Psychological Reports, 2000,86,763-770. Q Psychological Reports 2000

COMPULSIVE BUYING AND RISKY BEHAVIOR AMONG ADOLESCENTS ' JAMES A. ROBERTS AND JOHN F. TANNER. JR.

Summary.-Using a scale designed to measure compulsive buying of adolescents, a survey was conducted to assess the relationships of compulsive buying with several risky behaviors among 111 adolescents benveen the ages of 12 to 19 years. The correlacions suggcsr that self-report measures of cigarette smoking, alcohol use, drug use, and hkel~hoodof engaging in premariral sex were all significantly associated ( r s = .30 to .32) wich compulsive buying among teenagers. Implications of the results are discussed, and directions For research are offered.

Compulsive buying is characterized by inappropriate shopping and spending behavior that leads to psychosocial impairment (Black, Repertinget, Gaffney, & Gabel, 1998). Negative consequences can include personal distress, financial difficulties, such as bankruptcies and credit card debt, and marital and family disruption (Christenson, Faber, deZwaan, Raymond, Specker, Ekern, Mackenzie, Crosby, Crow, Eckert, Mussell, & Mitchell, 1994; Schlosser, Black, Repertinger, & Freet, 1994). Compulsive buying is also associated wich significant psychiatric comorbidity, including drug and alcohol abuse, sexual activity, depression, and personality disorder (Schlosser, et al., 1994). Compulsive buying is a definable chical syndrome which &splays features of both obsessive compulsive &sorder and impulse control &sorders (Christenson, et al., 1994). Compulsive buying is a logical, albeit negative, consequence of our consumer society. Ic is also quite common, with estimates of the prevalence of compulsive buying ranging from about 2 % to 8 % of the population (Black, e l al., 1998; Roberts, 1998). Cross-addiction is also a common phenomenon, indcating that addctive behavior may have psychological bases (firschrnan, 1992; Black, et al., 1998). When compulsive buying becomes an addtion, the primary reason for buying is no longer to obtain uthty or service from the purchased product but rather to satisfy personal needs through the buying process (Friese & Koenig, 1993). Compulsive buyers have been described as reacting to stress with higher anxiety than noncompulsive buyers (DeSarbo & Edwards, 1996). Escape from anxiety is widely cited as the primary motivation of persons exhibiting -

-

'Please address correspondence and reprint requests to James A. Roberts, Baylor University, Marketing Department, P.O. Box 98007, Waco, TX 76798-8007 or e-mail (Jim-Roberts@ Baylor.edu).

764

J. A. ROBERTS

& J.

F. TANNER, JR.

addictive or compulsive behaviors (DeSarbo & Edwards, 1996). Hirschman (1992) views compulsive buying as an addictive "process" whereby the affected individual seeks escape from stress and resulting anxiety through the compulsive buying activity. Incapable of managing stress and anxiety through internal means, the compulsive buyer resorts to external methods, such as buying, overeating, alcohol and drug abuse, sexual promiscuity, exercising, and other coping strategies. If denied a particular addction, those prone to compulsive or addictive behavior wdl switch from one substance or behavior to another (firschman, 1992). Research by Black, et al. (1998) showed that compulsive buyers were more Uely than control group members to have lifetime mood disorders and to exhibit more than one psychiatric disorder. First degree relatives of compulsive buyers were more likely than other relatives to show depression, alcoholism, and drug use and more psychiatric disorders in general. Black, et al.'s findings (1998) suggest that people may inherit a predisposition for poor impulse control. Whether this transmission occurs via genetics or learning is open to speculation, but research supports a relationship between compulsive buying and other addictive behaviors among family members. Valence, d'Astous, and Fortier (1988) found that, when other family members exhibited some form of dysfunction (alcoholism, gambling, bulimia, depression), compulsive buying scores of the respondents increased. Similarly, d'Astous, Maltais, and Roberge (1990) and Roberts (1998) reported a significant, positive relationship between the respondents' perceptions of their parents' compulsive buying tendencies and their own similar tendencies. - -

Psychiatric Comorbidzty of Compulsive Buying Compulsive buying has been classified as an impulse control disorder (Black, et al., 1998). However, it is unclear whether compulsive buying is related to mood disorders, substance use disorders, or impulse control disorders. Some still argue that compulsive buying possesses features of obsessive compulsive disorders (Christenson, et al., 1994). The evidence is more compelling, however, that compulsive buying is an impulse control dsorder (O'Guinn & Faber, 1989; McElroy, Pope, Hudson, Kleck, & White, 1991; Christenson, Faber, deZwaan, Raymond, Specker, & Mitchell, 1992; Faber, 1992; Christenson, et al., 1994; Schlosser, et al., 1994; Black, et al., 1998). Compulsive buying shares the basic characteristics of impulse control disorders such as (1) repetitive problematic behavior, (2) urges or mounting tension preceding the behavior, and (3) release of tension or following the behavior (Christenson, et al., 1994). Like drug use, compulsive buying can be seen as an attempt to reduce emotional tension (Valence, et al., 1988).

ADOLESCENTS' COMPULSIVE BUYING, RISKS

765

Regardless of how compulsive buying may be classified, the dlsorder is associated with psychiatric comorbidity, particularly mood, anxiety, and substance use disorders (Black, et al., 1998). Specifically, compulsive buying has been linked to alcohol and drug use and sexual promiscuity (Faber, O'Guinn, & Ktych, 1987; O'Guinn & Faber, 1989; Christenson, er al., 1994; McElroy, Kleck, Pope, Smith, & Strakowski, 1994; Schlosser, et al., 1994). Recent research on addiction and excessive behaviors, such as impulse control dsorders, have viewed such disorders as related problems having common characteristics and causes (Faber, 1992). It is Uely that these problem behaviors are the result of a combination of biological, psychological, and sociological factors (Faber, 1992). The primary objective of the present study is to examine the relationship between compulsive buying and several common risky behaviors among a group of adolescents: smoking, drug and alcohol use, and intent to engage in premarital sex. If a h k is established between compulsive buying and other risky behaviors, how we approach the diagnosis and treatment of compulsive buying and other risky behaviors exhibited by adolescents may change. An important criterion for assessing the usefulness of research should be its contribution to society and society's welfare (Peter, 1991).

This study was part of a larger study involving 11 public school dstricts and three privace schools. In the larger project, students were sampled in classes from Grades 7 through 12. Schools were randomly assigned questionnaires. For this study, one middle school with 48 seventh graders and one middle-high school involving 63 students across Grades 7 through 12 participated, out of 59 and 74, respectively. Each school represented a different district. Comparisons were made between the two districts; since there were no significant differences, the data were aggregated for further cornparIson. Students had to return a signed permission slip from their parents to participate in the study; students who did not return the slip were given another task. Since we had to rely on the children to take the permissions slips home and to return them, some nonparticipants could not participate because they lost the slip or forgot it. Reasons for not participating were not recorded for the 21 students who did not participate. Participation rate was 83% for useable questionnaires and a total sample of 111 students. Students ranged from 12 to 19 years of age, with an average age of 13.3 and a standard deviation of 1.3. See Table 1 for a summary of their characteristics. Surveys were conducted during health or biology classes. Comparisons were made for demographic make-up of the received questionnaires versus

J. A. ROBERTS & J. F. TANNER, JR

766

TABLE 1

SAMPLE'SCHARACTERISTICS (1V = 1 11) Aae, vr. 12-13 14-15 16-17 18-19

f

YO

84 21 4 2

75.7 18.9 3.6 1.8

Race Asian Black Hispanic Euro-American American Indian Other

Sex Male Female

f

% 42.3

47 64

57.7

84 1 12 4 2 2

77.8 3.7 11.1 3.7 1.9 1.9

Year in School 10 4 11 90 1 3

.9 3.6 10.0 81.8 .9 2.7

7 8 9 10 11 12

the total school population, and no significant differences were found. While the lack of significant dfferences does indicate some representativeness of the sample, not all students in any grade or any school were tested and some bias could occur simply as a function of the classes in which the surveys were conducted. Compulsive Buying Measure Compulsive buying behavior has been defined as a chronic, abnormal form of shopping and spending characterized by (in extreme forms) an overpowering, uncontrollable, and repetitive urge to buy, with disregard for the consequences (Edwards, 1993). The key elements of compulsive buying include the compulsion to buy and the negative effect such buying has on the affected individual. An 11-item Compulsive Buying scale was developed for the present study. Some items were borrowed and adapted from compulsive buying scales developed by Faber and O'Guinn (1992), Edwards (1993), and Valence, el a/. (1988). Other items were developed specifically for use with adolescents in mind, e.g., asking parents or others for money. AU responses were recorded on 5-point Llkert scales anchored by strongly agree (1) and strongly disagree ( 5 ) . A lower score on the compulsive buying scale represents higher compulsive buying. Ratings of the final five items were summed for a score. Extant scales were designed for adults and were deemed not appropriate for use with teenagers. For example, the most often used compulsive buying scale, developed by Faber and O'Guinn (19921, includes questions about writing checks, pay day, and credit cards. The 11 items of the compulsive buying scale exhibited adequate rehabdity (a=.76); however, a principal components factor analysis with varimax rotation suggested that several factors existed within the scale. Dropping the items which cross-loaded at .4 or above on more than one factor

7 67

ADOLESCENTS' COMPULSIVE BUYING. RISKS

resulted in a unidimensional five-item scale (see Table 2 for final scale items). The remaining five items all had factor loadings of .5 or above on the single factor which accounted for 45% of the variance. The mean for the 5-item measure was 18.1 with a standard deviation of 4.2. ITEMS

OF

TABLE 2 COMPULSNE BUYING SCALE*

Irem

M

SD

1. I feel I have to spend any money I have.' 2 . I am always asking my parents or other people €or money. 3 . I always buy something when I go shopping. 4. I buy things even though I cannot afford them.b 5 . When I want something but have no money, I will try to borrow it.

3.5 3.5 2.9 4.3

1.3 1.3

3.8

1.3

1.1 1.2

'Items retained for final scale. AU responses were recorded on >-point Llkert scales anchored by stron y agree (1) and strongly disagree (5). A higher score represents lower compulsive buying. k t i n g s of rhe iinal five items were summed LO create the score for this $ompulsiue Buying Scale. 'Adapted from the Valence, et al. (1988) Compulsive Buying Scale. From the Edwards (1993) Compulsive Buying Scale.

Other Measures The other focal variables for the present study included smgle-~tem measures of whether the respondent had ever (1) smoked cigarettes, (2) drunk alcohol, or (3) smoked or used an illegal drug. Responses to these questions were recorded as either yes ( I ) or no (2). A fourth question measured the respondent's behavioral intentions as they relate to sexual activity. The question asked "How do you plan to deal with sex?" Responses ranged from "have sex at the first opportunity" (I), "wait until I am in love" (21, "wait unul I am 18 or older" (31, to "wait until I am married" (4). Respondents were also asked to record their age, sex, race, and grade in school. AND DISCUSSION RESULTS

A correlational analysis was performed to assess the relationships of compulsive buying with several at-risk behaviors in adolescents. Caution is advised when interpreting the present findings. The sample size and number of variables studied suggests potential risk of type one error. Ratings on the five-item measure of compulsive buying were correlated with the measures of sexual activity, smoking, alcohol, and drug use. Demographic measures were also included in the analysis. As can be seen in Table 3, self-reported use of cigarettes, alcohol, and dlegal drugs were significantly associated with ratings of compulsive buying. As respondents' scores on the compulsive buying scale increased, so did the l~kehhoodthat they would report having smoked, drunk alcohol, and used drugs and were more &ely to participate in premarital sex, but no significant correlations were found between compulsive buying and the demographic measures of age, sex, race, and year in school.

J. A. ROBERTS & J. F. TANNER, JR.

Variable 1. 2. 3. 4. 5.

Compulsive Buying Smoking Alcohol Use Drug Use Sexual Intent 6. Age 7. Sex 8. Race 9. Year in School "p= .05. tp=.Ol.

1

2

309: .3 1t .32t 3O.t .02 -.01 .04 .09

,691 .23" .55f -.I8 -.I4 -.01 -.24"

3

4

5

.10 .45t .17 -.44t -.23" -.18 -.I5 .14 .17 -.02 .40t -.03 -.50t -.I1 -23

6

7

8

.01 -.lo .85t

.05 .12

.09

9

The relationships found between compulsive buying and several risky behaviors are of particular concern given earlier estimates that suggest compulsive buying is quite common (McElroy, et al., 1994; Schlosser, et al., 1994; Black, et a/., 1998; Roberts, 1998). Young people who display compulsive buying tendencies are more likely to smoke, drink alcohol, have used drugs, and be more i n c h e d to engage in premarital sex. If compulsive buying belongs to the family of impulse control disorders, the primary motivation for such behavior is considered alleviation of stress and anxiety through changes in arousal or enhanced self-esteem. The literature suggests a strong h k between self-esteem, stress reduction, arousal, and compulsive buying (O'Guinn & Faber, 1989). A close inspection of Tables 2 and 3 provides additional insight. The means on many of the compulsive buying items are quite high (greater than 4 on a 5-point scale). A higher score indicates a lower Ilkelhood of compulsive buying. Despite the above skewed distribution across the compulsive buying scale items, nearly 29% of the sample gave ratings of 3 or less. Because the present study used a new scale to measure compulsive buying, no norms have been developed to classify which of the study's respondents could be considered compulsive buyers; however, compulsive buying is best viewed as a continuum rather than as a dscrete characteristic (e.g., Edwards, 1993). Consumers can be classified along this continuum from normal (noncompulsive), recreational, borderline, compulsive, and addicted spenders in order of increasing compulsiveness in buying behavior (Edwards, 1993). The correlations between compulsive buying and the risky behaviors found in Table 3 suggest that these behaviors may share an underlying dimension. However, it is interesting to note that many of the correlations between the risky behaviors are higher than the correlations between the risky behaviors and compulsive buying (.3O to 3 2 ) . -

-

ADOLESCENTS' COMPULSIVE BUYING, RISKS

Directions for Research Planned research should take one of three drections. One may better characterize compulsive buying, define its boundaries, and create more valid measures, particularly for use among adolescents. The present measure of compulsive buying by adolescents should be developed further to become age-appropriate, talung into account the dkferences in younger teens versus older ones, who work, drive, and so forth. Yet, more conceptual development is needed first; this may require developmental studies of impulse control disorders and of coping skdls. Researchers might also focus on the elucidation of the antecedents of compulsive buying in relation to impulse control disorders, treating anxiety and stress as situational triggers of compulsive buying. Research could also focus on the prevention and treatment of compulsive buying. The relation of various coping slulls with the incidence of compulsive buying may clardy how to reduce such behavior among adolescents. REFERENCES BUCK, D. W., REPERTINCER, S., GAFFNN,G. R., &GABEL, J . (1998) Family history and psychiatric comorbidity in persons with compulsive buying: preliminary findings. A~nericanlournal of Pqchiaty, 155, 960-963. CHRISTENSON, G. A., FABER, R. 1.. DEZWAAN, M.. RAYMOND. N. C., SPECKER, S. M.. EKERN,M. D.. ~ ~ C K E N T. Z IB., E ,CROSBY, R. D., CROW,S. J., ECKERT,E. D., MUSSELL, M. P., &MITCHELL. J. E. (1994) Compulsive buying: descriptive characteristics and psychiatric comorbidity. Jozrnrai of Clinical Psychiatry, 55, 5-11. CHRISTENSON, G. A., FABER,R. 1.. OEZWMN.IM., RAYMOND. N. C., SPECUR.S . M.. &MITCHELL, 1. E. (1992) A descriptive study of compulsive shoppers. Paper presented at the American Psychiatric Association Conference, Washington, DC. o ' h o u s . A., MALTAIS. J., & ROBERCE. C. (1990) Compulsive buying tendencies of adolescent consumers. In M. E. Goldberg, G. Gorn, & R. W. Pollay (Eds.), Ahatrces in consumer research. Vol. 17. Provo, UT: Association for Consumer Research. Pp. 306-312. DESARBO,W. S.. & EDWARDS. E. (1996) T pologies of compulsive buyiig behavior: a consrrained clusterwise regression approacl. Jozrrnal of Corlrzrmer Prychology, 5 , 231-262. EDWAFUJS,E. A. (1993) Development of a new scale for measuring compulsive buying behavior. Financial Counseiirrg and Pla~rrritrg,4, 67-84. FAEIER, R. J. (1992) Money changes everything. Arnerico~rBehauioralScien/ir~,35, 809-819. FAEIER, R. 1.. & O'GUINN,T. C. (1992) A dinical screener for compulsive buying. Journal of Corrsrrmer Re.rearch, 19, 459-469. FABER, R. J.. O'GUINN. T. C.. & KRYCH.R. (1987) Compulsive consumption. In M. Wallendorf & P. Anderson (Eds.), Advances in corrsrrmer research. Vol. 14. Provo, UT: Association for Consumer Research. Pp. 132-135. FRIESE,S., &KOENIC, H. (1993) Shopping for rrouble. Advancing the Co~rsrimerIlrterest, 5, 2429. HIRSCHMAN, E. C. (1992) The consciousness of addiction: toward a general theory of compulsive consumption, Jorrrnal of Corrszrmer Research, 19, 155-179. MCELROY, S. L.. KLECK, P. E.. JR., POPE,H. G.. JR., SMITH. ]. M. R., &STRAKOWSKI, S. IW. (1994) Compulsive buying: a report of 20 cases. Jorrrrral of Clinical Psychralry, 55, 242-248. MCELROY, S. L., POPE,H. G., JR., HUDSON, J. I., KLECK,I? E., JR.,&WHITE,K. L. (1991) U e p comania: a report of nventy cases. Americari Jozirnal of Pychiatry, 148, 652-657. O'GUINN, T. C.. & FABER.R. J. (1989) Compulsive buying: a phenomenological exploration. Journal of Corrsrimer Research, 16, 147-157.

770

J. A. ROBERTS & J. F. TANNER, JR.

PETER.J. P. (1991) Philosophical tensions in consumer inquiry. In T. S. Roberrson & H. H. Kassarjian (Eds.), Handbook of corzsumer behauior. Englewood C M s , N J : Prendce-Hall. P p . 533-548. ROBERTS J A 1 9 9 8 ) Compulsive bu ing among coUe&e srudents: an investi ation of its antecedents, consequences, and impications for pub c pol~cy The ]orrrrza$of Corznitner /Iffairs, 32, 295-319. SCHLOSSER. 5.. BLACK. D. W., REPERTINGER, S., &FEET. D. (1994) Compulsive bu?:, demography, phenomenology, and comorbidiry in 46 subjects. Gerzerol H o s p ~ ~ aPsvc l tat?, 16. 205-212. VALENCE, G., D'ASTOUS,A.. & FORTIER,L. (1988) Compulsive buying: concepr and measurement. Jot~rnalof Conszimer Policy, 11, 419-433.

Accepted March 29, 2000