.

The authors say that there is typically some delay between a buying impulse and an impulsive purchase – even if for only a few seconds, consumers are still thinking, feeling, and evaluating various retail stimuli. Moreover, even highly impulsive consumers may experience normative encouragement or discouragement when the urge to buy something on impulse. They found that, when acting on impulse is socially appropriate and rational, consumers tend to have both a greater impulsive urge to buy and a greater likelihood of doing so. Individuals who have low impulsive tendencies and judge a possible impulse purchase negatively are unlikely to act on their buying impulses in such situations. However, when more impulsive consumers view a purchase as bad, they are likely to feel simultaneously a spontaneous urge to buy and a strong normative warning against acting on impulse. So, consumers’ normative evaluations can moderate the link between the trait, behavioral aspects of impulse buying and subsequent buying behavior. Even impulsive buyers seem able to reject making an impulsive purchase when negative normative evaluations reach some critical level.

2.4.7. Shopping with Others

Social influences may appear also when the influencing person shops with the person that is susceptible to influence, so when the influence is present. So, this paper also takes a look at the case of shopping with others and the consumer behaviors when the influence is present at shopping will be presented.

support, saying that peer group norms are relatively more receptive to the urge to purchase and induce less pressure to monitor one’s purchases, their presence in a purchasing situation can activate different expectations. Results of 2 experiments suggested that the presence of peers increases the urge to purchase, and the presence of family members decreases it. Thus, if the peer group norms encourage impulsive buying, then the tendency to engage in this behavior in the presence of peers should increase with the cohesiveness of the peer group in question. In contrast, if the family’s norms are antagonistic to impulsive purchasing, then the tendency for the presence of family members to inhibit impulsive purchasing should be greater when the family is cohesive. Moreover, this difference is greater when the group (peers or family) is cohesive and when participants are susceptible to social influence.

However, there are cultural differences to take into consideration when speaking of social influences and impulse buying. However, Lee and Kacen (2008) did not study the process of impulse buying when being with peers or family, but the satisfaction created by shopping on impulse alone or with someone for individualist and collectivists consumers. They found that the effect of being with another person during an impulse purchase has a differential effect across cultures. Consumers from collectivist countries were more satisfied with their impulse purchase when they were with an important other versus when they were alone at the time of purchase, while consumers from individualist countries showed no difference in satisfaction between these two purchase situations. For planned purchases, the presence of another person had no impact on post-purchase satisfaction for either collectivists or individualists.

2.5. The Need to Belong

As seen until now, existing research focused on shopping, impulse buying and the influence of norms on these behaviours. However, being around others has another function, too. Further on, the need to belong will be discussed, as well as its link to impulse buying.

relatedness, and self-esteem. Security may also be a need, which becomes salient in times of privation. Furthermore, by considering several criteria, Baumeister and Leary (1995) showed that the need to belong is a fundamental human motivation. The criteria that the authors believe should be fulfilled in order to conclude that the need to belong is a fundamental human motivation are the following: a) produce effects readily under all but adverse conditions; b) have affective consequences; c) direct cognitive processing; d) lead to ill effects (such as on health or adjustment) when thwarted; e) elicit goal-oriented behaviour designed to satisfy it (subject to motivational patterns such as object substitutability and satiation); f) be universal in the sense of applying to all people; g) not be derivative of other motives; h) affect a broad variety of behaviours; and i) have implications that go beyond immediate psychological functioning.

They found evidence of a basic desire to form social attachments. People form social bonds readily when having anything in common, when they share common experiences, or when they are just exposed to each other frequently. Moreover, people strongly resist losing attachments and breaking social bonds, even if there is no rational reason to maintain the bond and even if maintaining it would be difficult. They also found evidence that the need to belong shapes emotion and cognition. Forming or solidifying social attachments generally produces positive emotion (happiness, elation, contentment, and calm), whereas real, imagined, or even potential threats to social bonds generate a variety of unpleasant emotional states (anxiety, depression, grief, jealousy, and loneliness). Evidence was also found with regard to cognitive processing- people devote a disproportionate amount of cognitive processing to actual or possible relationship partners and interaction partners, and they reserve particular, more extensive patterns of information processing for people with whom they share social bonds.

Deficits in belongingness apparently lead to a variety of ill effects. Both psychological and physical health problems are more common among people who lack social attachments. Behavioural pathologies, ranging from eating disorders to suicide, are more common among people who are unattached.

less well off then people who can satisfy both. It is unclear whether the interactions must be pleasant or can be satisfactory if they are merely neutral. The evidence suggests only that aversive or conflictual interactions fail to satisfy the need.

The need to belong also appears to conform to motivational patterns of satiation and substitution. People need a few close relationships, and forming additional bonds beyond those few has less and less impact. When people brake a social bond, they seem to recover best by forming a new one, although each individual life tends to involve some particularly special relationships that are not easily replaced. People without intimate partners engage in a variety of activities to find partners, but people who have partners already are much less active at seeking additional relationships, consistent with the satiation hypothesis.

Although some issues remain unproved still, the authors conclude that the evidence found is sufficient to confirm that the need to belong is a fundamental human motivation.

As could be seen, according to Baumeister and Leary (1995), deficits in belongingness may lead to a variety of ill effects. They state that behavioural pathologies, ranging from eating disorders to suicide, are more common among people who are unattached. Impulse buying and compulsive buying even more, can be considered behavioural disorders. Taking this into consideration, Baumeister and Leary’s (1995) view could be extended to impulsive buying. Would belongingness substitute the need to buy impulsively, a whole new aspect of impulse buying would be discovered.

As it is unclear whether the interactions must be pleasant or can be satisfactory if they are merely neutral, the current hypotheses will suppose that it is just the interaction with a close friend that has a curing effect on the negative mood, and not any interaction.

influence of being with others, other variables were taken into consideration, so that the study would offer a more complete view on the impulse buying behaviour.

As the current research aims at studying if the presence of others, that is the presence of a friend, substitutes impulse buying when the subject is in negative mood, the hypotheses that arise are the following:

Hypothesis 1: The presence of a friend (a person with whom the subject has a close

relationship) substitutes the need for shopping on impulse when the subject is in a negative mood, acting as a mood elevator.

Hypothesis 2: The presence of a distant co-worker (a person with whom the subject

has not much in common, or does not developed a close relationship) does not substitute the need for shopping on impulse when the subject is in a negative mood, therefore not acting as a mood elevator.

For testing the fact that it is the need to belong that substitutes impulse buying as a mood improvement, and not just any positive event that occurs to the respondent, the case where a positive event occurs was used also. The case where the respondents encountered a positive event was not related to meeting with people and the effect of this meeting, but an event unrelated to belongingness which is supposed to be positive for everyone. The event concerned the finding of a lost object, which if would not have been found, would have caused the respondent much hassle.

Therefore, it is supposed that: Hypothesis 3: The occurrence of a positive event does not affect the impulse buying behaviour when the subject is in a bad mood, therefore not acting as a mood elevator.

Previous studies have not connected impulse buying, the need to belong and the negative feelings that have other causes than the unsatisfied need to belong. However, previous research (Baumeister, 2002; Dittmar et al., 1996; Verplanken et al., 2005; Mick and DeMoss, 1990; Silvera et al., 2008; Elliott, 1994) found evidence that negative mood generates the need to buy impulsively. Researches such as Baumeister and Leary’s (1995) suggest that the need to belong is a basic human need and the lack of belongingness may produce the appearance of negative feelings.

close friend if they are in a bad mood. Sometimes, just telling someone the bad events of a day, or telling the cause for the bad mood, one can feel relieved and have a more positive mood already.

Therefore, this research represents a first step in this direction, trying to bring new insight into impulse buying, a subject which has been research from so many perspectives. Consequently, a new model (Figure 4) is proposed based on the hypotheses presented above.

Figure 4 – Proposed model: Belongingness, a cure for negative mood and a buffer against impulsive purchase

The expectations of the study are that when a subject is in a negative mood, the presence of a friend is enough in order to improve the mood of the subject and therefore impulse buying does not occur anymore as a means of re-establishing the positive mood. So, it is expected that the hypotheses will be supported.

Negative mood Impulse buying as a cure for bad mood Lack of belongingness Belongingness - as a cure for bad mood

Different negative events

Improved Mood & no impulse buying Improved Mood &

3.

Research Design

3.1. Research Method

3.1.1. Participants and Design

For the choice of a sample, snowball sampling was used. First, a few participants were identified and asked to complete the questionnaire. They are employees from different financial institutions, health government organizations, education institutions, services companies, mobile telephony company, commerce companies and media companies, being located in different cities in Romania. Then, the initial participants were asked to pass the questionnaire to other individuals from the same organization. Only women were selected for this study due to their higher impulsiveness than men’s (Dittmar et al., 1996). Seventy-eight women participated in the study, twenty-six participants for each case. Participants were assigned randomly to one of the three scenarios. Responses were collected for each scenario from each institution.

The respondents are middle or highly educated persons, living in Romania, having Romanian or Hungarian nationalities. The cities where the participants live differ, so this fact may influence their answers, due to regional differences between people’s lifestyles. The age range was between 20 and 55.

of the expected results of the study would want to answer according to what they believe is the right answer. For this reason, a cover story was used for informing the interviewed people about the scope of the study.

3.1.2. Procedure

The data were collected by sending the questionnaires to the participants by e-mail (in Word format) or distributing the questionnaires (in hard copy) personally and after each respondent filled in the questions, the forms were gathered. Although this method has its drawbacks, such as not measuring the actual buying behaviour, it offers some advantages: perhaps more honest answers than in face-to-face interviews and convenience. As mentioned, this research misses the measurement of actual behaviour that participants have in a shopping environment, as no observation study was conducted. However, the observation method has its disadvantages also, because observing and interviewing shoppers in stores may change what is being observed about impulsive buying, thus leading to confounding the results by means of the “guinea pig effect” (Luo, 2005). Moreover, an observational study would not have been appropriate for this study, due to the uniqueness of factors and cases taken into consideration.

First, participants answered a few demographic questions (age, marital status, education etc.). Before starting the questionnaire, participants were informed that the study’s purpose was to understand the power of imagination in a shopping situation. They were also told that there are no right or wrong answers and the most important thing was to put themselves in the place of the imaginary person and decide taking into consideration their own opinions. Then, they were presented the imaginary situations. There were three scenarios. In all 3 cases, the fictive person was in a negative mood. However, each participant had to answer a set of questions related to only one scenario. So, there were different participants answering the questions related to each scenario. Below the 3 scenarios are presented:

Case 1 – Ann runs into best friend

to buy a swimming suit for the swimming course she has subscribed. Therefore, she decides to go to the mall after work to buy the swimming suit. Before entering the store, she runs into her best friend and they talk for some time. After saying goodbye, Ann goes into the store. As she is walking to the underwear and swim department, Ann sees a very nice pair of jeans on sale for 150 RON/ 40 euro.

Case 2 - Ann runs into a distant co-worker

Ann is a 24-year-old woman with a full-time job. She had a bad day at work - just one of those days when everything goes wrong. It is two days before Ann gets her next paycheck and she has only 100 RON/ 30 euro left for necessities. In addition, she has to buy a swimming suit for the swimming course she has subscribed. Therefore, she decides to go to the mall after work to buy the swimming suit. Before entering the store, she runs into a distant co-worker. They talk for a few minutes. After saying goodbye, Ann goes into the store. As she is walking to the underwear and swim department, Ann sees a very nice pair of jeans on sale for 150 RON/ 40 euro.

The third scenario was used in order to strengthen the first two hypotheses used by this research that suppose it is the need to belong which substitutes impulsive buying as a mood improvement, not any positive event encountered by a person.

Case 3 – Ann encounters a positive event

Ann is a 24-year-old woman with a full-time job. She had a bad day at work - just one of those days when everything goes wrong. It is two days before Ann gets her next paycheck and she has only 100 RON/ 30 euro left for necessities. In addition, she has to buy a swimming suit for the swimming course she has subscribed. Therefore, she decides to go to the mall after work to buy the swimming suit. Before entering the store, she receives a phone call. A person from the Police Department tells Ann that her lost driving license was found and will be returned to her. After saying goodbye, Ann enters the store. As she is walking to the underwear and swim department, Ann sees a very nice pair of jeans on sale for 150 RON/ 40 euro.

purchase decision alternatives Ann would make. These choice alternatives were designed to represent varying levels of buying impulsiveness. From low to high impulsiveness, these alternatives were: (1) buying the swimming suit only, (2) wanting the jeans but not buying it, (3) deciding not to buy the swimming suit, (4) buying both the swimming suit and jeans with a credit card, and (5) buying these plus matching blouse and a pair of ballerinas shoes, also with a credit card (Rook and Fisher, 1995).

Respondents used in the study of Luo (2005) were presented with the same purchasing situations and were asked to rate the impulsiveness of each purchase alternative along a 10-point scale ranging from 1 (lowest impulsiveness) to 10 (highest impulsiveness). The alternative “buy only the planned item” was rated lowest (M = 0.51); followed by “want the impulse item and not buy it” (M= 1.83); “decide not to buy the planned item” (M= 2.98); “buy both the planned item and the impulse item with a credit card” (M = 8.92); and “buy both, plus matching clothes, also with the credit card” (M= 9.34). It is important to note that “decide not to buy” (the third decision alternative) was reported to be more impulsive than either “buy only the planned item” (first decision alternative) or “want the impulse item and not buy it” (second decision alternative). Rook and Fisher (1995) explained this by noting that “because the imaginary person was described as planning to buy the planned item, some respondents appeared to viewed her change of plans as impulsive”. Alternatively, participants may have interpreted this response alternative to mean “decide to purchase the impulse item and not the planned item” rather than interpreting “no purchase” as impulsive in its own right. Therefore, this study used the same order for impulsiveness of the buying decisions as previous studies.

After answering the questions regarding urges to buy and buying decision, participants were asked questions about feelings (regret, disappointment and happiness) after the purchase decision. These feelings were measured on a 7-point scale ranging from 1 (not regretted / not disappointed / unhappy) to 7 (regretted/disappointed/ happy).

involvement and others. These questions are also useful for distracting the respondents from understanding the real focus of the study and consequently be influenced by normative influences regarding impulsive purchase (Rook and Fisher, 1995).

So, after the set of unrelated questions, the questionnaire continued with questions regarding the level of attention with which they read the scenario and the extend to which they projected themselves into the scenario; product involvement; IBT (to capture pre-existing differences in IB). The manipulation check was done by asking participants to indicate whether they recognized that the imaginary character encountered one of the types of encounters presented (yes/no) when they projected themselves into the shopping scenarios (Dahl et al., 2001).

Besides the independent and dependent variables, an exploratory variable was used – product involvement. This variable was used in order to eliminate the influence product involvement may have had on the participants to the research. The involvement scale’s items are: 1) “I like very much clothes.”, 2) “I enjoy browsing in clothes stores.”, 3) “I greatly enjoy buying clothes.”, 4) “Usually, I go more often into stores that sell clothes than other items.”, 5) “I am more pleased when I browse in clothes stores than in other types of store.”.

3.1.3. Independent Variables

The independent variables are: the type of encounter which will be manipulated and the impulse buying tendency (IBT) which will just be measured for each respondent, as it is a personality trait.

The type of encounter has 3 levels – encounter with a friend, a distant co-worker or a positive event. By exposing the participants to one of the 3 encounters, the differences in the relationships between the variables were measured. Therefore, the effect of belongingness (in the encounter with a friend) was measured and compared with the effects caused by the other 2 encounters.

3.1.4. Dependent Variables

The dependent variables are represented by: the urge to buy impulsively, impulsive purchase decisions, mood and regret.

In order to provide an index of impulsive urge, participants were asked to project themselves into the shopping scenario and to report their agreement with four items: a) I experienced a number of sudden urges to buy; b) I wanted to buy things even though they were not on the shopping list; c) I had strong urges to make impulsive purchases; and d) I felt a sudden urge to buy (Luo, 2005).

Impulse buying decision was assessed with the measure developed by Rook and Fisher (1995). So, respondents chose a purchase decision alternative from the following: a) buy only the planned item; b) want the impulse item and not buy it; c) decide not to buy the planned item; d) buy both the planned item and the impulse item with a credit card; and e) buy both, plus matching items, also with the credit card. Choices of these alternatives were coded 1 through 5, respectively.

3.2. Plan of Analysis

After finishing the questionnaire design, a questionnaire pretest has been conducted. By doing this, some problems were identified and solved, so that the actual sample would not have problems with the understanding of the questions and tasks. Following the pretest, some changes were made to the questionnaire.

Factor analysis was used in order to create the impulse buying tendency (IBT) index, while from the impulsive urge scale the index was developed by using the mean.

Kruskal-Wallis analysis of variance was used to examine the differences in the mean values of the dependent variable (urges to buy on impulse, impulse purchase) associated with the effect of the controlled independent variables (type of encounter), after taking into account the influence of the uncontrolled independent variables (IBT). This test was chosen as method of analysis due to the distribution of the dependent variables.

4. Data Analysis

4.1. Demographic Data

At the population level, the average age of participants was between 31 and 36 years (median= 3; representing the 3rd age group), with a std. dev. = 1.64. The average monthly income of the respondents was between 300 and 600 euro (median= 2; representing the 2nd income category), with a standard deviation of 0.74. So, it can be observed by comparing the standard deviations of the 2 variables, that age has a larger variability within respondents than the monthly income.

When taking into account the manipulation, in the case of encounter with best friend, the mean age was between 37 and 42 (median= 4, representing the 4th age group), and the average monthly income was between 300 and 600 euro (median = 2, representing the 2nd monthly income category). The standard deviation for age was 1.72 and for monthly income 0.77, therefore the age varying somewhat more within the set of data than income.

In the case of encounter with a distant co-worker, the mean values remained in the same category (median for age = 3.5, representing the 4th category of age; and median for income = 2, representing the 2nd income group), with slight changes in the standard deviations for age (std. dev. = 1.69) and monthly income (std. dev. = 0.74), as compared to the 1st manipulation.

For the 3rd case – encounter with positive event, the mean age was between 25 – 30 years (median = 2, representing the 2nd age group). The average monthly income was between 300- 600 euro (median=2, representing the 2nd income group). Once again, the variation of the data set for age (std. dev. = 1.52) was larger than for monthly income (std. dev. = 0.71).

So, it is shown that the distribution of ages, education and monthly income through the conditions is approximately the same for all 3 conditions. This distribution is preferred, as the demographic data might influence the responses of the participants and therefore not allow for a correct analysis of the variables of interest.

It is important to note that there are no systematic differences between conditions (encounter with best friend, distant co-worker or positive event), as would have been expected when using random assignment of participants to the manipulation cases.

4.2. Indexes

As some variables were measured on multiple item scales, it was necessary to create a single variable which incorporates each item of the respective scale. Two methods were used in order to create compound variables – either averaging the items, or using factor analysis. The choice of the technique used for each scale was based on previous researches, such as Rook and Fisher (1995) and Luo (2005) who conducted similar researches, similarity of the scale’s items and sphericity and sample adequacy tests.

Impulsive Urges Index

To provide an index of impulsive urge, participants were asked to project themselves into the shopping scenario and to rate their agreement with four items presented in the Research Design section. Responses were made on a multi-item Likert scale that ranged from 1 (strongly disagree) to 7 (strongly agree). Due to the similarity between the items, as well as considering the study of Luo (2005), the index of impulsive urge was created by averaging each items. For the impulsive urges scale, Cronbach’s α was 0.79, therefore the scale is reliable.

Appendix 3 shows that the mean value of impulsive urge took values from 1.5 to 7, measured on a scale from 1 to 7. At the population level, urges had a mean of 4.48 and standard deviation of 1.58. The values for impulsive urge tend to concentrate towards 4 (mode), representing a medium impulsive urge.

Involvement Index

averaging the scores given by the respondents to each of the statements. The statements of the scale were measured on a Likert-scale from 1 to 7. The involvement scale was found to be reliable (Cronbach’s α = 0.87).

Involvement was scored with a minimum of 1.2 and a maximum of 7 by respondents. Involvement’s mean was 5.18 and its standard deviation of 1.42. The values for involvement tend to concentrate towards 6.4 (mode), which is considered high involvement. There were found no significant (sig. = 0.3) differences between the involvement levels of participants exposed to the 3 manipulation conditions. Therefore, the results of the dependent variables will not be misrepresented by involvement.

Impulse Buying Tendency Index

Before condensing the 9 items of the IBT scale into one variable, item 8 - “I carefully plan most of my purchases.”- was reverse-coded. This was done because it was the only item stating the opposite of what the IBT scale represents. Therefore, by reverse-coding it, the results would be in accordance with the rest of the items in the scale. Then, factor analysis was used as a means for creating one variable from the 9-item scale. For ensuring the appropriateness of this method, KMO and Bartletts’s tests were done. The buying impulsiveness scale developed by Rook and Fisher (1995) scored 0.74 on the KMO test, demonstrating that factor analysis would result in 74% of the relationship between factors. A reliability test was also run, obtaining a Cronbach’s α of 0.81 (α > 0.7), therefore being a reliable scale. Appendix 4 shows the KMO test results, and Appendix 5 presents the Component Matrix which shows the correlation between the components of the index with the IBT index.

Further, the values obtained for IBT factor were ranked and thus, a new compound variable was created (NFAC1_1), having values from 1 to 5. This method was used for the easing the data interpretation.

Impulse buying tendency showed a similar variability of data (std. deviation = 1.41) as impulsive urges. The values for IBT tend to concentrate towards 2 (mode), which is considered low impulsiveness.

the measures of central tendency data per scenario. In the case of encounter with best friend, the means of the 3 variables had just slight variation from the case when the entire population was taken into analysis. Nor did the standard deviation show important changes.

4.3. Descriptive Analysis of Variables

Mood

Appendix 7 shows the frequencies of the mood states that the respondents felt after leaving the shop for each of the 3 conditions. Those who reported feeling in a better mood after leaving the store represented 80.8% from respondents in condition 2 (encounter with distant co-worker), followed by 73.1% from respondents in condition 1 (encounter with best friend), while only 57.7% reported feeling better after leaving the store from respondents in condition 3 (encounter with positive event).

Happiness

Chart 1 – Happiness of Participants per Condition

Impulsive Purchase Decisions

Observing Appendix 8, it can be noticed that the purchase decision which was chosen by the greatest percentage of respondents was “Wants to buy the jeans but does not buy them”, with 32.1%. However, at 1.3% difference, the purchase decision “Buys both the swimming suit and the jeans with a credit card” accumulated 30.8% of the total number of responses. “Buy the swimming suit only” has 25.6% of the responses. The other purchase decision options were chosen by a small percent of respondents – 5.1% “Decides not to buy the swimming suit” and 6.4% the most impulsive purchase decision – “Buys both the swimming suit and the jeans, plus a matching blouse and a pair of ballerina shoes, also with a credit card”.

Impulsive Buying Tendency

4.4. Distributions

The distribution of the dependent variables was checked, without considering the manipulation. Appendix 10 shows the histograms of the impulsive urge, impulsive purchase and mood after leaving the store, while the 3 histograms in Appendix 11 represent graphically the distribution of urges felt by the participants in each experiment condition. It can be noticed that the distribution of impulsive urges is not normal for the variable neither without considering the manipulation, nor when the manipulation was considered. Appendix 10 shows that neither do impulsive purchase decisions or mood after leaving the store have normal distribution, when the manipulation was not considered. It can also be stated that the distribution of mood after leaving the store is skewed to the right. Moreover, it was observed that the data had a graphical pattern that followed the trend of a normal distribution.

4.5. Correlations

4.5.1. General Correlations at Population Level

Bivariate correlation was used to discover whether the scores of one variable change in a similar way as the change of another variable change.

impulsive_urge

Impulsive purchasing

Mood after

leaving the shop Regret

Pearson Correlation 1,000 ,241* -,163 -,164 Sig. (2-tailed) ,033 ,153 ,150 impulsive_urge N 78,000 78 78 78 Pearson Correlation ,241* 1,000 -,388** -,308** Sig. (2-tailed) _,033 ,000 ,006 Impulsive purchasing N 78 78,000 78 78 Pearson Correlation -,163 -,388** 1,000 ,346** Sig. (2-tailed) _{,153 ,000 ,002}

Mood after leaving the shop

N 78 78 78,000 78

Pearson Correlation -,164 -,308** ,346** 1,000

Sig. (2-tailed) _{,150 ,006 ,002}

Regret

N 78 78 78 78,000

*. Correlation is significant at the 0.05 level (2-tailed). **. Correlation is significant at the 0.01 level (2-tailed). Table 2 – General Correlations

4.5.2. Correlation between Impulsive Purchase & Regret at Scenario Level

4.6. Main Analyses

As the distribution of the dependent variables is not normal and there were 3 independent samples to be compared, the Kruskal-Wallis non-parametric test was run and the analysis was based on the comparison of the rank sums for the 3 sets of sample data.

Hypothesis 1 states that the presence of a friend substitutes the need for shopping on impulse when the subject is in a negative mood. Hypothesis 2 and 3 support the idea presented in Hypothesis 1, stating that the presence of a distant co-worker, nor would the occurrence of a positive event substitute the impulsive purchase.

The Influence of Manipulation on Dependent Variables

4.6.1. K-W for Scenario & Impulsive Urge

Taking into consideration that impulsive urges should be substituted by belongingness in the case when the respondent participated in condition 1 (the person met with best friend), the differences between impulsive urges for participants in the 3 scenarios were studied.

It was observed that the average impulsive urge for the 1st condition was 4 (M= 4), with a standard deviation of 2. For the encounter with a distant co-worker, on the other hand, the impulsive urge was a bit higher (M=5), with a standard deviation of 1. In the 3rd condition, encounter with positive event, the average impulsive urge was equal to the 1st manipulation condition (M=4) and same std. dev. = 2.

Ranks

scenario N Mean Rank

encounter with best friend 26 39,75

encounter with distant

co-worker 26 40,25

encounter with positive event 26 38,50

Impulsive Urge

Total ₇₈

Table 3 - Mean Ranks for Impulsive Urges per Condition

Therefore, the null hypothesis was confirmed, meaning that there are no significant differences between the impulsive urges reported by the participants in the 3 conditions. It can be concluded that the type of encounter – with a friend, distant co-worker or positive event – does not influence the impulsive urges felt, the results being valid for the entire population. This is contrary to the expectations that the encounter with the best friend would produce an effect on the impulsive urges felt by the participants, in the sense that they would have experienced lower levels of urges to buy impulsively. One of the reasons for this result may be the sample used for this study, as well as other reasons which will be discussed more in-depth later in this paper.

4.6.2. K-W for Scenario & Impulsive Purchase

Impulsive purchase was assessed with a measure developed by Rook and Fisher (1995). Specifically, participants were asked to choose which one of five purchase decision alternatives the imaginary character Ann would make: 1) “buy only the swimming suit”; 2) “want the jeans and not buy it”; 3) “decide not to buy the swimming suit”; 4) “buy both the swimming suit and the jeans with a credit card”; and 5) “buy both, plus matching blouse and a pair of ballerina shoes, also with the credit card.” Choices of these alternatives were coded 1 through 5, respectively.

encounter with positive event. The least impulsive purchase decisions were taken by those respondents who were exposed to the 1st condition – encounter with best friend. The results of the Kruskal-Wallis test show that there are no significant differences at the population level. Moreover, there were not found significant differences (asymp. sig. = 0.53) of impulsive purchase decisions for the 3 conditions.

Ranks

scenario N Mean Rank

encounter with best friend 26 36,54

encounter with distant

co-worker 26 43,25

encounter with positive event 26 38,71

Impulsive purchasing

Total 78

Table 4 – Mean Ranks for Impulsive Purchase per Condition

So, for what concerns the 2 main analyses performed until now, it seems that participants exposed to one manipulation condition did not experience different levels of impulsive urges, nor did they take different purchase decisions as compared to participants who were exposed to another manipulation condition, as would have been expected.

4.6.3. Chi-Square Test for Scenario & Mood

The Chi-Square tested the null hypothesis, meaning that the 2 variables (type of encounter and mood) are independent from each other. The results of the test show that there is a significant (asymp. sig. = 0.05) relation between the type of encounter to which the respondents were exposed to and the reported mood after leaving the store.

a positive event. Moreover, participants who encountered a positive event were the most numerous being in a worse mood after leaving the store than when entering, as compared to the other 2 conditions.

Chart 2 – Mood After Leaving the Store per Condition

In order to determine the relation between the type of encounter and mood after leaving the store, regression analysis was performed. The R square of the model is 0.05 (Table 5), which does not provide a sufficient explanatory power, with a significance of 0.06 (sig. > 0.05). So, perhaps a larger sample would have provided enough data for obtaining a relevant result.

Model Summary Model R R Square Adjusted R Square Std. Error of the Estimate 1 ,217a ,047 ,034 ,788

a. Predictors: (Constant), scenario

Coefficientsa Unstandardized Coefficients Standardized Coefficients 95% Confidence Interval for B

Model _{B Std. Error Beta} t Sig.

Lower Bound Upper Bound (Constant) 1,064 ,236 4,509 ,000 ,594 1,534 1 scenario ,212 ,109 ,217 1,936 ,057 -,006 ,429

a. Dependent Variable: Mood after leaving the shop

Table 6 – Regression Analysis Result 2

4.6.4. K-W for Scenario & Regret

The Kruskal-Wallis test was used in order to discover the relationship between the scenario that respondents projected themselves into and the reported regrets they felt with regard to the imaginary character’s impulsive purchase decision. The results show there is no significant difference (asymp. sig. = 0.67, sig. > 0.05) between the regrets reported by the respondents for the 3 manipulation cases (mean rank = 37.5 for 1st condition, 38.3 for the 2nd condition and 42.6 for the last condition). So, the type of encounter has no effect on the regrets felt by the respondents participating in one of the 3 manipulations. This means that the fact that some participants encountered the best friend, others the distant co-worker and others the positive event, did not change their levels of regrets felt with regard to the impulsive purchase.

The Influence of IBT on the Dependent Variables

4.6.5. K-W for IBT & Impulsive Urge

4.6.6. K-W for IBT & Impulsive Purchase

Once again, the Kruskal-Wallis test showed no significant differences (asymp. sig. = 0.33, sig. > 0.05) in the purchase decision took by respondents with different levels of impulse buying tendency. So, it seems that impulse buying tendency does not influence the way in which people behave in purchase situations. This is an interesting finding, considering the previous finding which suggests that IBT influences urges to buy impulsively. It can be concluded that although people with high impulsiveness feel the urge to buy impulsively, they control themselves and do not take more impulsive purchase decisions than those participants with lower levels of IBT and urges to buy impulsively.

4.6.7. K-W for IBT & Mood

After running the Kruskal-Wallis test for determining if there are differences in the stated moods of the respondends influenced by IBT, results show that there are no significant differences (asymp. sig = 0.26). So, the fact that some respondents are more impulsive than others does not influence their mood after leaving the store. This could mean that although respondents with high levels of IBT felt higher levels of urge to buy impulsively, did not buy more on impulse than those who had low levels of urge to buy impulsively, and did not feel better or worse due to their choice as compared to the lower IBT respondents.

4.6.8. K-W for IBT & Regret