Does Personality Affect Compulsive Buying? An Application of the Big Five Personality Model

Compulsive Buying (CB) is ‘a consumer’s tendency to be preoccupied with buying that is revealed through repetitive buying and a lack of impulse control’ (Ridgway et al., 2008). Although Kraepelin (1915) and Bleuler (1976) described this phenomenon many years ago, it was the study of Faber et al. (1987) which increased the interest in that topic. For more than 20 years empirical studies have been conducted to increase our knowledge about this maladaptive consumer behavior. The main focus of these studies has been on identifying the antecedents and consequences of CB. First of all, CB is viewed as a method of coping with life’s challenges (Scherhorn, 1990), low self-esteem (d’Astous et al., 1990; Hanley & Wilhelm, 1992; O’Guinn & Faber, 1989) and/or internal psychological tension (Valence et al., 1988). It is also seen as a process whereby the consumer wants to escape from negative feelings, stress and anxiety (Edwards, 1993; Ergin, 2010; O’Guinn & Faber, 1989). Buying a product causes release of the internal tension. At the same time however feelings of guilt and frustration appear as the person could not control the inner drive to buy an unneeded product. The consumer is caught in a vicious circle.


Introduction
Compulsive Buying (CB) is 'a consumer's tendency to be preoccupied with buying that is revealed through repetitive buying and a lack of impulse control' (Ridgway et al., 2008).Although Kraepelin (1915) and Bleuler (1976) described this phenomenon many years ago, it was the study of Faber et al. (1987) which increased the interest in that topic.For more than 20 years empirical studies have been conducted to increase our knowledge about this maladaptive consumer behavior.The main focus of these studies has been on identifying the antecedents and consequences of CB.First of all, CB is viewed as a method of coping with life's challenges (Scherhorn, 1990), low self-esteem (d' Astous et al., 1990;Hanley & Wilhelm, 1992;O'Guinn & Faber, 1989) and/or internal psychological tension (Valence et al., 1988).It is also seen as a process whereby the consumer wants to escape from negative feelings, stress and anxiety (Edwards, 1993;Ergin, 2010;O'Guinn & Faber, 1989).Buying a product causes release of the internal tension.At the same time however feelings of guilt and frustration appear as the person could not control the inner drive to buy an unneeded product.The consumer is caught in a vicious circle.
Despite its negative consequences and a high comorbidity with mood, substance use and eating disorders (McElroy et al., 1994;Schlosser et al., 1994), CB is not officially recognized as a mental disorder (American Psychiatric Association [APA], 2000).Moreover, there is a disagreement about the nosology of CB: some researchers consider CB as an addictive disorder (Scherhorn, 1990), others as an obsessive-compulsive disorder (Ridgway et al., 2008).Nonetheless, there exist criteria for compulsive buying disorder (CBD), widely accepted by the researchers in the domain: 1) frequent preoccupation with shopping or irresistible buying impulses; 2) buying more than is needed or/and can be afforded; 3) distress related to buying behavior; 4) significant interference with work or social areas of functioning (McElroy et al., 1994).Currently CB is most often diagnosed as 'Impulse Control Disorder Not Otherwise Specified'.
Though no consensus about the classification of CB as a mental disorder was yet found there is no doubt that CB is a serious psychological problem which causes significant impairments in financial, legal, social and occupational areas of functioning of an affected person.Therefore it is of major importance to investigate the nature of CB and its antecedents in more detail.Knowing which variables significantly influence CB will help clinicians and affected persons to understand what the 'driving force' behind this maladaptive behavior is and to address the potential source of the problem during the therapy.
Because of their stability, personality traits are considered very useful in predicting behaviour.Although few studies were conducted in order to find the link between CB and personality (eg., Mowen & Spears, 1999;Balabanis, 2001;Wang & Yang, 2008), the inconsistent results prevent from drawing reliable conclusions.The aim of the current study is to explore the causes of these inconsistencies and to give an ultimate answer to the question whether Big Five personality traits differentiate compulsive and non-compulsive buyers.

Compulsive buying and the Big Five
'Personality' is 'the set of psychological traits and mechanisms within the individual that are organized and relatively enduring and that influence his or her interactions with, and adaptations to, the intrapsychic, physical, and social environments' (Larsen & Buss, 2010, p. 4).For many years efforts in research on personality have been focused on identifying the elemental personality traits.In 1934 Thurstone suggested that there are five independent common factors underlying personality, a notion further supported by Fiske (1949), Tupes & Christal (1961) and a host of other researchers (e.g., Costa & McCrae, 1985;John, 1999;Goldberg, 1992;Wiggins, 1996).The Five Factor model of personality, also known as the Big Five (Goldberg, 1992) (Costa and McCrae, 1992).
Since the Big Five has achieved the greatest degree of consensus of all trait taxonomies (Larsen & Buss, 2010), it has been used to explain individual differences in a large amount of research (e.g., Endler & Spear, 1998).Although the first attempts to understand and predict consumer behavior through the use of personality variables have yielded disappointing results (Kassarjian, 1971), there has recently been a growing interest in personality influences in consumer research (e.g., Egan & Taylor, 2010;Lin, 2010;Mowen, 2000).The first attempt to identify a link between personality and compulsive buying behaviour was made by Mowen and Spears (1999).They employed Allport's hierarchical approach where surface traits are explained by central traits which are predicted by cardinal psychological traits.With regard to this approach the Five Factor Model of personality was used to explain materialism and the needs for arousal, which in turn predict compulsive buying behaviour.With respect to personality traits the results showed that low Conscientiousness, high Agreeableness and low Stability predict compulsive buying directly.
A similar approach was applied in two studies by Mowen (2000) with impulsiveness and compulsiveness as central traits.When data were analyzed excluding mediators from the model, only two personality traits -Neuroticism and Agreeableness, appeared to be significant predictors of compulsive buying.
Also Balabanis (2001) used the Five Factor Personality Model to explain compulsive buying behaviour.The results indicated that Extraversion and Intellect, the two factors which did not appear to directly influence compulsive buying in previous studies, were found to be related to buying compulsiveness in lottery tickets and scratch-cards.Extraversion was positively and Intellect negatively related to compulsive buying behaviour.Surprisingly, in contradiction to the findings by Mowen and Spears (1999), Balabanis (2001) found a negative correlation between Agreeableness and CB.
A few years later Wang and Yang (2008) published their paper on the influence of personality and compulsive buying in an online shopping context.In accordance with the findings of Mowen and Spears (1999), the researchers found Stability to be negatively correlated with compulsiveness in buying.With regard to the other Big Five personality traits no significant differences between compulsive and non-compulsive buyers were found.
It is important to notice that also Schlosser et al. (1994) and Mueller et al. (2010) were interested in personality influences on CB.However since their studies were conducted in clinical samples, their aim was not to compare compulsive with non-compulsive buyers in a general consumer population but to 'assess the overall life-style and problems of subjects already recognised as compulsive shoppers' (Schlosser et al., 1994, p. 205) or 'identify personality prototypes in treatment seeking patients with CB' (Mueller et al., 2010, p. 930 As demonstrated above, with the exception of Neuroticism, no consistent conclusions can be drawn from the previous research on the relationship between compulsive buying behavior and personality (see Table 1).This can be due to the fact that these studies were investigating CB behavior in different buying contexts.For example Balabanis (2001) focused on scratch-cards and lottery tickets buying behavior whereas other studies examined compulsive buying behavior in general.A second ground for the inconsistent results might be due to the different samples investigated.The majority of above mentioned studies neglect to reveal the percentage of compulsive buyers in their sample.It is therefore possible that the authors were investigating the (direct or indirect) link between (compulsive) buying tendency (measured on a continuum basis) and personality with a very limited amount of actual compulsive buyers in their sample.Furthermore, all research samples of above mentioned studies under consideration consisted of only students.
Although research based on student samples might be a useful source of information, it cannot be generalized to the buying population.
As a result the question about the relationship between CB and personality still remains unresolved.Taking into account the generally established significance of personality in determining consumer behavior (Haugtvedt et al., 1992;Kassarjian, 1971), the objective of the current study is to explore personality differences between consumers with and without compulsive buying tendency.In this study the above mentioned shortcomings of previous research will be addressed and the prevalence of compulsive buyers in the general (not exceptionally student) consumer population in Flanders will be revealed.
'Agreeableness' involves trust, sympathy and altruism (Costa & McCrae, 1992).It deals with motives for maintaining positive relations with others (Jensen-Campbell & Graziano, 2001).The fact that compulsive buying is not a socially desirable behavior and therefore can prevent maintaining positive relations with others leads to the development of the second hypothesis: www.intechopen.comDoes Personality Affect Compulsive Buying?An Application of the Big Five Personality Model 135 H2: Compulsive buyers score lower on 'Agreeableness' than non-compulsive buyers.
People scoring high on 'Conscientiousness' are well-organized and consider the consequences of their actions carefully.Since compulsive buying behavior is an impulsive, irresponsible act we expect that: H3: Compulsive buyers score lower on 'Conscientiousness' than non-compulsive buyers.
Individuals scoring high on 'Neuroticism' tend to have low self-esteem and high levels of anxiety and depression (Costa & McCrae, 1985).Each of these variables has been associated with compulsive buying (Edwards, 1993;Ergin, 2010;O'Guinn & Faber, 1989), hence we hypothesize that: H4: Compulsive buyers score higher on 'Neuroticism' than non-compulsive buyers.
'Intellect' is characterized by original, imaginative and broad interests (McCrae & Costa, 1985).Since people scoring high on this trait are intellectually curious and have a better perception of reality, we do not expect them to engage in compulsive buying behavior.Therefore we presume that: H5: Compulsive buyers score lower on 'Intellect' than non-compulsive buyers.

Participants and procedure
Data for the current study were collected over a three-month period in the beginning of 2010 in Flanders, the Dutch speaking part of Belgium.To obtain a representative sample of the Flemish consumer population we have used several ways to spread our online questionnaire.First of all we used invitations on online forums such as Flair and Libelle (women magazines popular in Belgium), a forum for shopping addicts (http://www.yes.nl/cafe/) and a forum for seniors (http://www.seniorennet.be/forum/).We have advertised the survey in our university newsletter and asked colleagues and friends to forward the questionnaire.To encourage potential respondents we promised an incentive of €100 for one randomly chosen respondent.
A total number of 2298 respondents actually participated in the study.After a careful data cleaning procedure (only Flemish adults, aged 18 years or older, who answered all the questions within a reasonable time frame, were taken into account), 1522 participants were retained from the primary dataset serving as the basis for the further analyses.The sample was clearly dominated by female participants (72,6%).Approximately 80% of the respondents was younger than 30 years with 21-24 years old participants being represented most (42%).Taking into account the focus of our study (general consumer population) and the fact that young females do the majority of shopping, this does not affect the representativity of our sample.

Measures
Although the majority of the researchers in the domain use the clinical screener of Faber and O'Guinn (1992) to measure compulsive buying tendency, Ridgway et al. (2008) identified two major shortcomings of this scale.First of all, it does not measure the obsessive-compulsive dimension of buying, since all the items focus on the impulse control aspect.Secondly, items concerning income and financial consequences prevent consumers with higher incomes from being recognized as compulsive buyers, as they can afford their unusual spending habits.Ridgway et al. (2008, p. 625) conclude that 'unless consumers have already been identified with a buying disorder, the clinical screener may misclassify some of them as compulsive buyers, because of its dependence on income-related items and the weights assigned to those items in the scoring algorithm'.Moreover, according to Ridgway et al. (2008), compulsive buying measurements should not focus on identifying consequences of maladaptive consumer behaviour, but on the underlying behavioural tendencies.
Therefore we opted to measure compulsive buying tendency with the 6-item scale developed by Ridgway et al. (2008).This scale does not include items concerning income and financial consequences, it incorporates both characteristics of obsessive-compulsive behaviour as well as the impulse-control dimensions of buying and is the first to appropriately assess the extent of compulsive buying tendency in the general population of consumers.Four items are measured on 7-point Likert scale from 'strongly disagree' to 'strongly agree' ('My closet has unopened shopping bags in it', 'Others might consider me a shopaholic', 'Much of my life centers around buying things' and 'I consider myself an impulse purchaser') and two items are measured on a 7-point scale from 'never' to 'very often' ('I buy things I don't need' and 'I buy things I did not plan to buy') (all items translated to Dutch with back-translation following international guidelines on test translation1 ).According to the guidelines provided by Ridgway et al. (2008), respondents who obtained a score of 25 or more points are considered compulsive buyers.
The Mini-IPIP measure of personality (Donnellan et al., 2006) was used to assess the respondents' Big Five personality traits.The scale consists of 20 items and is a short form of the 50-item International Personality Item Pool-Five-Factor Model measure (Goldberg, 1999).With 4 items per trait (items translated to Dutch with back-translation following international guidelines on test translation), the scale assesses the Big Five personality dimensions namely Extraversion, Agreeableness, Conscientiousness, Neuroticism and Imagination/Intellect.Confirmatory factor analysis performed in AMOS 18.0 revealed an acceptable goodness of fit and composite reliability (CR) of the scale in the current study.

Confirmatory factor analysis of the Big Five measure
The goodness of fit of the Big Five model of Donnellan et al. (2006) was verified in our sample.Though all items loaded sufficiently on their scales (see Graph 1), the initial model indicated a poor fit.
Therefore three changes were entered in the model.Firstly, one item of Neuroticism ('I am relaxed most of the time') was deleted, since it seemed to load not only on the latent variable Neuroticism but also on Extraversion.Secondly, the necessity of the introduction of 3 paths of covariance between two error terms of Conscientiousness, two error terms of Agreeableness and two of Intellect were indicated by the modification indices2 .Finally the error terms of one item of Neuroticism and one of Intellect were linked.This lead to an acceptable goodness of fit: χ² (df=138)= 994; GFI= .932;AGFI= .907and RMSEA= .064.Since χ² statics are influenced by the sample size (Dickey, 1996) other test statistics were taken into consideration when evaluating the model fit.The Goodness of Fit Index and the Adjusted Goodness of Fit Index well exceed the required minimum levels of respectively .85 and .80.Also the Root Mean Square error of approximation was below the cut-off value of .08 (Brown & Cudeck, 1993).

Differences between compulsive and non-compulsive buyers
Using the cut-off point of 25 validated by Ridgway et al. (2008, see measurements for further explanation), 93 out of 1522 respondents (6,1%) were classified as compulsive buyers (see Table 2).There were significantly more women in the compulsive buying group than in the non-compulsive buying group (Chi²= 23,771; p<,001).There were also significant differences in age between both groups (Chi²= 16,710; p=,005).Compulsive buyers seem to be younger than other consumers (average age 22years for compulsive and 26years for non-compulsive buyers, the difference is significant at p<.000).With regard to education (Chi²= 3,385;p=,336) and income (Chi²= 1,719; p=,887) no significant differences between compulsive and noncompulsive buyers were found.

Conclusion
Since no consistent findings could be drawn from previous research on the relationship between CB and personality, the aim of this study was to put the findings of the respective studies into perspective and to repeat the research on a large sample and with a reliable measure of compulsive buying behaviour in the general buying population.In the sample of 1522 respondents, 93 (6,1%) were classified as compulsive buyers.This is in line with previous research on the topic (e.g., Black, 2007;Koran et al., 2006).The research revealed significant demographical differences between compulsive and non-compulsive buyers, with young women having the most chance to buy compulsively.This tendency can be due to fact, that emotional and identity-related dimensions of shopping are more important for women than for men (Babin et al., 1994;Dittmar et al., 2004) and reflect the developmental needs of young people to explore consumer activities or to establish an adult identity by material goods (Dittmar, 2005).
The findings of the current study confirm the utility of the personality approach to comprehend consumer behaviour and extend our understanding of compulsive buying behaviour.The study provides strong empirical support that personality does have an impact on CB.First of all, the results confirm H1, meaning that outgoing, sociable, active people are more prone to buy compulsively.Opposite to what we expected (H2), the t-test results showed that compulsive buyers score higher on Agreeableness as compared to noncompulsive buyers.This effect deserves extra attention and can explain why Balabanis (2001) found a negative relation between these two variables.If we think of compulsive buying as a not socially desirable behaviour we expect Agreeable people to score lower on CB since they are highly motivated to maintain positive relations with others (Jensen-Campbell & Graziano, 2001).For scratch-cards and lottery ticket buyers this appeared to be true (Balabanis, 2001) but not in our research where a general tendency to buy was measured.This is an important finding as it shows that Agreeableness can be sensitive to the retail context under investigation and might point out, that excessive buying does not always have to be considered as socially undesirable.

www.intechopen.com
Psychology -Selected Papers 140 Also H4 and H5 were confirmed by the study.As expected, Neuroticism was positively related to CB. Emotional unstable people who often experience a variety of negative feelings, such as anxiety, depression, anger or embarrassment (McCrae & Costa, 1985) are more prone to buy compulsively.On the other hand, Intellect was negatively related to CB which means that the more imaginative, daring, reasonable a person is, the lower tendency she or he has to buy compulsively.Finally, the results didn't confirm H3 meaning that the trait Conscientiousness does not differentiate between compulsive and non-compulsive buyers.This means that responsibility, precision and self-discipline are characteristic which neither prevent nor make people more prone to engage in excessive buying behavior.
Some limitations of the current study have to be reported.First of all in any given Internet community, there are some individuals who are more likely than others to complete an online survey.This tendency can lead to a systematic bias known as self-selection bias (Thompson et al., 2003).Secondly, the focus of the current study was an investigation of the differences between compulsive and non-compulsive buyers with regard to the Big Five personality traits.It would be interesting to see whether different, more specific measures of personality could predict CB better (eg., Personality Adjective Check List, Strack ,1991).Furthermore, since our study didn't focus on a particular retail context, it would be interesting to investigate whether the prevalence of CB and the personality of compulsive buyers differ depending on the shopping context.Finally, with online shopping gaining in importance, additional studies on online compulsive buying behaviour should be undertaken.
The current research provides strong empirical evidence that compulsive buying is very present in the Flemish consumer population.Based on the representative consumer sample this study revealed that some specific inner characteristics of a consumer can make a person more prone to CB.How individuals seek reinforcement is valuable information for the affected individuals and their families, and can be taken into account by clinicians working with consumers seeking help.It is important to realize that the consumers' personality will influence help-seeking behaviour and also responses to treatment of compulsive buying.

Table 1 .
). Relationship between CB and the Big Five Personality Traits: the Overview of Previous Research Results

Table 2 .
Socio-Demographic Profile of the Respondents Independent sample t-tests were conducted between compulsive and non-compulsive buyers regarding their personality traits (see Table3).Compulsive buyers appeared to be significantly more Extravert (p<.001, d=.50),Neurotic (p<.001, d=.36)and Agreeable (p=.003, d=.35) in comparison to non-compulsive consumers.On the other hand, non-compulsive buyers scored higher on the trait Intellect (p<.001, d=.35) when compared to compulsive buyers.No significant difference between both groups could be revealed with regard to the trait Conscientiousness (p= .089,d=.20).

Table 3 .
Differences in the Average Scores on the Big Five Personality Measure between Compulsive and Non-Compulsive Buyers