Testing the boundary for sequential mitigation effect using an international sample: an individual difference in self-monitoring.
The main goal of the current research is to examine the boundary for sequential impulsive purchases on consumers' purchase intentions. Dholakia, Gopinath, and Bogozzi (2005) have found that impulse purchases decline when individuals engage in prior impulsive behaviors, which they referred to as sequential mitigation effect (SME). These researchers argue that "decision makers will become less likely to choose an impulsive option on account of having participated in a prior impulsive choice task" (Dholakia et al., 2005, p 180). They claim that the desire for impulsive options is a limited motivational resource and individuals may consume it after engaging in prior impulsive purchases, which accounts for the drop in sequential impulsive purchases. Given these researchers' findings, the current research contributes to the impulse buying literature by introducing a factor that may put a boundary for SME. In the current research, the author argues that SME works in a different way for low and high self-monitoring individuals. Also, the current research is different from Dholakia's et al. (2005) research in three-folds. First, this paper utilizes an international sample to expand on and test for the SME effect. Second, unlike Dholakia et al. (2005), the author of the current research examines the effect of SME over a longer period of time (days compared to minutes) in order to determine whether SME can be sustained after a longer time gap between two sequential tasks. Third, the author examines SME across different impulsive tasks to show the robustness of SME. In particular, some materialistic tasks such as impulsive purchases (versus impulsive donation) may be viewed by some people as socially undesirable behaviors and thus these people will more likely avoid these tasks to maintain their image in society. Therefore, it is important to understand the boundary for SME and whether it works under different levels of impulsive tasks that are viewed positively or negatively by society.
From a managerial and practical standpoint, impulsive purchases have been estimated to account for more than $4 billion in retail store sales (Dolliver, 1998; Mogelonsky, 1998). Retailers are continually trying to increase the number of impulsive purchases in stores through product displays and store and package designs (Hoyer and Maclnnis, 1997; Jones et al., 2003). In addition, contemporary marketing innovations expand impulsive buying opportunities (Kacen and Lee, 2002) and repeated impulsive purchases. Hence, knowing whether SME will influence repeated-impulsive situations is vital to firms' revenues. In addition, most consumer behavior or decision-making is based on repeated decisions (Bagozzi, 1981; Betsch, Fiedler, and Brinkmann, 1998; Betsch et al., 2001; Betsch and Haberstroh, 2005; Bentler and Speckart, 1979; Norman and Smith, 1995; Verplanken, Aarts, and van Knippenberg, 1997). In these repeated buying situations, previous decisions can systematically influence later ones. By knowing the boundary for SME, managers can easily influence consumers' impulsive decisions. However, there is a gap in the literature as to how the influence of SME operates in making subsequent decisions. To fill this gap, this research will investigate the impact of previous impulsive decisions on subsequent ones and whether this impact is qualified by the individual difference in self-monitoring.
In general, impulsive behaviors have been a target of philosophical discussion for many years. Specifically, extensive research on impulse buying behaviors began in the early 1950s and focused on investigating those purchase decisions that are made after the consumer enters a retail environment (Rook, 1987). Stern (1962) provides the foundation for defining impulse buying behavior, which classifies it as planned, unplanned, or impulse. Based on this classification, planned buying behavior involves a time-consuming information search followed by rational decision making (Piron, 1991; Stern, 1962). Whereas, unplanned buying refers to all purchases made without such advance planning and includes impulse buying.
Although impulsive behaviors can occur in any setting, consumer impulse buying, in particular, is an extensive everyday context for it. Rook (1987, p. 191) defines consumer impulse buying as "a sudden, often powerful and persistent urge to buy something immediately. Also, impulse buying is prone to occur with diminished regard for its consequences." In the same vein, Hoch and Loewenstein (1991) explain the impulse buying as a struggle between the psychological forces of desires and willpower. An impulse buying as described by Rook (1987) tends to disrupt the consumer's routine behavior and is more emotional than rational.
In a different vein, impulse buying is more likely to be perceived as "bad" than "good" (e.g., in the areas of personal finance, post-purchase satisfaction, social reaction, or overall self-esteem) (Rook, 1987; Rook and Hoch, 1985). In addition, consumers are more likely to feel out-of-control when buying impulsively than when making thoughtful purchases (Rook, 1987). Yet, it is possible to imagine situations in which impulse buying would be viewed as normatively neutral or even positively legitimate behavior (e.g., donating money, spontaneous gift of an ill person, taking advantage of a two-for-one in-store special, or a sudden decision to pick up the tab for a meal) (Rook and Fisher, 1995).
Beginning in the 1990s, researchers began taking a deeper look inside the consumers, especially in terms of whether his or her spending behavior was dictated by mood or generalized willpower. Among these studies, there is one recent stream of research that seems to be promising for understanding the causes of impulse buying. This stream of research examines the relationship between impulse buying and depletion of self resources. According to Hoch and Loewenstein (1991), consumer decisions represent an ever-shifting conflict between desire and willpower. That is, when the desire for a product surpasses consumers' intentions not to make the purchase, impulse buying can occur. For the most part, this view emphasizes the two separate mechanisms involved in impulsive spending: (1) the desire to buy and (2) the ability to exercise self-control over this urge (Vohs and Faber, 2007).
Similar to the previous stream of research, Dholakia et al. (2005) claims that the desire to engage in impulsive behaviors may be caused by individuals' consumption of motivational resources. Once these motivational resources are left unconsumed, individuals will feel the urge to engage in impulsive purchases. However, by the time these resources are consumed or depleted in the first impulsive choice, individuals will feel less motivated to engage in impulsive purchases in later tasks. With respect to Dholakia's et al. research, in the current research the author argues that SME works differently for high and low self-monitoring individuals. Also, supporting Dholakia's et al. findings, the author argues that SME still applies for sequential impulsive choices that are separated by a time gap; although, the author believes that SME can sustain a longer time gap (i.e., days compared to minutes) and is limited to low self-monitors. That is, the author believes that low self-monitors have less desire (compared to high self-monitors) to behave impulsively in the second choice following days of having engaged in the first impulsive choice, which is due to the depletion of their motivational resources.
SELF-MONITORING AND SME
Among the various personality traits that have been associated with marketplace behaviors, self-monitoring has attracted particular attention from marketing researchers (Browne and Kaldenberg, 1997). Self-monitoring is the tendency to notice cues for socially appropriate behaviors and modify one's behavior accordingly (Snyder, 1974). Snyder (1987) has argued that self-monitoring influences consumer behavior because it is linked with consumers' interest in leaving a positive impression that convey an image of the self to other people. This interest make high self-monitor appear to be different people in different situations (Browne and Kaldenberg, 1997).
If high self-monitors emphasize leaving positive impressions, one might ask what relationship self-monitoring has to being an impulse buyer. Impulse buying is a behavior which the literature and consumers both state is normatively wrong (Hausman, 2000; Bellenger et al., 1978; Cobb and Hoyer, 1986; Han et al., 1991; Kollat and Willet, 1967; Rook and Fisher, 1995; Weinberg and Gottwald, 1982). These negative evaluations of impulse buying behaviors originate from psychological studies of impulsiveness that characterize impulse behaviors as signs of immaturity and lacking of behavioral control (Levy, 1976; Solnick et al., 1980; Vohs and Faber, 2007) or as irrational, risky, and wasteful (Ainslie, 1975; Levy, 1976; Rook and Fisher, 1995; Solnick, et al. 1980). Sharma, Sivakurmaran, and Marshal (2010) found that self-monitoring has a negative association with impulse buying. This personality trait leads low and high self-monitors to show different behaviors in various consumer behavior contexts, including impulsive behavior contexts. For instance, high self-monitors may have a desire to appear rational when they feel that their decisions may come under scrutiny by others (Lerner and Tetlock, 1999; Sharma et al., 2010). Since impulse buying is sometimes being perceived by society as normatively wrong, which results in post-purchase negative affect, guilt, and unfavorable evaluation of purchase decision (Dittmar and Drury, 2000; Rook, 1987; Trocchia and Swinder, 2002), high self-monitors might be less likely to give in to their impulses (Sharma et al., 2010). Also, high self-monitors may have greater motivation compared to low self-monitors to control their impulses simply because they want to appear rational (Luo, 2005).
Although tendencies to be either a high or low self-monitor have been related to attention to many marketing activities including materialism and impulse buying, in the current research the author argues that this personality trait may explain the boundary for SME. In their research, Dholakia et al. (2005) believe that the desire for impulsive options is a limited motivational resource of the decision maker. Meaning, participation in a prior impulsive task consumes this motivational resource, which results in its depletion. Therefore, individuals may experience less desire for impulsive options in the subsequent choice. Thus, in the current research, the author believes that SME works for low self-monitors, but not high self-monitors. For one thing, since high self-monitoring individuals care more about their external image and less about their personal value systems, they may not use up their motivational resources to buy impulsively at all simply because impulse buying could be a socially unacceptable behavior. Also, since these individuals have the tendencies to maintain a consistent self-presentation across situations that harm their image, they will maintain their impulsivity level in the second impulsive task. Furthermore, since high self-monitors have greater motivation compared to low self-monitors to control their impulses (Luo, 2005), they do not deplete their motivational resources for impulsivity in the first impulsive task and hence will sustain their urges in the second task as well.
On the other hand, low self-monitors are more interested in satisfying their personal value systems and private realities (Browne and Kaldenberg, 1997) and hence engaging in a prior impulsive tasks will fulfill their personal value systems and deplete some of their motivational resources for impulsivity. Therefore, unlike high self-monitors, low self-monitors will more likely experience SME and lesser desire for impulsive options in the second impulsive choice, supporting the notion that motivation resources for impulsivity may be consumed in the first choice for low self-monitors.
In the next experiment, the author examines this notion using hypothetical impulsive scenarios in two sequential tasks. As noted by Dholakia et al. (2005), SME applies to sequential impulsive choices where the options between the two choices themselves are not directly related to each other and does not apply to those gambling situations (see Dholakia et al., 2005 for SME constraints). Therefore, this experiment is designed to replicate SME using impulsive scenarios in two unrelated sequential tasks (i.e., impulsive donation and purchase) and show that SME works for low self-monitors, but not high self-monitors (see Research Framework). Also, as noted earlier, since impulse buying can have both negative and positive consequences for the consumers, one may argue that high self-monitors might act differently if they engage in a second impulsive task that has positive consequences, which makes them socially desirable. Therefore, the author shows that engaging in impulsive tasks that have positive social outcomes still result in no SME effect for high self-monitors because the motivational resources for impulsivity will not be consumed in the first impulsive choice; hence, it will not result in a drop in impulsivity in the second impulsive choice. As a result, the author proves that this SME boundary is robust to impulsive tasks that are socially desirable and socially undesirable. Furthermore, the author demonstrates that the effect of SME is maintained even after days of having engaged in the first impulsive task, but limited to low self-monitors.
In this experiment, impulsive choices were operationalized using hypothetical scenarios developed by Dholakia et al. (2005). These scenarios were selected based on high levels of identification with such situations reported by American students of both genders in prior pretests and research (cf., Dholakia et al., 2005). For the current research, the scenarios were slightly altered to fit the study purpose and local culture (as detailed below). Participants were 118 undergraduate students from a large Middle Eastern university. They participated in exchange for partial course credit as a part of a principle of marketing course and were told that they were participating in a series of studies that were designed to better understand consumers' responses. Therefore, they were asked to participate in two research sessions. Each research session lasted for approximately twenty minutes. All questionnaires, including used scales, were translated into Arabic using back to back translations, and reviewed by two bilingual experts in the field for content validity. Translated versions of the questionnaires were agreed upon by the experts. Students were given the questionnaires in the language of their preference, English or Arabic. Both languages were made available because of the bilingual capabilities of the university population. Almost all participants chose the Arabic version (%96.61).
To verify the realism of the scenarios, the author asked a separate group of students from the same sample pool (N = 47) to read each scenario and (a) rate whether they believed the scenario was realistic (1 = strongly disagree to 5 = strongly agree); (b) rate whether they believed the scenario was believable (1 = strongly disagree to 5 = strongly agree), and (c) indicate how likely they would be to encounter a situation similar to the one described in the scenarios (1 = very unlikely to 5 very likely). Reliability analysis indicated that the three items could be combined to form a scale ([alpha] = .85). Importantly, the mean on this three-item realism scale for the least realistic scenario (M = 3.35, SD = 0.99) was above the scale midpoint of 3, t (45) = 2.43, p < .05. Analysis of the other scenarios yielded identical results (all ps < .001). Taken together, these results suggest that participants viewed the scenarios as realistic.
In the first research session, participants completed two individual difference measures, including the trait Buying Impulsiveness Scale (BIS; Rook and Fisher, 1995) and Self-monitoring Scale (SMS; Lennox and Wolfe, 1984; O'Cass, 2000). Then, participants were assigned randomly to one of the two treatments (i.e., experimental and control). Participants in the experimental treatment were asked to complete a section, which consisted of an impulsive scenario and some measures related to the scenario, while participants in the control treatment did not complete any task and were asked to stay silent until the session ended.
In keeping with Rook's original definition of impulsivity, in the current research, the author focuses on the urge to engage in impulsive choices, assuming it represents an important precursor of actual impulsive behavior (cf. Beatty and Ferrell, 1998; Dholakia, 2000; Dholakia et al., 2005; Herabadi, Verplanken, and van Knippenberg, 2009). Thus, participants in the experimental treatment were asked to imagine themselves in either one of these impulsive scenarios: impulsive donation or impulsive purchase (adapted from Dholakia et al., 2005). Two different impulsive scenarios were used to examine the robustness of SME boundary under high or low socially desirable impulsive contexts (impulsive donation, impulsive purchase).
Impulsive donation. "Imagine that you have received a bonus of K.D. 200 (K.D.1 is equivalent to $3.5) from work. A few days later, you unexpectedly get a call from a well-known charity seeking contributions from you."
In response to the scenario, participants indicated their likelihood to give money to a charity with the following item: "What is the likelihood that you would donate to the charity?" with a 7-point scale (1 = Very Unlikely to 7 = Very Likely). Participants' impulsiveness to donate to the charity was measured using the item: "If you were in this situation, you would want to donate to the charity." using a 7-point scale (1 = Strongly Disagree to 7 = Strongly Agree).
Impulsive purchase. "Imagine, on a weekend, after a busy and productive week at work, you go to the mall with your friend to buy a pair of shoes for an up-coming event. As you are walking in the mall, you see a great looking shirt on sale. The helpful salesperson tells you that this shirt is of the most recent style. You also find that the shirt is available in your size and in your favorite color."
In response to the scenario, participants indicated their likelihood to purchase the shirt with the following item: "What is the likelihood that you would purchase the shirt?" with a 7-point scale (1 = Very Unlikely to 7 = Very Likely). Then, participants indicated their impulsiveness to buy the shirt with the following measure: "If you were in this situation, you would want to purchase the shirt." using a 7-point scale (1 = Strongly Disagree to 7 = Strongly Agree).
Five days later, all participants were asked to complete the second research session in order to fulfill the research requirement. As noted by Dholakia et al. (2005), SME applies to sequential impulsive choices where the options between the two choices themselves are not directly related to each other. Therefore, half of the participants in the control treatment and those in the experimental treatment who were given the impulsive donation scenario in the first session were given the following impulsive purchase scenario:
Impulsive purchase. "Imagine that you enjoy exercising and running and like to eat health food. On a weekend, after a busy and productive week at work, you go to the mall with your friends to enjoy your time. Walking through the mall, you and your friends decide to buy dinner from the food court. You head to your favorite restaurant and as you are looking through the menu, you decide to buy a healthy meal and then you see a mouth-watering tray of sweet."
Participants indicated their likelihood to purchase the sweet with the following item: "What is the likelihood that you would purchase the sweet?" with a 7-point scale (1 = Very Unlikely to 7 = Very Likely). Then, participants indicated their impulsiveness to buy the sweet with the following measure: "If you were in this situation, you would want to purchase the sweet." using a 7-point scale (1 = Strongly Disagree to 7 = Strongly Agree).
For the other half of the control treatment participants and those in the experimental treatment who were given the impulsive purchase scenario in the first session were given the following impulsive donation scenario:
Impulsive donation. "Imagine that you have received a letter from the General Manager asks whether you want to donate to the marathon walk, which will be held to help children with cancer. Donation will be made according to the number of kilometers, a half dinar per km. That is, you will donate 10 dinars if you walk 20 km."
In response to the scenario, participants indicated their likelihood to donate to the charity with the following item: "What is the likelihood that you would donate to the charity?" with a 7-point scale (1 = Very Unlikely to 7 = Very Likely). Participants' impulsiveness was measured using the item: "If you were in this situation, you would want to donate to the charity." using a 7-point scale (1 = Strongly Disagree to 7 = Strongly Agree).
The trait BIS ([alpha]= .83) and SMS ([alpha] = .67) were found to be non-significant ([M.sub.Experimental] = 3.69, [M.sub.Control] = 3.92, t (1, 57) = .774, p = .442; [M.sub.Experimental] = 3.69, [M.sub.Control] = 3.92, t (1, 57) = .774, p = .442, respectively) across the two treatments.
It was expected that SME on participants' impulsivity would be qualified by the individual difference in self-monitoring. Moreover, the author anticipated that this moderation should hold under both socially desirable (e.g., impulsive donation) and socially undesirable behaviors (e.g., impulsive purchase). As noted above, to determine the robustness of the SME boundary, the author used two types of impulsive scenarios (impulsive purchase and impulsive donation) to test whether impulsivity type could impact the SME boundary. For each scenario, participants were randomly assigned to either the control or experimental treatment. To analyze the data, the author first conducted a 2 (Treatments: experimental versus control) x 2 (Self-monitoring: high versus low) x 2 (Impulsivity type: purchase versus donation) between-subjects MANOVA on the purchase likelihood and impulsiveness measures (see Table 1). The main goal in this analysis was to determine whether the treatments, self-monitoring, and impulsivity type interacted with one another. For the purchase likelihood and impulsiveness measures, results revealed a non-significant three-way interaction, F (1, 110) = .48, p = .49; F (1, 110) = .49, p = .48, respectively.
Results of the three-way interaction revealed that the SME boundary was not influenced by the impulsivity type. Accordingly, for the primary analysis, the author collapsed across impulsivity type and focused solely on the interaction effect of treatments and self-monitoring on purchase likelihood and impulsiveness measures using a 2 (Treatments: experimental versus control) x 2 (Self-monitoring: high versus low) between-subjects MANOVA on the purchase likelihood and impulsiveness measures (see Table 2).
For the purchase likelihood and impulsiveness measures, the results revealed a significant two-way interaction between treatments and self-monitoring, F (1, 114) = 6.00, p < .05; F (1, 114) = 6.61, p < .01, respectively. Furthermore, in order to understand the two-way interaction, a series of t-tests were conducted to compare low and high self-monitors under both treatments.
As illustrated in Figure 1, for low self-monitors, those in the experimental treatment had lower purchase likelihood than those in the control treatment, [M.sub.Experimental] = 4.77, [M.sub.Control] = 5.67, t (58) = 2.23, p < .05. However, for high self-monitors, the treatments did not differ in the purchase likelihood measure, [M.sub.Experimental] = 3.82, [M.sub.Control] = 4.00, t (56) = 1.37, p =.18.
Similarly, Figure 2 indicates that, for low self-monitors, those in the experimental treatment had lower impulsivity than those in the control treatment, [M.sub.Experimental] = 4.50, [M.sub.Control] = 5.70, t (58) = 2.78, p < .01; while, for high self-monitors, the treatments did not differ in the impulsiveness measure, [M.sub.Experimental] = 5.28, [M.sub.Control] = 4.69, t (56) = 1.07, p =.29.
An additional analysis was conducted to provide further empirical evidence that, in general, low self-monitors (compared to high self-monitors) are more likely to be under the influence of SME. In order to conduct this analysis, an index was created by computing the difference in dependent measures (purchase likelihood and impulsiveness measures) between the first-session and second-session for participants in the experimental treatment. A one-sample t-test showed that the difference in dependent measures for low self-monitors was significantly different from zero for the purchase likelihood measure (t (28) = 2.47, p < .05) and impulsiveness measure (t (28) = 2.21, p < .05), while the difference in dependent measures for high self-monitors was non-significant for the purchase likelihood measure (t (26) = 0.00, p = 1.00) and impulsiveness measure (t (26) = .19, p = .85). These results provide further support to the notion that SME is more likely to occur for low self-monitors than high self-monitors. That is, low self-monitors have less desire (compared to high self-monitors) to behave impulsively in the second choice following days of having engaged in the first impulsive choice, which is caused by the depletion of motivational resources.
The primary goal of this research was to evaluate the boundary for SME and examine whether the individual difference in self-monitoring would influence the occurrence of SME. Also, the secondary goal was to test the robustness of SME by showing that individuals are vulnerable to SME under different levels of socially desirable impulsive choices. Thus, the results indicated that, in different impulsive contexts, the difference in impulsivity between the treatments was significant for low self-monitors, while this difference was non-significant for high self-monitors. These findings provide empirical evidence that low self-monitors' motivational resources were in fact being depleted after engaging in a prior impulsive task and hence resulted in SME. However, the deletion of these resources requires that the person develops a goal to act impulsively in the first place. Meaning, if the person does not have tendencies or goals to engage in impulsive tasks, the person will not consume the motivational resources for impulsivity in the first place and hence SME will not occur. Since high self-monitors have greater motivation to control their urges under different social contexts and do not have tendencies to act impulsively simply because engaging in impulsive behaviors may have a negative representation in societies, these individuals attempt to appear normal and hence do not act impulsively in the first place to protect their image.
Possible explanation of the SME boundary
In general, motivation is defined as "the process of allocating personal resources in the form of time and energy to various acts in such a way that the anticipated affect resulting from these acts is maximized." (Naylor, Pritchard, and Ilgen, 1980; p. 7). Since low self-monitors are more likely to allocate personal resources to satisfy their personal value systems and private realities (Browne and Kaldenberg, 1997), engaging in a prior impulsive tasks will fulfill their personal value systems and hence deplete some of their personal resources for impulsivity in subsequent impulsive tasks. On the other hand, since high self-monitors care more about their social image in the society, they have less motivation (compared to low self-monitors) to engage in impulsive purchases simply because impulsivity is sometimes viewed by society as normatively wrong (Dittmar and Drury, 2000; Rook, 1987; Trocchia and Swinder, 2002). Therefore, the different motivational directions that low and high self-monitors have toward impulsivity cause them to act differently when they encounter sequential impulsive tasks and hence results in the SME boundary.
This research has important managerial and practical implications. First, most consumer behavior or decision-making is based on repeated decisions (Bagozzi, 1981; Betsch et al., 1998; Betsch et al., 2001; Betsch and Haberstroh, 2005; Bentler and Speckart, 1979; Norman and Smith, 1995; Verplanken et al., 1997). In these repeated-buying situations, previous decisions can systematically influence later ones. This repeated-decision situation can also occur frequently in impulsive purchases. Specifically, consumers are frequently exposed to repeated situations in retail stores that tempt them to buy things impulsively. If retailers create situations in which impulsive purchases occur in subsequent manners, they would want to know whether this could have an effect on consumers or whether there are limitations for this effect. In sum, this research can provide managerial implications for marketing activities, including price management, product positioning, and product display. With various marketing mix tools, marketing managers can easily influence consumers' decision experiences. These different experiences can then affect subsequent choices. Marketing managers can strategically use decision experiences to maximize their profits based on the findings of this research.
LIMITATIONS AND FUTURE RESEARCH
In this research, the author attempted to show the boundary for and robustness of SME. Yet, like any other study, this research has a few limitations, which may impact the findings. For one thing, this paper used a hypothetical scenario to operationalize impulsive behaviors. Although this approach has been used in previous research (Dholakia et al., 2005), future researchers should look into this issue and test the concept in real-world settings. Also, while the present results were consistent with the author's predictions, the present study did not directly test for the mediating mechanism underlying the examined effect. The author believes that high self-monitors have the tendencies to maintain a consistent self-presentation across situations that harm their image and hence they do not engage in impulsive behaviors in the first place, which in turn makes them less susceptible to SME. Unlike high self-monitors, low self-monitors are more interested in satisfying their personal vales and hence engaging in a prior impulsive tasks fulfills their personal value systems and depletes some of their motivational resources for impulsivity. Therefore, unlike high self-monitors, low self-monitors will more likely experience SME and lesser desire for impulsive options in the second impulsive choice. Future researchers could help provide empirical evidence to support the suggested mediating mechanism that causes low self-monitors to be under the influence of SME while high self-monitors to be less influenced by it.
Ainslie, G. (1975). Specious reward: a behavioral theory of impulsiveness and impulse control. Psychological Bulletin, 82(4), 463-96.
Bagozzi, Richard P. (1981). Attitudes, Intentions, and Behavior: A Test of Some Key Hypotheses. Journal of
Personality and Social Psychology, 41, 607-627. Bellenger, D.N., D.H. Robertson, & E.C. Hirschman (1978). Impulse buying varies by product'. Journal of
Advertising Research, 18, 15-18. Bentler, Peter M. & George Speckart (1979). Models of Attitude-Behavior Relations. Psychological Review, 86, 452-464.
Betsch, Tilmann, Klaus Fiedler, & Julia Brinkmann (1998). Behavioral Routines in Decision Making: The Effects of Novelty in Task Presentation and Time Pressure on Routine Maintenance and Deviation. European Journal of Social Psychology, 28, 861-878.
Betsch, Tilmann, Susanne Haberstroh, Andreas Glockner, Thomas Haar, & Klaus Fiedler (2001). The Effects of Routine Strength on Adaptation and Information Search in Recurrent Decision Making. Organizational
Behavior and Human Decision Processes, 84, 23-53.
Betsch, Tilmann & Susanne Haberstroh (2005). The Routines of Decision Making, Lawrence Erlbaum Association.
Browne, B.A. & D.O. Kaldenberg (1997). Conceptualizing self-monitoring: links to materialism and product involvement. Journal of Consumer Marketing, 14, 31-44.
Cobb, C.J. & W.D. Hoyer (1986). Planned Versus Impulse Purchase Behavior. Journal of Retailing, 62(4), 384-409.
Dittmar, H. & J. Drury (2000). Self-image--is it in the bag? A qualitative comparison between "ordinary" and "excessive" consumers. Journal of Econ Psychology 21, 109-42.
Dholakia, U.M. (2000). Temptation and resistance: an integrated model of consumption impulse formation and enactment. Psychology and Marketing, 17, 955-982.
Dholakia, U.M., M. Gopinath & R. P. Bagozzi (2005). The role of desires in sequential impulsive choices. Organizational Behavior and Human Decision Processes, 98, 179-194.
Dolliver, Mark (1998, May), I'll Just Grab This, Adweek, 39, 19.
Han, Y.K., G.A. Morgan, A. Kotsiopulos, & J. Kang-Park (1991). Impulse buying behavior of apparel purchasers. Clothing and Textile Research Journal, 9(3), 15-21.
Hausman, A. (2000). A multi-method investigation of consumer motivations in impulse buying behavior. Journal of Consumer Marketing, 17, 403-419.
Herabadi, A.G., B. Verplanken, & A. van Knippenberg (2009). Consumption experience of impulsive buying in Indonesia: emotional arousal and hedonistic considerations. Asian Journal of Social Psychology, 12, 20-31.
Hoch, S.J. & G.F. Loewenstein (1991). Time-in-consistent preferences ad consumer self-control. Journal of Consumer Research, 17(4), 492-507.
Jones, M.A., K.E. Reynolds, S. Weun, & S.E Beatty (2003). The-product-specific nature of impulse buying tendency. Journal of Business Research, 56(7), 505-511.
Kacen, J.J. & J.A. Lee (2002). The influence of culture in consumer impulsive buying behavior. Journal of Consumer Psychology, 12(2), 163-176.
Kollat, D.T. & R.P. Willett (1967). Consumer impulse purchasing behavior. Journal of Marketing Research, 4, 21-31.
Lennox, R.D & R.N. Wolfe (1984). Revision of the self-monitoring scale. Journal of Personal Social Psychology, 46, 1349-64.
Lerner, J.S. & P.E. Tetlock (1999). Accounting for the effects of accountability. Psychology Bulletin, 125, 255-75.
Levy, M. (1976). Deferred gratification and social class. Journal of Social Psychology, 100, 123-35.
Luo, X. (2005). How does shopping with others influence impulsive purchasing? Journal of Consumer Psychology, 15(4), 288-94.
Mogelonsky, Marcia, (1998, July). Keep Candy in the Aisles. American Demographics 20, 32.
Naylor, J. C., Pritchard, R. D., & Ilgen, D. R. (1980). A theory of behavior in organizations. New York: Academic Press.
Norman, Paul & Lawrence Smith (1995). The Theory of Planned Behaviour and Exercise: An Investigation into the Role of Prior Behaviour, Behavioural Intentions and Attitude Variability. European Journal of Social Psychology, 25(4), 403-415.
O'Cass, A. (2000). A psychometric evaluation of a revised version of the Lennox and Wolfe revised self-monitoring scale. Psycho Marketing, 17(5), 397-419.
Piron, F. (1991). Defining Impulse Purchasing. Advances in Consumer Research, 18, 509-514.
Rook, D.W. (1987). The buying impulse. Journal of Consumer Research, 14, 189-99.
Rook, D.W. & R.J. Fisher (1995). Normative influences on impulsive buying behavior. Journal of Consumer Research, 22, 305-13.
Rook, D.W. & S.J. Hoch (1985). Consuming impulses. Advances in Consumer research, 12, 23-27.
Sharma, P., B. Sivakumaran, & R. Marshall (2010). Impulse buying and variety seeking: a trait-correlates perspective. Journal of Business Research, 63, 276-283.
Snyder, M. (1974). Self-monitoring of expressive behavior. Journal of Personality and Social Psychology, 30, 526-37.
Snyder, M. (1987). Public Appearances and Private Realities. W.H. Freeman and Co., New York, NY.
Solnick, J.V., C.H. Kannenberg, D.A. Eckerman, & M.B. Waller (1980). An experimental analysis of impulsivity and impulse control in humans. Learning and Motivation, 11, 61-77.
Stern, Hawkins (1962). The Significance of Impulsive Buying Today. Journal of Marketing, 26 (April), 59-62.
Trocchia, P.J. & J. Swinder (2002). An investigation of product purchase and subsequent non-consumption. Journal of Consumer Marketing, 19, 188-204.
Verplanken, Bas, Henk Aarts, & Ad van Knippenberg (1997). Habit, Information Acquisition, and the Process of Making Travel Mode Choice. European Journal of Social Psychology, 27, 539-560.
Vohs, K.D. & R.J. Faber (2007). Spent resources: self-regulatory resource availability affects impulse buying. Journal of Consumer Research, 33(4), 537-547.
Abdullah J. Sultan, Kuwait University
Table 1: Three-Way Interaction Source DV Type III SS Df MS Corrected Model Purchase Likelihood 79.52 (a) 7 11.36 Impulsiveness 97.77 (b) 7 13.97 Intercept Purchase Likelihood 2822.58 1 2822.58 Impulsiveness 2801.52 1 2801.52 Self-Monitoring Purchase Likelihood 2.87 1 2.87 Impulsiveness 0.86 1 0.85 Treatments Purchase Likelihood 3.88 1 3.88 Impulsiveness 12.43 1 12.43 Impulsivity Type Purchase Likelihood 37.79 1 37.79 Impulsiveness 59.21 1 59.21 Self-Monitoring Purchase Likelihood 10.69 1 10.69 * Treatments Impulsiveness 15.71 1 15.71 Self-Monitoring Purchase Likelihood 17.85 1 17.85 * Impulsivity Type Impulsiveness 10.54 1 10.54 Treatments * Purchase Likelihood 1.17 1 1.17 Impulsivity Type Impulsiveness 0.03 1 0.03 Self-Monitoring * Purchase Likelihood 1.37 1 1.37 Treatments * Impulsiveness 1.49 1 1.49 Impulsivity Type Error Purchase Likelihood 311.64 110 2.83 Impulsiveness 335.02 110 3.05 Total Purchase Likelihood 3442.00 118 Impulsiveness 3433.00 118 Corrected Total Purchase Likelihood 391.15 117 Impulsiveness 432.79 117 Source F P Corrected Model 4.01 .001 4.58 .000 Intercept 996.30 .000 919.86 .000 Self-Monitoring 1.01 .317 0.28 .597 Treatments 1.37 .244 4.08 .046 Impulsivity Type 13.34 .000 19.44 .000 Self-Monitoring 3.77 .055 * Treatments 5.16 .025 Self-Monitoring 6.30 .014 * Impulsivity Type 3.46 .065 Treatments * 0.41 .521 Impulsivity Type 0.01 .923 Self-Monitoring * 0.48 .488 Treatments * 0.49 .485 Impulsivity Type Error Total Corrected Total Table 2: Two-Way Interaction Source DV Type III SS df MS Corrected Model Purchase Likelihood 21.88a 3 7.29 Impulsiveness 26.99b 3 8.99 Intercept Purchase Likelihood 3047.24 1 3047.24 Impulsiveness 2998.17 1 2998.17 Self-Monitoring Purchase Likelihood 2.12 1 2.12 Impulsiveness 0.41 1 0.41 Treatments Purchase Likelihood 0.23 1 0.23 Impulsiveness 2.78 1 2.78 Self-Monitoring Purchase Likelihood 19.45 1 19.45 * Treatments Impulsiveness 23.52 1 23.52 Error Purchase Likelihood 369.27 114 3.24 Impulsiveness 405.80 114 3.56 Total Purchase Likelihood 3442.00 118 Impulsiveness 3433.00 118 Corrected Total Purchase Likelihood 391.15 117 Impulsiveness 432.79 117 Source F P Corrected Model 2.25 .086 2.53 .061 Intercept 940.72 .000 842.26 .000 Self-Monitoring 0.66 .420 0.11 .736 Treatments 0.07 .791 0.78 .379 Self-Monitoring 6.00 .016 * Treatments 6.61 .011 Error Total Corrected Total Figure 1: Using self-monitoring to show the boundary for SME on purchase likelihood Self-monitoring Purchase likelihood Control Experimental Low SM 5.67 4.77 High SM 4.59 5.31 Note: Table made from bar graph. Figure 2: Using self-monitoring to show the boundary for SME on impulsiveness Impulsiveness Control Experimental Low SM 5.70 4.50 High SM 4.69 5.28 Note: Table made from bar graph.