Chapter 2 Literature Review
All aspects of consumer behavior (e.g., the acquisition, use and disposition of goods) have an integral ethical component. The Hunt-Vitell Model is a major comprehensive theoretical model that can be readily applied to individual consumer behavior (Ferrell &
Gresham, 1985; Hunt & Vitell, 1986, 1993). As depicted in the Hunt-Vitell model (1993), the individual decision maker’s perception of an ethical problem in a situation is followed by the perception of various possible alternatives that might be used to resolve the problem. When the consumer has determined a set of perceived alternatives or courses of action, two major ethical evaluations might take place: deontological evaluation and/or a teleological evaluation.
The deontological evaluation focuses on the specific actions or behaviors of the consumer. In making a deontological evaluation, the individuals attempt to evaluate the inherent rightness versus wrongness of various behaviors or alternatives. On the other hand, teleology focuses on the consequences of the behaviors. The key issue in the teleological evaluation holds that the moral worth of actions or practices is determined solely by the consequences of the actions or practices. A particular behavior is considered the most ethical if the consequences bring a greater balance of good over bad than any other alternative.
2.1 Ethical Judgments
In most situations a consumer’s ethical judgments are likely to be a function of both the deontological and teleological evaluations. Ethical judgments (the extent of which one believes that an action is moral rightness or wrongness) determine behavior through the intention, but ethical judgments might differ somewhat from intentions because one’s teleological evaluation also affects intentions directly.
General ethical judgments refer to an individual’s subjective probability beliefs concerning various ethically-questioned consumer behaviors represented by four basic consumer ethical dimensions (e.g., Chan et al., 1998; Muncy & Eastman, 1998). To study this construct, Muncy and Vitell (1992) developed a consumer ethics scale in which questions with ethical implications can be divided into four classes. They were used for measuring ethical judgments. The first class, “actively benefiting from illegal activities”, consists of questions regarding actions that are initiated by the consumer and that are almost universally perceived as illegal (e.g., drinking a can of soda in a supermarket without paying for it). The second class, “passively benefiting at the expense of others”, includes questions regarding consumers taking advantages of mistakes by sellers (e.g., not saying anything when the waitress miscalculates the bill in your favor). The third class, “actively benefiting from questionable actions or behaviors”, includes questions regarding situations in which the consumer is actively involved in some deception, but their actions may not necessarily be considered illegal (e.g., using a coupon for merchandise they did not buy). Finally, the last class, “no harm”, includes questions regarding situations in which consumers perceive their actions as doing little or no harm/foul (e.g., using computer software or games they did not buy). Since the measures of general ethical judgments focus on the actions and consequences of those behaviors, individuals who believed in the teleological concept are more likely to agree with the questionable consumption behaviors, whereas individuals who believed in the deontological concept are more likely to use higher ethical standards in judging the questionable consumption behaviors.
2.2 Ethically Questionable Behavior
There are a number of studies concerning ethically questionable behavior (EQB) in consumption. Research on EQB comprises two streams (Fukukawa, 2002). The first stream concerns specific EQB; investigating the decision-making in relation to a specific issue of
EQB, seeking to understand components of attitude, a formation of intention and causes of behavior in relation to that specific issue, including tax evasion (e.g., Hessing et al., 1988), shoplifting (e.g., Babin & Griffin, 1995; Krasnovsky & Lane, 1998), software piracy (e.g., Glass & Wood, 1996; Wagner & Sanders, 2001), and counterfeiting (e.g., Albers-Miller, 1999).
The second stream investigates ethical judgments regarding EQB in different settings (Muncy
& Vitell, 1992). That is, Muncy and Vitell (1992) investigated four dimensions of ethical judgments (actively benefiting from illegal activities, passively benefiting at the expense of others, actively benefiting from questionable actions, and No harm), consisting of 27 different kinds of EQB, in different settings (e.g., Austria, Rawwas, 1996; other EU countries (e.g., Greece, Italy and Spain), Polonsky et al., 2001). However, there is a paucity of research investigating the relationship between general ethical judgments and specific kinds of EQB.
Specifically, examining how certain judgments may be related to specific kinds of EQB is valuable.
2.3 Conceptual Framework
Personality, Ethical Judgments, and Behavioral Intention to Not Repay
Figure 2.1 introduces a conceptual model, depicting the relationships among personality, ethical judgments, and intention to not repay. The first part of the model emphasizes the personality traits of locus of control and risk-taking propensity. These traits are selected because they are theoretically related to EQB. According to Mudrack (1993), one individual difference variable worth examining within the ethics literature is locus of control (LOC).
Some studies have stated that LOC can influence the unethical behavior of individuals (e.g., Bass et al., 1999). Additionally, Rallapalli et al. (1994) studied the interrelationships between general ethical judgments and personality traits of consumers. Of the many possible
personality traits (e.g., aggressiveness; achievement), only risk propensity was significantly and positively related with all dimensions of ethical judgments.
The second part of the model depicts that these personality traits will have influence on each of the four dimensions of ethical judgments. An individual’s ethical judgment is the degree to which he or she considers various ethically-questioned consumer behaviors morally acceptable (e.g., Chan et al., 1998; Muncy & Eastman, 1998). Muncy and Vitell (1992) developed a consumer ethics scale in which questions with ethical implications can be divided into four classes. They were used for measuring ethical judgments, including actively benefiting from illegal activities, passively benefiting at the expense of others, actively benefiting from questionable actions or behaviors, and no harm.
The third part of the proposed model describes anticipated relationships between ethical judgments and intention to not repay. Actual behavior is ultimately the topic of interest, but this study has difficulty in measuring actual behavior of this nature. That is, we may be violating individuals’rights to privacy by asking about unethical behavior. Though some factors and situations may interfere or constrain an individual’s ability to act upon his or her intentions, intention is still an important construct found to relate significantly to actual behavior (March & Woodside, 2005). As a result, behavioral intention is used to serve as an adequate surrogate measure (Jones & Kavanagh, 1996). Additionally, the model postulates a direct influence of locus of control on intention to not repay.
Figure 2.1 Hypothesized Model of Personality Traits, Ethical Judgments, and Behavioral Intention
Actively benefiting from illegal activities (F2)
Passively benefiting at the expense of others (F3)
Actively benefiting from questionable actions (F4)
No harm (F5)
Behavioral intention to not repay (F1)
Personality Traits Ethical Judgments Behavioral Intention
Locus of control (LC)
Risk-taking propensity (R)
Although not shown in the model, individual age and level of education are included as control variables. Age appears related to ethical judgments, with older consumers being more ethical (Muncy & Vitell, 1992; Rawwas & Singhapakdi, 1998; Vitell, 2003; Vitell et al., 1991).
Studies also indicate that people with high debt levels tend to be younger and less educated (Dessart & Kuylen 1986; Sullivan et al., 1989). To reduce the likelihood that individual demographic characteristics would confound the relations examined in this study, it is necessary to control for these two characteristics.
Personality Influences on Ethical Judgments
The majority of research on ethical judgments (the second stream of research regarding EQB) has focused on examining consumer ethical judgments towards EQB in different settings (e.g., Rawwas, 1996). Previous studies generally regard ethical judgments as dependent variables and explore their antecedents (e.g., individual ethical ideologies) or examine the relationships between these ethical judgments and other variables (e.g., attitudinal characteristics; materialism) (Chan et al., 1998; Muncy & Eastman, 1998; Vitell &
Muncy, 1992). Little research has examined the relationship between personality and general ethical judgments. Only Rallapalli et al. (1994) suggested that there were significant interrelationships between consumer ethical judgments and personality traits. Rallapalli et al.
(1994) also argued that individual personality traits influenced consumer ethical judgments/actions.
A few studies on ethics have explored individual differences in how people deal with moral judgments. Each of these studies has attempted to delineate the ethical frameworks underlying ethical judgments (e.g., Brady, 1985; Kohlberg, 1984; Velasquez, 1992).
Underlying each of these is a belief that the ethical judgments of individuals are affected by relatively stable individual differences in ethical ideologies. In light of the contingency
framework for understanding ethical decision making in marketing (Ferrell & Gresham, 1985), personality traits are an integral part of individual factors that can potentially affect ethical/unethical decision-making. In a related study, Munch et al. (1991) suggested that understanding consumer personalities could help understand the process used by consumers to resolve ethical dilemmas. However, the type of personality traits that influence consumer ethical judgments and how they do so is less clear.
The most widely studied personality in the ethics literature is locus of control, which has gained acceptance through several models of ethical decision-making (e.g., Stead et al., 1990;
Treviño, 1986). In accordance with Rotter’s (1966) argument, LOC is a personality variable manifested by the extent to which individuals believe events are contingent upon their own behavior or characteristics. An individual with an internal LOC perceives outcomes to be a direct result of his or her efforts, whereas an individual with an external LOC perceives outcomes to result from external forces beyond his or her control, such as fate or the actions others. Hence, individuals with an internal LOC are more inclined to take responsibility for the consequences of their behavior than are individuals with an external LOC, and also have higher ethical standards.
Some empirical studies have concluded that individuals with external locus of control probably form intentions to behave unethically because they are inclined to ascribe responsibility to others or to situational factors (Dozier & Miceli, 1985; Hegarty & Sims, 1978, 1979; Lefcourt, 1982; Singhapakdi & Vitell, 1991; Treviño & Youngblood, 1990). Thus, it appears that, generally, individuals with external locus of control may see an increased likelihood of shirking responsibility and thus may display low ethical standards. Thus, this study proposes a positive relationship between external LOC and ethical judgments regarding ethically questionable behaviors.
H1a: External locus of control is positively related to the ethical judgments regarding actively benefiting from illegal activities.
H1b: External locus of control is positively related to the ethical judgments regarding passively benefiting at the expense of others.
H1c: External locus of control is positively related to the ethical judgments regarding actively benefiting from questionable actions.
H1d: External locus of control is positively related to the ethical judgments regarding no harm/ no foul.
The trait of risk-taking propensity can also be reasonably expected to be associated with ethical decision-making. Fukukawa (2002) suggested that individual attitude towards performing EQB may be influenced by one important explanatory factor: risk-taking.
Risk-taking is related to the extent to which a person seeks to be exposed to uncertain situations, especially uncertainty involving financial gains or losses. This factor could be interpreted as a type of personality trait concerned with the inclination of an actor to either seek or avoid risk (Kowert & Hermann, 1997). Studies have shown that risk-taking predicts criteria related to unethical behavior. For example, in Jackson et al. (1972), the relationship between risk-taking and willingness to behave unethically in various hypothetical situational dilemmas reaches 0.44. Rallapalli et al. (1994) suggested that individuals with higher risk propensity tended to have stronger belief in benefiting from both illegal and questionable actions than individuals with lower risk propensity. Such individuals with higher risk propensity are likely to display greater need for sensation seeking, and may seek risk regardless of whether the decisions are framed as gains or losses (Tokunaga, 1993). Vitell et al.
(1990) also showed that individuals with high risk taking propensity were more willing to take positions that were less socially desirable or morally questionable. Such individuals
appear to place greater emphasis on “substance over form”, and may consider breaking rules if doing so would best serve the group (Cohen et al., 1995). On the basis of the above arguments, it is conceivable that under conditions such as an individual possessing higher risk-taking propensity, that individual may be more likely to benefit substantially from unethical behavior. Thus, this study proposes a positive relationship between risk-taking propensity and ethical judgments regarding ethically questionable behaviors.
H2a: Risk-taking propensity is positively related to the ethical judgments regarding actively benefiting from illegal activities.
H2b: Risk-taking propensity is positively related to the ethical judgments regarding passively benefiting at the expense of others.
H2c: Risk-taking propensity is positively related to the ethical judgments regarding actively benefiting from questionable actions.
H2d: Risk-taking propensity is positively related to the ethical judgments regarding no harm/ no foul.
Ethical Judgments and Behavioral Intention
Past studies have identified the link between attitudes or judgments concerning an action and intention to perform that action. For example, Randall (1989) surveyed empirical studies that had examined the Fishbein and Ajzen (1975) model and found that the link between judgments and intentions was substantiated. Furthermore, some studies have argued that ethical judgment for certain behavior influences this unethical behavioral intention/behavior (Barnett & Valentine, 2004; Bass et al., 1999; Kwong et al., 2003). Such studies have
any particular alternative will be chosen”and postulated that “ethicaljudgments impact behavior through the intervening variable of intentions”. Moral evaluation is a key component in the model of ethical decision-making (e.g., Akaah & Riordan, 1989; Hunt & Vitell, 1986).
Individuals confronting an ethical dilemma evaluate it on the basis of relevant ethical cognitions about themselves. Interestingly, an individual’s low ethical standards may be the key variable that breeds unethical behavior (Vitell et al., 1991). Thus, similar to the ethical judgment concerning specific action, general ethical judgments (actively benefiting from illegal activities, passively benefiting at the expense of others, actively benefiting from questionable actions, and no harm) are also essential in the formation of behavioral intention regarding ethical or moral issues, because they are integral to individual attitudes toward ethical issues (Chiu, 2003). Therefore, this study predicts the following four relationships between ethical judgments regarding ethically questionable behaviors and intention to not repay.
H3a: The ethical judgments regarding actively benefiting from illegal activities are positively related to intention to not repay.
H3b: The ethical judgments regarding passively benefiting at the expense of others are positively related to intention to not repay.
H3c: The ethical judgments regarding actively benefiting from questionable actions are positively related to intention to not repay.
H3d: The ethical judgments regarding no harm/no foul are positively related to intention to not repay.
Personality and Behavioral Intention to Not Repay
There are two reasons for theorizing a linkage between the trait of LOC and intention to
not repay. First, following Hypotheses 1a~1d and Hypotheses 3a~3d, LOC should indirectly influence intention to not repay through ethical judgments. For example, individuals with external LOC are more likely to attribute responsibility to others or to situational factors, leading to their lower standard of ethics in judging the acceptability of questionable actions, thereby yielding unethical behavioral intentions. Second, external LOC may influence intention to not repay directly in terms of psychological characteristics. Individuals with a strong external LOC are more likely to get into debt, such that they easily suffer credit-related problems, because they view money from a more extrinsic, materialistic perspective than a utilitarian one, and also display anxiety about their inability to handle money (Tokunaga, 1993). In addition, Perry and Morris’s (2005) findings indicated that, although financial knowledge and income are important, those who believe that financial outcomes are due to chance or powerful others, i.e., externals, will be slightly less likely to take steps to manage their finances. Thus, this study adds a direct link from LOC to intention to not repay.
H4: There is a direct positive relationship between external locus of control and intention to not repay.
Based on the above hypothesis development, ethical judgments (actively benefiting from illegal activity, passively benefiting at the expense of others, actively benefiting from questionable actions and no harm/foul) may play two roles in the relationships between personality (LOC and risk-taking propensity) and intention to not repay. First, ethical judgments may partially mediate the effect of LOC on intention to not repay. This relation suggests that LOC may have a direct influence on intention to not repay, but may also be related to intention to not repay through the role in shaping ethical judgments. Individuals with higher external LOC may be more likely to not repay credit card expenses partly because of their lower ethical standards and partly because they view money from an extrinsic and
money regardless of ethical judgments. Second, ethical judgments may fully mediate the relationship between risk-taking propensity and intention to not repay. Risk-taking propensity may be associated with the intention only through ethical judgments made during credit card use. Individuals with higher risk-taking propensity may be more likely to seek uncertain situations and to exhibit low uncertainty avoidance, making him or her have lower ethical standard, and in turn more probably lead to intention to not repay.
Chapter 3 Methods
3.1 Sample
The data were collected by questionnaires, of which 448 usable questionnaires from those holding credit cards in Taiwan (at least 20 years old) were obtained. In recent years, heavy debt levels of cardholders and over unethical marketing practices by credit card solicitors have created many social problems in Taiwan. It appears that samples from Taiwanese cardholders should be appropriate for this study. We sampled four areas in Taiwan (the areas in the North, Middle, South and East) to conduct the field survey. In each area, several representatives helped to distribute questionnaires. A convenience sample was used in this study, but sample diversification (see Table 3.1) and sample relevance have been taken into account (Sackett & Larson, 1990). In our study the sample possessed the essential person and setting characteristics (i.e., using credit cards to consume) that defined membership in the intended target population/universe.
Overall, of the 500 that were distributed, 448 completed questionnaires were received.
Of the 448 credit cards consumers, 271 (60.5%) were male and 410 (91.5%) were 20~40 years old. The sample was highly educated, with 69.2% holding a college degree and 12.9%
with a graduate degree. During the survey for this study, respondents were assured that all information would be kept strictly confidential in order to increase the response rate and acquire more accurate information.
3.2 Measures
Locus of Control
We measured locus of control by using an abbreviated scale of LOC having 11 items developed by Barnett and Lanier (1995). Subjects were asked to respond to the items in the original forced-choice format by selecting the statement from each pair which best reflected their thoughts. Scores for each respondent ranged from 0 for an individual who selected all
“internal”statements, to a score of 11 for an individual who selected all “external”statements.
We measured a person’s LOC by using scores summed on 11 items. The coefficient alpha for the LOC measure was 0.63 in the current study, less than the commonly used criterion of 0.7.
However, since the measure items used the 0-1 scale and their items reflected widely different situations, the reliability of our measurements should be acceptable. As argued by Nunnally and Bernstein, the “heterogeneity would be a legitimate part of the test if it were part of the domain of content implied by the construct”(Nunnally & Bernstein, 1994: 312).
Risk-Taking Propensity
Risk-taking propensity was adapted from the Jackson Personality Inventory (JPI) (Jackson, 1976). A subset of the statements from the original JPI were selected in such a way that they can better reflect culture heterogeneity. After interviewing three scholars with related expertise, the scale for risk-taking propensity was composed of 12 statements. Respondents were asked to indicate “disagree”or “agree”to a series of 12 statements. A score of 0 was given for “disagree”and 1 for “agree”for each statement. The summation of responses yielded risk-taking propensity scores from 0 to 12. Higher scores reflected individuals who were prone to higher risk-taking. The coefficient alpha for this measure was 0.71.