Journal of Economic Psychology 32 (2011) 79–92

Contents lists available at ScienceDirect

Journal of Economic Psychology

journal homepage: www.elsevier.com/locate/joep

Shopping without pain: Compulsive buying and the effects of credit

card availability in Europe and the Far East
Hui-Yi Lo a,b,⇑, Nigel Harvey b
a
College of Management, Yuan Ze University, Taiwan
b
Division of Psychology and Language Sciences, University College London, UK

a r t i c l e i n f o a b s t r a c t

Article history: The financial consequences of compulsive buying are obvious given the large amount of
Received 7 April 2008 debt reported by compulsive buyers in many studies. Credit cards allow consumers to bor-
Received in revised form 14 November 2010 row money very easily in order to satisfy their desire to purchase. In two web-based exper-
Accepted 3 December 2010
iments, we found that compulsive shoppers often overspent and were rarely influenced by
Available online 15 December 2010
price. Their overspending was partially mediated by their excessive use of credit cards.
Furthermore, compulsive shoppers were less conscious of their budgets, especially when
JEL classification:
they used credit cards. They also obtained more pleasure from accomplishing a shopping
D14
trip and were more distressed by delayed product delivery than normal shoppers. Finally,
PsycINFO classification: compulsive shoppers in Taiwan were more compulsive than those in the United Kingdom:
3920 they displayed many of the above symptoms of compulsive buying more saliently.
3233 Ó 2010 Elsevier B.V. All rights reserved.

Keywords:
Compulsive buying
Credit card usage

1. Introduction

Researchers vary in how they define compulsive buying behavior. Kraepelin (1915) originally noted ‘‘oniomania’’ in the
psychiatric literature and Bleuler (1924) also referred to oniomania or buying mania. Essentially the symptoms of compul-
sive buying are uncontrollable urges that repeatedly compel a person to buy, bringing temporary relief from psychological
needs (e.g. tension and low self-esteem). After buying, an individual feels guilty about succumbing to the urges and suffers
financial harm (Faber & O’Guinn, 1989; Garcia, 2007; Manolis & Roberts, 2008; McElroy, Keck, Pope, & Hudson, 1994;
McElroy, Satlin, Pope, Keck, & Hudson, 1991). With rare exceptions (Koran, Faber, Aboujaoude, Large, & Serpe, 2006), com-
pulsive buying behavior is largely restricted to middle-income and low-income individuals (Faber & O’Guinn, 1992; Faber,
O’Guinn, & Krych, 1987), most of whom are women (O’Guinn & Faber, 1989; Schlosser, Black, Repertinger, & Freet, 1994).
Compulsive shoppers are more likely to possess low levels of self-esteem (Coopersmith, 1990; D’Astous, 1990; Lee,
Lennon, & Rudd, 2000; O’Guinn & Faber, 1989; Scherhorn, Reisch, & Raab, 1990; Yurchisn & Johnson, 2004) and highly mate-
rialistic attitudes (Desarbo & Edwards, 1996; Mowen, 2000; O’Guinn & Faber, 1989; Roberts, 2000).
Typically, they suffer from high levels of depression, and buy things in their attempts to alleviate their negative mood
(Black, 2007; Desarbo & Edwards, 1996; Faber, O’Guinn, & Krych, 1987). Negative feelings or outcomes can be guilt or regret

⇑ Corresponding author. Address: College of Management, Yuan Ze University, Room 60910, 135 Yuan-Tung Road, Chung-Li city 32003 Taiwan. Tel.: +886
3 463 8800x2093.
E-mail addresses: huiyi.lo@saturn.yzu.edu.tw (H.-Y. Lo), n.harvey@ucl.ac.uk (N. Harvey).

0167-4870/$ - see front matter Ó 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.joep.2010.12.002
80 H.-Y. Lo, N. Harvey / Journal of Economic Psychology 32 (2011) 79–92

that purchasing has led to financial difficulties or they can arise from the harmful impact that their behavior has on their
interpersonal relationships. The financial consequences of compulsive buying are clear from the large debts reported by
compulsive buyers in many studies (Christenson, Faber, de Zwaan, & Raymond, 1994; O’Guinn & Faber, 1989).
Debt is an increasing social problem in most developed countries at present. For example, the total insolvencies in
England and Wales grew from 25.6 thousand in 1991 to 46.6 thousand in 2004 (Department of Trade & Industry, 2006).
The dramatic growth in credit card usage among young adults and adolescents is likely to have contributed to this problem.
The increase in the number of young adults and adolescents holding credit cards and incurring credit card debts has gener-
ated concern that their finances are overextended but that they remain unaware of the long-term consequences of severe
indebtedness (Davies & Lea, 1995; Roberts & Jones, 2001; Roberts & Tanner, 2000).
Credit cards allow consumers to borrow money very easily in order to satisfy their purchasing desires. Those who pay by
credit card are more likely to make additional purchases and increase the magnitude of their spending (Feinberg, 1986;
Hirschman, 1979; Soman, 2001). Consequently, they can rapidly go into debt. Thus the easy availability of credit is likely
to be one of the main causes of excessive spending. For example, in Taiwan, some consumers overuse credit cards in order
to satisfy their urges to spend, and their abnormal consumption leads to severe financial hardship. These people are known
as ‘‘Card Slaves’’. According to data from the Taiwanese Bank Association (2006), there are an increasing number of card
slaves pleading for help. In 2006 there were 2,622,412 credit card holders with severe debt problems seeking assistance from
the Association (Taiwanese Bank Association, 2006).
The convenience of credit may lure compulsive shoppers to live beyond their means because credit cards are useful to
those people who are less able to pay but eager to acquire more material possessions (Hirschman, 1979) and because the
money involved in credit card transactions appears unreal and abstract. Behavioral studies indicate that consumers spend
more (Burman, 1974) or spend more quickly (Feinberg, 1986) when they carry credit cards. They also display poor knowl-
edge of the implications of credit card borrowing (Wiener et al., 2007). Compulsive shoppers may exhibit these problems
more severely than others because they typically have lower self-control; as a result, they are likely to spend more money
(Baumeister, 2002).
O’Guinn and Faber (1989) found that compulsive buyers have more bank credit cards than other people (mean compul-
sive buyers: 3.7; mean general consumers: 2.2) and pay back fewer of them in full each month (mean compulsive buyers:
1.0; mean general consumers: 2.0). Furthermore, they keep more credit cards within $100 of their credit limit than general
consumers do (mean compulsive buyers: 1.8; mean general consumers: 0.4). McElroy et al.’s (1994) study of psychiatric pa-
tients with problematic buying behavior showed that credit cards triggered or increased their compulsive buying. Other
studies have also suggested that compulsive shoppers tend to misuse credit facilities (Magee, 1994; Roberts, 1998; Roberts
& Jones, 2001).
Here, we report two web-based experimental studies that were designed to examine: (1) whether payments by credit
card lead compulsive shoppers to overspend more dramatically and to spend more in total than the rest of population
and (2) whether credit cards affect shoppers’ budget-consciousness and their mood reactions to events related to their pur-
chasing behavior.
We also studied whether differences between normal and compulsive shoppers vary between Taiwan and the United
Kingdom. There are two reasons why they might do so. The first is cultural. We know that compulsive shoppers are partic-
ularly concerned with purchasing products that are fashionable or symbolic of high status (O’Guinn & Faber, 1989) and that
Confucian societies place more emphasis on the importance of displaying ownership of such products than Western societies
do (Wong & Ahuvia, 1998). Hence, we expect Taiwanese compulsive shoppers to be more compulsive than their British coun-
terparts. In other words, effects associated with compulsive shopping should be larger in the Taiwanese sample. Secondly,
cross-national differences may arise because current ways of paying for products differ between the two countries (NCCC,
2008; Visa Europe, 2006): certain payment mechanisms may facilitate compulsive shopping more than others.

2. Identifying compulsive buying

2.1. Compulsive buying scales

To explore what is special about compulsive buyers, they have to be identified. Research on the diagnosis of compulsive
buying has been reported in many studies (e.g. Faber & O’Guinn, 1989; Manolis & Roberts, 2008; Manolis, Roberts, &
Kashyap, 2008; McElroy et al., 1991; O’Guinn & Faber, 1989; Valence, D’Astous, & Fortier, 1988). This work has resulted
in two scales designed to assess degree of compulsive buying behavior, one developed by Faber and O’Guinn (1992) and
the other by McElroy et al. (1991). Faber and O’Guinn’s (1992) compulsive buying scale (CBS) was used here because it
was expressly designed as a screening device to distinguish individuals who tend toward compulsive purchasing from
the normal population.
Faber and O’Guinn (1992) used logistic regression to select those scale items (from a pool of 29 candidate items) that were
best able to distinguish a group of compulsive shoppers from a general population comparison group. The resulting model com-
prised seven items and used a 5-point Likert scale to assess frequency or degree of agreement with each one. Two of the seven
items pertain to emotional reactions to shopping (e.g., make oneself feel better, feeling anxious when not buying). The remaining
five relate to financial aspects of buying (e.g., the feeling of having to spend money that is left at the end of a pay period).
 H.-Y. Lo, N. Harvey / Journal of Economic Psychology 32 (2011) 79–92 81

Overall CBS scores are derived from a scoring equation. Individual scores on each of the seven questions range from one
(strongly associated with compulsive shopping) to five (strongly associated with an absence of compulsive shopping). Scores
on each question receive their own weight. Weighted scores are then combined and then added to a negative constant
( 9.69). Thus more negative scores are more indicative of a tendency to shop compulsively. The development of the scale
used logistic regression because Faber and O’Guinn’s (1992) aim was to produce a clinical screening device: their analyses
showed that a cut-off point of 1.34 to was best able to determine whether an individual has a clinically significant tendency
to compulsive buying behavior.
To test the scale’s validity as a screening device, Faber and O’Guinn (1992) measured 12 variables associated with com-
pulsive buying behavior (e.g. self-esteem, materialism, envy, credit card ownership, etc.) and compared their mean scores
across three groups: compulsive buyers identified by the screening device; self-identified compulsive buyers; general con-
sumers. This showed that there were no significant differences between the compulsive shopper group identified by the
screening device (N = 22) and the self-identified clinical group (N = 22) on any of the 12 variables but that such differences
did exist between these groups and the comparison group. Later work confirmed that use of the scale with the recommended
cut-off is associated with high levels of reliability and validity (Faber, Christenson, Zwaan, & Mitchell, 1995; Magee, 1994;
Manolis & Roberts, 2008; Roberts, 1998; Roberts & Jones, 2001).

2.2. Compulsive buying behavior: Additional external validation of the CBS

The research carried out to assess the validity of the CBS that we outlined above required participants to evaluate them-
selves mainly on psychological variables (e.g. self-esteem, materialism). Arguably, there are problems with this approach. For
example, compulsive shoppers may assess themselves as more materialistic than other people but exhibit no greater mate-
rialism when shopping.
A more behavioral approach to assessing the scale’s external validity would increase confidence that it measures what it
purports to measure. To achieve this, types of buying behavior (i.e. buying style) that the theoretical work cited above sug-
gests should be more frequent in compulsive buyers than in the general population should be identified. Then a study should
be carried out to determine whether they are indeed more frequent in those identified as compulsive buyers by the CBS.
A study of just this type was reported by Lejoyeux, Mathieu, Embouazza, Huet, and Lequen (2007). They carried out an
investigation at the entrance of a department store. Analysis of the buying style of their 200 female participants indicated
differences between compulsive shoppers and normal shoppers were significant and in the expected direction. For example,
48% of compulsive shoppers’ decisions to buy were most often made while visiting shops whereas only 24% of normal shop-
pers’ decisions to buy were made then. Also, 23.4% of compulsive shoppers used the item they had bought less than expected
whereas this was true for only 14.4% of the normal shoppers. Furthermore compulsive shoppers were significantly more
likely to make purchases because they want to impress others, because those purchases were considered personally grati-
fying, and because they were influenced by the brand of the product.
Unfortunately, Lejoyeux et al.’s (2007) study used McElroy et al.’s (1991) compulsive buying scale rather than Faber and
O’Guinn’s (1992) CBS. To obtain the same type of external validation for the CBS, we used it in a study that otherwise closely
replicated Lejoyeux et al.’s (2007) investigation. Details are as follows.

2.2.1. Procedure
To ensure comparability with Lejoyeux et al. (2007), all participants were female (N = 202). They were recruited at the
door of a branch of the Taiwanese SOGO department store between 11:30 am and 8:00 pm. To ensure confidentially, we re-
corded respondents’ ages but no other data that could be used to identify them. They each completed the CBS and then re-
sponded to the buying behavior questions that Lejoyeux et al.’s (2007) used in their study. We also included one additional

Table 1
Results of external validation study of Faber and O’Guinn’s (1992) compulsive buying scale.

Buying style CS NCS Difference statistically Consistent with

(n = 21) (n = 181) significant? Lejoyeux et al. (2007)?
Necessary purchases vs. self-gifts or gifts to others (%) 61.9 (gifts) 22.7 (gifts) Fisher exact test, p < .001 No
Decision to buy an object most often taken before entering in 23.8 58 Fisher exact test, p = .005 Yes
the shop (%)
Shopping more often alone than in company of others (%) 23.8 35.9 Fisher exact test, p > .05 Yes
Purchases made during sales (%) 52.4 56.9 Fisher exact test, p > .05 Yes
Purchases considered as occasions not to be passed up (%) 71.4 69.6 Fisher exact test, p > .05 No
Purchases used less than expected (%) 66.7 27.1 Fisher exact test, p = .001 Yes
Feelings of being deceived after purchase (%) 42.9 8.3 Fisher exact test, p < .001 Yes
Purchases made to impress others (mean) 4.24 2.51 t (N = 202) = 4.76, p < .001 Yes
Purchases most often considered as personally gratifying (%) 90.5 64.6 Fisher exact test, p = .015 Yes
Purchases linked to social status (%) 61.9 21 Fisher exact test, p < .001 Yes
Importance of the brand on the decision to buy an item (%) 81 32.6 Fisher exact test, p < .001 Yes
No. of times in the last 6 months were returned (mean) 2.57 .07 t (N = 202) = 12.08, p < .001 No
No. of items bought in this shopping trip (mean) 5.24 2.22 t (N = 202) = 7.41, p < .001 Additional question
82 H.-Y. Lo, N. Harvey / Journal of Economic Psychology 32 (2011) 79–92

question that previous work suggests should differentiate between compulsive buyers and the general population: How
many items have you bought on this shopping trip?

2.2.2. Results
Results are shown in Table 1. It can be seen that 10 of the 13 buying behaviors that have been identified as characteristic
of compulsive buying were different in those identified by the CBS as compulsive buyers and those who were identified as
non-compulsive buyers. Furthermore, results were highly consistent with those obtained by Lejoyeux et al. (2007) using
McElroy et al.’s (1991) scale. These findings provide further evidence that the CBS scale is a valid screening device that dis-
tinguishes the different buying styles of compulsive and normal buyers.

3. Experiment 1

Both our studies employed a web-based simulated shopping environment to observe the behavioral processes underlying
compulsive buying. Although implementation of psychological investigations via the Internet is now widespread and accepted,
only a few studies into consumer behavior have so far used this approach. It provides opportunities for obtaining large, demo-
graphically varied and well-specified samples at reasonable cost. For detailed discussions of web-based testing in psychology,
see Birnbaum (2000, 2004), Reips (2002), Skitka and Sargis (2006), and Reimers (2007). Wiener et al. (2007) and Maimaran and
Wheeler (2008) provide examples of web-based experiments specifically designed to examine consumer behavior.

3.1. Participants

A sample of respondents was collected over a period of 58 days via the Internet. The web-link for the experiment was
placed on several Taiwanese and British discussion forums and web-based experiment websites (college subject pools,
the Google discussion forum, the Yahoo knowledge forum, community overview-ebay, the campus discussion forum, online
psychology research UK and psychological research on the net). Participants were recruited from these websites; data were
gathered from individuals who visited these forums and were willing to complete the experiment. Respondents were of two
nationalities: Taiwanese and British.
There were 464 British respondents. They comprised 328 valid respondents and 136 invalid respondents. The latter failed
to complete the experiment, had completed it previously, or were not British participants. The valid response rate was thus
71%. There were 174 Taiwanese respondents who comprised 125 valid respondents and 49 invalid respondents. This repre-
sents a valid response rate of 72%.
The valid British respondents were 38% male and 62% female. The percentage of British compulsive shoppers was 11% (4%
Male and 7% Female). The majority of annual incomes of compulsive shoppers were in a low-income categories £9999 and
below (31 of 36).
The valid Taiwanese respondents were 42% male and 58% female. The percentage of compulsive shoppers in this group
was 17% (6% Male and 11% Female). The monthly income of most compulsive shoppers was either in a low-income category
NTD 25,001–35,000 (8 of 21) or in a middle-income category NTD 45,001–55,000 (8 of 21).
These demographic results are consistent with the literature which suggests that compulsive buying behavior tends to be re-
stricted to middle-income or low-income individuals (Faber & O’Guinn, 1992; Faber, O’Guinn, & Krych, 1987). With rare excep-
tions (Koran et al., 2006), women constitute the majority of compulsive shoppers (O’Guinn & Faber, 1989; Schlosser et al., 1994).

3.2. Experimental design

The design of this experiment included two types of choice: skip purchasing versus pay by cash and skip purchasing versus
pay by credit card. Fig. 1 shows an example of these choices in each condition. Participants saw a sequence of 20 such prod-
ucts. They comprised 13 low price items and seven high price items. After every five items, participants had to decide
whether to pay for what they had chosen or whether to give up the items. Two sets of five items could be purchased only
by credit card and two sets only by cash. For approximately half the participants (those whose number at registration was
odd), the method of payment for the two sets of five items was ordered cash, credit card, cash, credit card. For the rest (those
whose number at registration was even), the sequence was credit card, cash, credit card, cash.
After this choice task, nine questions were used to assess participants’ credit card usage. These questions were selected
from Roberts and Jones’ (2001) credit card usage scale. They produced a credit card usage score for each participant of 1
(never), 2 (rarely), 3 (sometimes), 4 (often) or 5 (very often).

3.3. Results

3.3.1. Spending behavior: the effects of credit card availability

On the one hand, credit cards allow consumers to borrow money very easily. On the other hand, the availability of credit
cards can lead them rapidly into debt. Thus, compulsive shoppers are likely to use credit cards more unwisely and to over-
spend more frequently than normal shoppers.
 H.-Y. Lo, N. Harvey / Journal of Economic Psychology 32 (2011) 79–92 83

Fig. 1. Experiment 1: product displays in each condition.

As participants used different currencies, a z-score was used to transform data (i.e. total spending and overspending
data1). Then t-tests were used to test for significance of differences. Compulsive shoppers used credit cards more often than
non-compulsive shoppers (Mcs = 2.97, Mncs = 1.87, t (453) = 10.18, p < .001), spent more in total than non-compulsive shop-
pers (Mcs = 1.19, Mncs = 0.17, t (453) = 8.56, p < .001), and overspent more severely than normal shoppers (Mcs = 1.19,
Mncs = 0.17, t (453) = 8.59, p < .001).
Mediation analysis used to clarify the effects of compulsive buying and credit card usage on overspending and total
spending. To identify the degree of mediation for overspending, we first regressed overspending onto CBS score, then re-
gressed credit card usage behavior onto CBS, and finally regressed overspending onto both credit card usage behavior and
CBS score. The degree to which the influence of CBS score on overspending is reduced when accounting for the influence
of credit card usage expresses the degree of mediation.
The results of these regression analyses are shown in Fig. 2. They reveal that the effect of credit card usage did partially
mediate the influence of the CBS score on overspending because: (a) the CBS score predicted overspending

1
Overspending = (monthly income) – (spending on monthly essential goods) – (total spending) < 0. Here, data concerning monthly income and spending on
monthly essential goods were derived from the information obtained in the registration section.
84 H.-Y. Lo, N. Harvey / Journal of Economic Psychology 32 (2011) 79–92

Fig. 2. Mediation model for overspending in Experiment 1.

(F (1, 451) = 106.79, p < .001); (b) the CBS score predicted credit card usage (F (1, 451) = 248.76, p < .001); (c) credit card
usage predicted overspending (F (1, 451) = 94.22; p < .001); and (d) the influence of the CBS score on overspending dropped
significantly when controlling for the effect of credit card usage (F (1, 450) = 21.69, p < .001).
Because total spending is given by a constant minus overspending, the mediation analysis for total spending produces a
mediation model with the same coefficients as the one shown in Fig. 2 for overspending. Thus, we can say that usage of credit
cards encourages compulsive shoppers to overspend and thereby helps to lure them into spending more in total.
Analyses of variance were used to examine the effects of compulsive shopping behavior and nationality on the dependent
variables. Overspending and total spending were influenced only by the compulsive versus non-compulsive shopping
variable. However, as Fig. 3 shows, the frequency of credit card use was influenced by compulsive buying behavior
(F (1, 449) = 168.56, p < .001), nationality (F (1, 449) = 11.99, p < .001) and the interaction between these variables
(F (1, 449) = 8.60, p < .01). Scheffé’s test indicated that Taiwanese compulsive shoppers (M = 3.33) used credit cards more
excessively than British compulsive shoppers (M = 2.76), but there was no statistically significant difference between
Taiwanese normal shoppers (M = 1.91) and British normal shoppers (M = 1.86), F (3, 449) = 59.42, p < .001 (Fig. 3).

3.3.2. Deciding whether to purchase and chosen mode of payment

As mentioned in design section, participants had to decide whether to pay for what they had chosen after each set of five
items. For two sets of items, deciding to buy involved payment by credit card and, for two sets, it involved payment by cash.
We expected that people who tended not to buy would score higher on the CBS (i.e. be less compulsive in their purchasing)
than those who tended to buy. We also expected that people who tended to purchase by credit card would score lower on
the CBS (i.e. be more compulsive in the purchasing) than people who tended to purchase by cash. Also, if we are correct in
arguing that Taiwanese compulsive buyers are more compulsive than British compulsive buyers, the differences are likely to
be larger in the former group.
To test these hypotheses, we divided our participants into four groups. Participants who bought more items by credit card
than the average number bought by credit card across all participants and who bought more items by cash than the average
number bought by cash across all participants, we labeled purchasers. Participants who bought fewer items by credit card
than the average number bought by credit card and who bought fewer items by cash than the average number bought by
cash, we labeled non-purchasers. Participants who bought more items by credit card than the average number of items
bought by credit card and who bought fewer items by cash than the average number bought by cash, we labeled credit-card

Fig. 3. Experiment 1: effects of compulsive buying and nationality on credit card usage.
 H.-Y. Lo, N. Harvey / Journal of Economic Psychology 32 (2011) 79–92 85

Fig. 4. Upper panel: the influence of payment mode on compulsive buying behavior in Experiment 1 (Purchasers vs. non-purchasers). Lower panel: the
influence of payment mode on compulsive buying behavior in Experiment 1 (credit-card purchasers vs. cash purchasers).

purchasers. Finally, participants who bought fewer items by credit card than the average number bought by credit card and
bought more items by cash than the average number bought by cash, we labeled cash purchasers.
Our first analysis compares the marginal means of the CBS score for purchasers and non-purchasers (Fig. 4, upper panel).
Positive scores on the ordinates of these graphs show CBS scores that are higher (less compulsive) than average and negative
scores show CBS scores that are lower (more compulsive) than average.
Non-purchasers were less compulsive than purchasers (F (1, 449) = 71.15, p < .001). Furthermore, an effect of nationality
(F (1, 449) = 9.89, p < .01) and an interaction between nationality and purchasers/non-purchasers (F (1, 449) = 10.41,
p = .001) arose because Taiwanese shoppers who tended to purchase were more compulsive than UK shoppers who tended
to purchase (Fig. 4, upper panel).
Our second analysis compares marginal CBS scores for credit-card purchasers and cash purchasers (Fig. 4, lower panel).
Cash purchasers were less compulsive than credit-card purchasers (F (1, 120) = 18.81, p < .001). Also, an effect of nationality
(F (1, 120) = 5.30, p < .05) and an interaction between nationality and preferred mode of payment (F (1, 120) = 12.38,
p = .001) arose because Taiwanese shoppers who tended to pay by credit card were more compulsive than UK shoppers
who tended to pay by credit card (Fig. 4, lower panel).

3.4. Summary

In comparison with the rest of population, compulsive shoppers used credit cards more frequently. They also overspent
more and had higher total spending than normal shoppers, a phenomenon that was partially mediated by their higher credit
card usage. In addition, those who selected more items during their shopping trip and those who paid for these items by
credit card rather than cash tended to be more compulsive. A number of these effects were more evident in Taiwanese than
in British participants.
86 H.-Y. Lo, N. Harvey / Journal of Economic Psychology 32 (2011) 79–92

4. Experiment 2

Experiment 1 showed that compulsive shoppers use credit cards more frequently than other people. Thus, compulsive
shoppers may use credit cards excessively because they pay more attention to satisfying their cravings than to the possibility
of future debt. If this is so, their moods may be more easily affected by factors that disrupt their shopping processes. With
this consideration in mind, Experiment 2 was designed to examine shoppers’ budget-consciousness and the way that their
moods reacted to interventions affecting their use of credit cards.
Experiment 1 also showed that participants were more prone to overspend than normal shoppers. In Experiment 2, we
examined whether the mode of payment and the price of products affects shoppers’ intentions to buy and whether there is
any difference between compulsive shoppers and normal shoppers in this respect. We also investigated the possibility that
compulsive shoppers have a greater preference for luxury over non-luxury goods than non-compulsive shoppers.

4.1. Participants

A sample of respondents was gathered over a period of one month, and was again drawn from two nationalities (i.e. Brit-
ish and Taiwanese). As before, the web-link for the experiment was placed at several Taiwanese websites (Yahoo knowledge
forum, pay-easy online shopping and the Nanya Institute of Technology) and British websites (UCL and its psychology par-
ticipant pools, the Google discussion forum, community overview-ebay, and the University of London campus discussion for-
um). Data were gathered from individuals who visited these forums and were willing to do the experiment.
In this study, participants were informed that, by registering and participating in the entire experiment, they would be
entered in a prize-draw lottery and that first prize winner would win £50 cash (NT$ 3000) and five runners-up would
win £20 (NT$ 1000) cash. The winners were notified by e-mail.
Total number of British respondents (707) included 521 valid respondents and 186 invalid respondents, thereby giving a
valid response rate of 73.69%. Total number of Taiwanese participants (250) consisted of 200 valid respondents and 50 in-
valid respondents, thereby providing a valid response rate of 80%.
The valid British respondents were 38% male and 62% female and the percentage of them who were compulsive shoppers
was 12% (4% Male and 8% Female). Most annual incomes of British compulsive shoppers were in categories £9999 and below
(56 of 64). The valid Taiwanese respondents were 39% male and 60% female and the percentage of them who were compul-
sive shoppers was 16% (6% Male and 10% Female). The monthly income of most Taiwanese compulsive shoppers was either
in a low-income category (below NTD 25,000) or at the middle-income level (NTD 35,001–45,000). Again, these demographic
results are consistent with those reported in the literature.

4.2. Experimental design

In this experiment, participants were told that they had £150 (or $NT 10,000) cash as well as their credit cards. They were
asked to imagine that they were looking for three products in an Internet shop: sunglasses, handbags, and printers. They had
two choices in each category: one was a high price luxury brand, and the other was a low price mundane brand. If they did not
have enough cash, they could pay by credit card, but they were penalized £20 (or $NT 1500) cash each time they did so. After
each transaction, their cash account was calculated automatically, and their remaining cash balance was shown on the screen
at the time of their next transaction. After the sequence of three choices, participants automatically entered a checkout section.
The fixed account feature of the design enabled us assess budget-consciousness by measuring whether participants over-
spent. Also, the instantaneous provision of cash balance information allowed us to examine whether participants noticed this
during each transaction. Furthermore, the £20 (or $NT 1500) cash penalty was used to throw light on whether compulsive
buyers’ cravings leads them to ignore this financial loss when satisfying those urges.
To assess how participants’ moods reacted to events that disrupted their purchasing behavior, three interventions were
used. These were ‘‘CANNOT OWN’’, ‘‘BUDGETARY SUGGESTION’’ and ‘‘DELAYED DELIVERY’’. In addition, two prompts elicited
their moods in response to successful purchasing. These were ‘‘HAPPINESS IF OWNED PRODUCTS IN REAL WORLD’’ and
‘‘HAPPINESS WITH PURCHASES’’. The conditions that led to these interventions and prompts were as follows.
Overall spending that was more than the affordable limit prompted the first intervention, ‘‘CANNOT OWN’’. This involved
telling participants that they could not own the products that they had selected because their overall spending was beyond
their affordable range. They were asked to express their mood in reaction to this on a five point scale that ranged from one
(no influence) to five (extremely disappointed). After this first intervention, the next screen provided the second one, a
‘‘BUDGETARY SUGGESTION’’. This informed them that their spending was not affordable and asked them to be careful to con-
trol their budget. They were then asked to express their mood in reaction to this suggestion, using a scale ranging from one
(very unhappy) to five (very happy).
Participants who had not overspent were treated differently. They were told that they had not overspent and given the
first mood prompt, ‘‘HAPPINESS IF OWNED PRODUCTS IN REAL WORLD’’. This involved presenting them with this message:
‘‘Congratulations, you accomplished your shopping trip. How happy would you be if you could own the products you have
purchased in the real world?’’. In response, they rated their mood on a scale from one (very unhappy) to five (very happy).
Those who had not overspent but had paid for at least one of their purchases by credit card then experienced the third type
 H.-Y. Lo, N. Harvey / Journal of Economic Psychology 32 (2011) 79–92 87

of intervention, ‘‘DELAYED DELIVERY’’. This involved telling them that they must wait for one of the products they had pur-
chased for seven days2 until their bank validated their spending and then asking them to rate their mood in reaction to this
information using a scale ranging from one (no influence) to five (extremely disappointed). However, participants who did
not overspend and did not use credit cards in any transactions received instead the second mood prompt, ‘‘HAPPINESS WITH
PURCHASES’’. This involved asking them to rate how happy they were with their purchases on a scale from one (very unhappy)
to five (very happy).

4.3. Results

4.3.1. Credit card usage

Participants could decide to pay by credit cards or by cash in the experiment but, if they used credit cards, they were
penalized £20 (or $NT 1500) cash. In the British sample, 72% of compulsive shoppers preferred to use a credit card for at least
one of their purchases even though they lost some cash but only 45% of non-compulsive shoppers elected to use one (Fisher’s
exact test, p < .001). Results from the Taiwanese sample were similar: 78% of compulsive shoppers preferred to use a credit
card for at least one of their purchases but only 53% of non-compulsive shoppers chose to use one (Fisher’s exact test,
p < .001). These findings indicate that compulsive shoppers were more willing than other shoppers to use credit cards, de-
spite the £20 (or $NT 1500) cash penalty that this incurred.

4.3.2. Budget-consciousness
For British participants, Fisher’s exact test indicated that most shoppers overspent, presumably because they did not no-
tice their instantaneous cash account information during each transaction. However, non-compulsive shoppers were less
likely to overspend than compulsive shoppers (64% versus 86%). This implies that they were more conscious of their bud-
geting constraints than compulsive shoppers. Within the Taiwanese sample, a majority of compulsive shoppers (75%) over-
spent, presumably because they were not conscious of their account information. However, most non-compulsive shoppers
(60%) did not overspend and so appear to have been more conscious of this information (Fisher’s exact test, p < .001).
To assess how their budget awareness reacted to credit card usage, a three-way ANOVA was carried out on the overspend-
ing3 data (Fig. 5). The amount of overspending was greater when participants used credit cards than when they did not
(F (1, 713) = 127.75, p < .001) and when they were compulsive shoppers than when they were not (F (1, 713) = 63.85,
p < .001). The difference between compulsive and non-compulsive shoppers was, on average, larger in the Taiwanese sample
(F (1, 713) = 14.20, p < .001). Also the effect of greater overspending when participants used cards was enhanced when they
were compulsive shoppers (F (1, 713) = 38.26, p < .001). Finally, an interaction between all three variables (Fig. 5) showed that
this enhancement was greater in Taiwanese than in British participants (F (1, 713) = 13.08, p < .001). So, again, we conclude that
non-compulsive shoppers are more conscious of their budget than compulsive shoppers, that Taiwanese compulsive shoppers
are more compulsive than British compulsive shoppers, and that credit card availability reduces shoppers’ consciousness of
their budget constraints.

4.3.3. Degree of preference for luxury products

In this experiment, participants were shown handbags, sunglasses, and printers. Each category consisted of a luxury (or a
famous) brand and a mundane (or an economy) brand. In each category, participants had to select one of the products. Com-
pulsive shoppers were expected to prefer luxury (or famous) brands to economy (or mundane) brands.
The three categories of product were split into those which were fashionable (handbags and sunglasses) and those which
were functional (printer). Within each of these groups, selection of a luxury brand scored one and selection of a mundane
brand scored 0. Thus, for fashionable products, the total luxury brand score ranged between 0 (all non-luxury) and 2 (all lux-
ury choices). Analyses of variance with post hoc comparisons were used to examine the effects of compulsive shopping
behavior and nationality on the mean score.
For fashionable products, the analysis of variance showed that compulsive shopping behavior (F (1, 717) = 62.80, p < .001)
and nationality (F (1, 717) = 41.52, p < .001) had significant effects on the choice of brand type but there was no interaction
between these two variables. This indicates that compulsive shoppers selected luxury brands more than non-compulsive
shoppers and Taiwanese shoppers chose luxury brands more than British shoppers (Fig. 6).
This last result suggests that Taiwanese compulsive shoppers were more compulsive than British compulsive shoppers.
Thus we used a t-test on the CBS scores to compare the intensity of compulsive buying between British and Taiwanese com-
pulsive shoppers. Results show that Taiwanese compulsive shoppers (M = 3.46) were more compulsive than British com-
pulsive shoppers (M = 2.54), t (96) = 3.93, p < .001. Could this difference have arisen because Faber and O’Guinn’s (1992)
scale picked up only the most compulsive Taiwanese shoppers but classified a broader range of British shoppers as compul-
sive? Assuming the shapes of the two distributions were not dramatically different, this would have resulted in a lower per-
centage of Taiwanese than British shoppers being classified as compulsive. However, a higher percentage of Taiwanese were

2
Normally, the validation of a credit card from banks is real-time when we buy a product online and we receive the product within a few days. If we are
asked to wait for a period of seven days for our spending to be validated and told that the shop will not deliver the product until then, we would be waiting for
longer than usual.
3
Overspending = total spends – budget constraint (i.e., British: £150; Taiwanese: $TW 10,000).
88 H.-Y. Lo, N. Harvey / Journal of Economic Psychology 32 (2011) 79–92

Fig. 5. Upper panel: the effect of credit card use and compulsive buying behavior on overspending, an indicator of budget awareness, in Experiment 2 (UK).
Lower panel: the effect of credit card use and compulsive buying behavior on overspending, an indicator of budget awareness, in Experiment 2 (Taiwan).

classified as compulsive (16% versus 12%). Hence, it appears that the greater compulsivity of Taiwanese compulsive shoppers
accounts for why they chose luxury brands more than British compulsive shoppers did.
For functional products, there were no significant effects of nationality or shopping style on choice of brand types.

4.3.4. Mood ratings after interventions and in response to queries

There were three potentially disruptive interventions (‘‘CANNOT OWN’’, ‘‘BUDGETARY SUGGESTION’’ and ‘‘DELAYED
DELIVERY’’) that could influence participants’ moods and two queries about participants’ moods in response to making pur-
chases (‘‘HAPPINESS IF OWNED PRODUCTS IN REAL WORLD’’ and ‘‘HAPPINESS WITH PURCHASES’’). Two-way ANOVAs were
used to examine how compulsive shopping behavior and nationality affected participants’ moods in response to each of
these cases (Table 2). These analyses show that the moods of compulsive shoppers were lower than those of non-compulsive
shoppers after the interventions but higher after successfully making purchases. These effects were larger in the Taiwanese
than in the UK sample for the two interventions that occurred after overspending had stopped participants owning products.
When participants encountered two of the interventions (‘‘CANNOT OWN’’ and ‘‘BUDGETARY SUGGESTION’’) and one of
the queries (‘‘HAPPINESS IF OWNED PRODUCTS IN REAL WORLD’’), some of them had made one or more purchases with a
credit card and some had not used a credit card at all. For these cases, we added this variable into the ANOVAs to ask how
credit card use influences the effects describe above. Thus three three-way analyses ANOVAs were carried out with nation-
ality, type of shopper, and card use as between-participant factors. (Table 3 shows the mean mood ratings for each combi-
nation of these variables.) These analyses broadly confirmed the results of the two-way ANOVAs but also revealed that
participants’ moods were influenced by whether they had used credit cards to make their purchases.
For ‘‘CANNOT OWN’’, the effect of type of shopper (F (1, 210) = 56.66, p < .001), nationality (F (1, 210) = 5.80, p < .05), and
the interaction between these two variables (F (1, 210) = 12.89, p < .001) were all significant. However, the ANOVA also
revealed that the degree to which the difference between compulsive and non-compulsive shoppers was greater in the
Taiwanese sample was less when a credit card had been used than when it had not (F (1, 210) = 4.07, p < .05). The reason
 H.-Y. Lo, N. Harvey / Journal of Economic Psychology 32 (2011) 79–92 89

Fig. 6. Experiment 2: effects of compulsive buying and nationality on preference for luxury brands.

Table 2
Experiment 2: mood ratings in response to interventions (two-way ANOVAs).

Conditions Type of Nationality N Mean SD Source

shoppers
CBS Nationality CBS  nationality
1. Can’t own products* CS British 22 3.41 1.01 F (1, 214) = 72.46, F (1, 214) = 3.83, F (1, 214) = 12.48,
p < .001 p > .05 p = .001
Taiwanese 13 4.54 0.78
NCS British 117 2.38 1.11
Taiwanese 66 2.06 1.09

2. Budgetary suggestion** CS British 22 2.73 0.70 F (1, 214) = 18.05, F (1, 214) = 2.46, F (1, 214) = 15.25,
p < .001 p > .05 p < .001
Taiwanese 13 1.92 0.86
NCS British 117 2.78 0.74
Taiwanese 66 3.12 0.81

3. Delayed delivery* CS British 33 3.88 0.86 F (1, 289) = 18.89, F (1, 289) = 2.53, F (1, 289) = 0.03,
p < .001 p > .05 p > .05
Taiwanese 15 3.60 1.24
NCS British 177 3.06 1.11
Taiwanese 68 2.72 1.29

4. Happiness if owned products in CS British 42 3.81 1.09 F (1, 499) = 17.55, F (1, 499) = 0.02, F (1, 499) = 2.89,
real world** p < .001 p > .05 p > .05
Taiwanese 19 4.05 1.13
NCS British 340 3.43 0.96
Taiwanese 102 3.15 1.20

**
5. Happy with purchases CS British 42 3.81 1.09 F (1, 206) = 19.41, F (1, 206) = 0.01, F (1, 206) = 1.04,
p < .001 p > .05 p > .05
Taiwanese 19 4.05 1.13
NCS British 340 3.43 0.96
Taiwanese 102 3.15 1.20

Note. CS: compulsive shoppers; NCS: non-compulsive shoppers; CBS: compulsive buying score.
*
5-Point scale (5: extremely disappointed; 1: no influence).
**
5-Point scale (5: very happy; 1: very unhappy).

for this was that, for non-compulsive shoppers, using a credit card made the intervention more depressing, especially in Tai-
wan. In contrast, for compulsive shoppers, using a credit card made the intervention more depressing in the British sample
but less depressing in the Taiwanese one. Hence, using a card made moods of compulsive and non-compulsive Taiwanese
shoppers more similar but those of British compulsive and non-compulsive shoppers more dissimilar.
For ‘‘BUDGETARY SUGGESTION’’, the effects of type of shopper (F (1, 210) = 8.87, p < .01) and the interaction between
nationality and type of shopper were significant (F (1, 210) = 6.68, p = .01). However, the analysis also produced a marginally
90 H.-Y. Lo, N. Harvey / Journal of Economic Psychology 32 (2011) 79–92

Table 3
Experiment 2: mood ratings in response to interventions (three-way ANOVAs).

Conditions Type of shoppers Card Nationality N Mean SD

1. Can’t own products* CS No British 9 2.89 .60
Taiwanese 3 5.00 .00
Yes British 13 3.77 1.09
Taiwanese 10 4.40 .84
NCS No British 88 2.34 1.08
Taiwanese 45 1.89 .91
Yes British 29 2.52 1.18
Taiwanese 21 2.43 1.36

2. Budgetary suggestion** CS No British 9 2.67 .50

Taiwanese 3 2.67 .58
Yes British 13 2.77 .83
Taiwanese 10 1.70 .82
NCS No British 88 2.77 .69
Taiwanese 45 3.16 .85
Yes British 29 2.79 .90
Taiwanese 21 3.05 .74

3. Happiness if owned products in real world** CS No British 9 3.78 .97

Taiwanese 4 5.00 .00
Yes British 33 3.82 1.13
Taiwanese 15 3.80 1.15
NCS No British 163 3.45 .94
Taiwanese 34 3.35 1.15
Yes British 177 3.41 .98
Taiwanese 68 3.04 1.22

Note. CS: compulsive shoppers; NCS: non-compulsive shoppers; CBS: compulsive buying score; Card: using credit card.
*
5-Point scale (5: extremely disappointed; 1: no influence).
**
5-Point scale (5: very happy; 1: very unhappy).

significant interaction between nationality and using a card (F (1, 210) = 3.29, p < .05). To investigate this further, we carried
out a two-way ANOVA with card use and nationality as factors. This showed the interaction to be significant (F
(1, 214) = 5.12, p < .05). British participants’ responses to the intervention were very similar when they used a credit card
and when they did not. However, Taiwanese participants’ ratings showed them to be happier after the intervention when
they did not use a card than when they did.
For ‘‘HAPPINESS IF OWNED PRODUCTS IN REAL WORLD’’, there were effects of type of shopper (F (1, 495) = 18.22,
p < .001) and an interaction between nationality and type of shopper (F (1, 495) = 5.09, p < .05). However, the ANOVA also
revealed a main effect of using a credit card (F (1, 495) = 4.20, p < .05) and an interaction between that variable and nation-
ality (F (1, 495) = 4.19, p < .05). These effects arose because British participants’ happiness with owning the products was not
affected by whether they had been bought with a credit card but that Taiwanese were happier when they had not used a
credit card to buy them than when they had used one.
These findings are consistent with the notion that card use influences people’s moods in response to buying products and
in response to factors that obstruct buying behavior. However, the nature of its influence depends on cultural factors: only
Taiwanese shoppers were less happy with their purchases and less happy with receiving budgetary suggestions when they
had used a credit card.

4.4. Summary

Compulsive shoppers used credit cards more. They overspent more and their overspending was greater when they used
credit cards. For fashionable products, they were more likely to select luxury brands. They were more disappointed by inter-
ventions that obstructed their buying behavior and reacted more negatively to budgetary suggestions. However, they had
higher moods after successfully purchasing products. Most of these effects were larger in Taiwanese compulsive shoppers
than in British compulsive shoppers.
This was not the only difference between nationalities. In the Taiwanese sample, using credit cards accentuated the neg-
ative effects on mood of factors that obstructed buying but depressed the positive effects on mood of successfully obtaining
products. (The one exception was for Taiwanese compulsive shoppers who were told that they had overspent and, hence,
could not obtain the products: they were somewhat less disappointed if they had used a credit card.) In contrast, there
was little evidence in the British sample that credit cards mediated the effects of these factors on mood.
 H.-Y. Lo, N. Harvey / Journal of Economic Psychology 32 (2011) 79–92 91

5. General discussion

Debt is one of the most disruptive effects of compulsive shopping. Indeed, credit cards are useful to those people, espe-
cially compulsive shoppers, who are less able to pay but are eager to acquire more material possessions. This is probably
because the money involved in credit card transactions appears unreal and abstract. Hence, credit cards allow compulsive
shoppers to borrow money to satisfy their cravings without them being faced with the immediate consequences of doing
so. They may lack self-control or they may underestimate the effects of their purchases on their financial condition. In either
case, the short-term satisfaction that they gain from their buying activities may be increasingly offset by their awareness of
their rising debt. As a result, they may seek strategies to counteract their cravings to buy.
Compulsive shoppers used credit cards more frequently (Fig. 3) and spent more money on their purchases. In fact, use of
credit cards partially mediated the tendency of compulsive shoppers to spend more than other shoppers (Fig. 2). People
using credit cards were also less conscious of their budget than people using cash and this was especially true of compulsive
shoppers (Fig. 5). Furthermore, compulsive shoppers’ moods were more easily affected by factors that disrupted their shop-
ping (Table 2). These findings extend results of earlier studies indicating that credit card purchases increase both the fre-
quency of spending and its excessiveness (Feinberg, 1986; Hirschman, 1979; Soman, 2001).
Results from Experiment 2 indicate that compulsive shoppers do not welcome being reminded of increasing financial
problems while they are shopping: they are not happy with suggestions that draw their attention to their budgetary con-
straints. They are also more disturbed than normal shoppers by problems (delays in product delivery) that prevent them
from immediately satisfying their urges.
Differences between Taiwanese and British shoppers could arise either from underlying cultural differences or from more
superficial differences associated with variations in the way people can make purchases in the two countries. We shall con-
sider these possibilities in turn.
Wong and Ahuvia (1998) point out that the consumption culture in Confucian societies is based on an interpersonal con-
strual of the self. Members of those societies tend to behave in ways that emphasize social roles and public perception. As a
result, their consumers tend to place more emphasis on public or visible possessions than Western consumers. They more
often use products symbolically to lay claim to a desirable position within the socioeconomic hierarchy (Wong & Ahuvia,
1998). According to this point of view, it is not surprising that Taiwanese compulsive shoppers chose more luxury goods than
British compulsive shoppers and used credit cards more excessively than British compulsive shoppers. These differences ar-
ose because cultural differences led to Taiwanese compulsive shoppers to be more compulsive than British compulsive
shoppers.
There are cross-national variations in the payment methods that are available to people. For example, direct debit cards
are a well-established method of payment in the UK but Smart pay, the Taiwanese equivalent of debit cards, is only now
being promoted. British people probably regard their debit cards as closely analogous to cash because money is taken from
their bank account directly and immediately. Any cross-national differences in the way cash and debit cards are used by
compulsive shoppers in Taiwan and the UK could arise from this difference in availability and familiarity with the debit card
system.
In contrast, credit cards are widely used in both countries. Cross-national differences in their use cannot be attributed to
differences in how familiar people are with them as a payment method. Compulsive shoppers in both countries differ from
non-compulsive shoppers by their clear preference for payment by credit card over other modes of payment. Thus the
greater use of credit cards by Taiwanese than by British compulsive shoppers is best explained by cultural factors (Wong
& Ahuvia, 1998) that induce them to be more compulsive than their UK counterparts.

References

Baumeister, R. F. (2002). Yielding to temptation: Self-control failure, impulsive purchasing, and consumer behavior. Journal of Consumer Research, 28,
670–676.
Birnbaum, M. H. (2000). Psychological experiments on the internet. San Diego, CA, US: Academic Press.
Birnbaum, M. H. (2004). Human research and data collection via the internet. Annual Review of Psychology, 55, 803–832.
Black, D. W. (2007). Compulsive buying disorder: A review of the evidence. CNS Spectator, 12, 124–132.
Bleuler, E. (1924). Textbook of psychiatry. New York: MacMillan.
Burman, D. (1974). Do people overspend with credit cards? Journal of Consumer Credit Management, 5, 98–103.
Christenson, G. A., Faber, R. J., de Zwaan, M., & Raymond, N. C. (1994). Compulsive buying: Descriptive characteristics and psychiatric comorbidity. Journal of
Clinical Psychiatry, 55, 5–11.
Coopersmith, S. (1990). Self-esteem inventories. Palo Alto, CA: Consulting Psychologists Press.
D’Astous, A. (1990). An inquiry into the compulsive side of normal consumers. Journal of Consumer Policy, 13, 15–31.
Davies, E., & Lea, S. E. G. (1995). Student attitudes to student debt. Journal of Economic Psychology, 16, 663–679.
Department of Trade and Industry (2006). Number of individual insolvencies, England & Wales. UK: National Statistics. Social Trend 36.
Desarbo, W. S., & Edwards, E. A. (1996). Typologies of compulsive buying behavior: A constrained clusterwise regression approach. Journal of Consumer
Psychology, 5, 231–262.
Faber, R. J., Christenson, G. A., Zwaan, M. D., & Mitchell, J. (1995). Two forms of compulsive consumption: Comorbidity of compulsive buying and binge
eating. Journal of Consumer Research, 22, 296–304.
Faber, R. J., & O’Guinn, T. C. (1989). Classifying compulsive consumers: Advances in the development of a diagnostic tool. Advances in Consumer Research, 16,
738–774.
Faber, R., & O’Guinn, T. C. (1992). A clinical screener for compulsive buying. Journal of Consumer Research, 19, 459–469.
Faber, R. J., O’Guinn, T. C., & Krych, R. (1987). Compulsive consumption. Advances in Consumer Research, 14, 132–135.
Feinberg, R. A. (1986). Credit cards as spending facilitating stimuli: A conditioning interpretation. Journal of Consumer Research, 13, 348–356.
92 H.-Y. Lo, N. Harvey / Journal of Economic Psychology 32 (2011) 79–92

Garcia, I. (2007). Compulsive buying: An irresistible impulse or a reflection of personal values? Revista de Psicologia Social, 22, 125–136.
Hirschman, E. C. (1979). Differences in consumer purchase behavior by credit card payment system. Journal of Consumer Research, 6, 58–66.
Koran, L. M., Faber, R. J., Aboujaoude, E., Large, M. D., & Serpe, R. T. (2006). Estimated prevalence of compulsive buying behavior in the United States.
American Journal of Psychiatry, 163, 1806–1812.
Kraepelin, E. (1915). Psychiatrie (8th ed.). Leipzig: Verlag Von Johann Ambrosius Barth.
Lee, S.-H., Lennon, S. J., & Rudd, N. A. (2000). Compulsive consumption tendencies among TV shoppers. Family and Consumer Sciences Research Journal, 28,
463–488.
Lejoyeux, M., Mathieu, K., Embouazza, H., Huet, F. O., & Lequen, V. R. (2007). Prevalence of compulsive buying among customers of a Parisian general store.
Comprehensive Psychiatry, 48, 42–46.
Magee, A. (1994). Compulsive buying tendency as a predictor of attitudes and perceptions. Advances in Consumer Research, 21, 590–594.
Maimaran, M., & Wheeler, S. C. (2008). Circles, squares, and choice. The effect of shapes arrays on uniqueness and variety seeking. Journal of Marketing
Research, 45, 731–740.
Manolis, C., & Roberts, J. A. (2008). Compulsive buying: Does it matter how it’s measured? Journal of Economic Psychology, 29, 555–576.
Manolis, C., Roberts, J. A., & Kashyap, V. (2008). A critique and comparison of two scales from fifteen years of studying compulsive buying. Psychological
Reports, 102, 153–165.
McElroy, S. L., Keck, P. E., Pope, H. G., & Hudson, J. I. (1994). Compulsive buying: A report of 20 cases. Journal of Clinical Psychiatry, 55, 242–248.
McElroy, S. L., Satlin, A., Pope, H. G., Keck, P. E., & Hudson, J. I. (1991). Treatment of compulsive shopping with antidepressants: A report of three cases. Annals
of Clinical Psychiatry, 3, 199–204.
Mowen, J. C. (2000). The 3M model of motivation and personality: Theory and empirical applications to consumer behavior. Boston: Kluwer.
National Credit Card Center (NCCC) (2008). Debit Card. 8-12-0008. <http://www.nccc.com.tw/dr_card/cardint5.htm>.
O’Guinn, T. C., & Faber, R. J. (1989). Compulsive buying: A phenomenological exploration. Journal of Consumer Research, 16, 147–157.
Reimers, S. (2007). The BBC internet study: General methodology. Archives of Sexual Behavior, 36, 147–161.
Reips, U. D. (2002). Standards for internet-based experimenting. Experimental Psychology, 49, 243–256.
Roberts, J. A. (1998). Compulsive buying among college students: An investigation of its antedecents, consequences, and implications for public policy.
Journal of Consumer Affairs, 32, 295–319.
Roberts, J. A. (2000). Consuming in a consumer culture: College students, materialism, status consumption, and compulsive buying. Marketing Management
Journal, 10, 76–91.
Roberts, J. A., & Jones, E. (2001). Money attitudes, credit card use, and compulsive buying among American college students. The Journal of Consumer Affairs,
35, 213–239.
Roberts, J. A., & Tanner, J. F. Jr., (2000). Compulsive buying and risky behavior among adolescents. Psychological Reports, 86, 763–770.
Scherhorn, G., Reisch, L. A., & Raab, G. (1990). Addictive buying in West Germany: An empirical study. Journal of Consumer Policy, 13, 355–387.
Schlosser, S., Black, D. W., Repertinger, S., & Freet, D. (1994). Compulsive buying: Demography, phenomenology, and comorbidity in 46 subjects. General
Hospital Psychiatry, 16, 205–212.
Skitka, L. J., & Sargis, E. G. (2006). The internet as psychological laboratory. Annual Review of Psychology, 57, 529–555.
Soman, D. (2001). Effects of payment mechanism on spending behavior: The role of rehearsal and immediacy of payments. Journal of Consumer Research, 27,
460–474.
Taiwanese Bank Association (2006). The statistical data for debt consultation. Bank Association.
Valence, G., D’Astous, A., & Fortier, L. (1988). Compulsive buying: Concept and measurement. Journal of Consumer Policy, 11, 419–433.
Visa Europe (2006). SEPA and the Internal Market for Payments: Liberating the potential of a truly efficient European payments market. 8-12-2008. <http://
www.visaeurope.com/pressandmedia/reports/main.jsp>.
Wiener, R. L., Holtje, M., Winter, R. J., Cantone, J. A., Gross, K., & Block-Lieb, S. (2007). Consumer credit card use: The roles of creditor disclosure and
anticipated emotion. Journal of Experimental Psychology: Applied, 13, 32–46.
Wong, N. Y., & Ahuvia, A. C. (1998). Personal taste and family face: Luxury consumption in Confucian and Western societies. Psychology & Marketing, 15,
423–441.
Yurchisn, J., & Johnson, K. K. P. (2004). Compulsive buying behaviour and its relationship to perceived social status associated with buying, materialism, self-
esteem, and apparel-product involvement. Family and Consumer Sciences Research Journal, 32, 291–314.