Integrated Moral Conviction Theory of Student Cheating: An Empirical Test

Abstract

In this article, we develop an integrated moral conviction theory of student cheating by integrating moral conviction with (a) the dual-process model of Hunt–Vitell’s theory that gives primacy to individual ethical philosophies when moral judgments are made and (b) the social cognitive conceptualization that gives primacy to moral identity. We found empirical support for our proposed model in a study with 311 business students where moral conviction predicted student moral disengagement and subsequent unethical decision making related to academic dishonesty not only directly but also indirectly through ethical philosophy and moral identity. Based on these results, we derive specific implications for teaching and learning practice.

Keywords

moral conviction student cheating ethics moral identity moral disengagement

Introduction

Some students cheat. It is an unfortunate characteristic of the college environment that professors routinely deal with student dishonesty cases. Numerous studies have found that a significant proportion of students behave dishonestly in classes, with rates as high as 66% across all students, while business students, at approximately 84%, show even higher amounts (McCabe & Treviño, 1995). Considering the pervasiveness of cheating among college students, understanding why some students cheat while others do not is important to teachers’ ability to secure and maintain academic integrity among all of their students. Gholami and Tirri (2012) found that educators who initiate an empathetic, yet assertive, dialog and listen to student responses may better understand why some students disengage morally and engage in academic cheating. Educators may find that some students frame academic cheating as a moral issue, while other students perceive cheating as a nonmoral issue and a matter of personal preference or conventional practice. By understanding student framing of the decision to cheat, we begin to deconstruct the role of moral disengagement in the unethical decision-making process related to student cheating.

Moral disengagement, a propensity to engage in self-deception and make unethical decisions appear more acceptable (Bandura, Caprara, Barbaranelli, Pastorelli, & Regalia, 2001), refers to the extent to which students tend to disengage their internalized self-regulation by attempting to reframe a moral issue and neutralize it as nonmoral and thus make it appear justifiable and permissible (Duffy, Aquino, Tepper, & O’Leary-Kelly, 2005). We posit that a previously unexamined construct driving moral disengagement and unethical decision making is moral conviction, which is individual variation in the extent to which people frame an issue, choice, or situation as moral (Skitka, Bauman, & Sargis, 2005; Skitka, Washburn, & Carsel, 2015). This framing of issues, choices, or situations by individuals increases individual vulnerability to make snap decisions about morality with automaticity driven by bounded ethicality (Chugh, Bazerman, & Banaji, 2005; Kern & Chugh, 2009). Variation in moral conviction has important implications for behaviors influenced by controversy and therefore, might be relevant to student moral disengagement and their subsequent (un)ethical decisions regarding academic cheating.

We focus on the relationship between the concepts of moral conviction and moral disengagement to address whether variation in the extent to which students moralize the issue of cheating influences their moral disengagement and misconduct. We also investigate whether moral conviction extends this influence beyond that of individual ethical philosophy and moral identity. To address these issues, we first integrate ethical moral philosophies (i.e., consequentialist vs. nonconsequentialist moral philosophies), and follow that with a discussion of moral conviction theory. Next, to model different paths leading to students’ propensity to use mechanisms of moral disengagement, we incorporate moral conviction theory with both the dual process version of Hunt–Vitell’s theory and the social cognitive theory of moral identity. Building on these integrated theoretical perspectives, we then propose and test a theoretically derived model. Finally, we discuss the results of our analysis and summarize the implications of our findings.

Theoretical Foundations

Researchers using novel theoretical perspectives and models (e.g., Haidt’s [2001] social intuitionist model) have added to cognitive and socioemotional factors with the potential to explain differences in processing information about moral and nonmoral (i.e., preferential and/or conventional) issues that trigger the (un)ethical decision-making process (Hunt & Vitell, 1993). Originally, Hunt and Vitell (1986) proposed a cognitive model of moral decision making that included both rule-based and cost-benefit cognitive reasoning in the process of unethical decision making. They suggest that these two individual ethical philosophies about moral issues are the best predictors of unethical decision making and subsequent (un)ethical behavior (e.g., cheating). They posit that individual variations in weighing of different evaluations of ethical philosophies with regard to perceived consequences (i.e., teleology; Huang, 2004) and formulated rules (i.e., deontology; Ashby, 1950) influences the (un)ethical decision about a moral issue (see also Buckley, Fedor, & Marvin, 1994; Longenecker & Ludwig, 1990; Vitell & Hunt, 2015).

Recent empirical studies based on social cognitive theory suggest that an individual’s moral identity also influences moral judgment or unethical decision making (Aquino & Reed, 2002). From a social cognitive perspective, moral identity is a source of moral motivation that reflects the extent to which values, goals, and virtues are central elements of a person’s identity (Aquino & Reed, 2002; Hardy, 2006). Past studies identified two dimensions in which moral identity is self-important to individuals (Aquino & Reed, 2002). The first dimension reflects one’s private moral identity or the degree to which the individual’s self-concept defined their moral traits (internalization), while the second reflects one’s public moral identity or the degree to which these moral traits manifest publicly through an individual’s actions in the social context (symbolization). As internalization and symbolization vary in dominance in the self-concept, so too does the extent to which individuals construct their moral identity in terms of the moral traits (e.g., fairness and honesty; Weaver, 2006).

Most recently, Skitka and associates (Skitka, 2010; Skitka, Bauman, & Lytle, 2009; Skitka et al., 2015) proposed a new theoretical perspective emphasizing the role of moral conviction about morality or immorality of specific issues (e.g., academic cheating). Specifically, they suggest people making moral judgments vary in their “upfront understandings” or convictions that guide their feelings whether something is right or wrong. These moral convictions are particularly salient when addressing issues that are complex and controversial with the potential to provoke not only a cognitive but also an emotional response.

This article combines the above three theoretical perspectives (i.e., Hunt–Vitell’s model of moral philosophies, Aquino and Reed’s theory of moral identity, and the theory of moral conviction proposed by Skitka and her colleagues) to propose an integrated moral conviction model of academic cheating among students (particularly business students). The main contribution of our model is a novel proposition that moral conviction likely influences student propensity to use moral disengagement mechanisms through upheld and self-regulatory moral standards of academic honesty, moral identity, and ethical philosophies. Extending other research (e.g., Dalton, 2015; Dalton & Crosby, 2010), we propose that a high level of moral conviction assists students in making ethical decisions by deterring them from engaging in academic cheating (e.g., Woodbine & Amirthalingam, 2013). Our proposed theoretical model is shown in Figure 1, and depicts multiple paths through which moral conviction influences unethical decision making. The following sections provide theoretical and empirical support for our specific hypotheses which account for indirect influences that are conveyed via moral identity, ethical philosophy, and moral disengagement mechanisms.

Figure 1.

Integrated moral conviction model of academic dishonesty.

Integration of Moral Philosophies on Moral Judgment¹

There are two broad perspectives of normative ethical philosophies on moral judgment and moral behaviors (Ferrell, Crittenden, Ferrell, & Crittenden, 2013; N. E. Landrum, 2001): consequentialism and nonconsequentialism. Consequentialist-based ethics are a family of philosophies which emphasize the effect of an agent’s action on other people and the “goodness/badness” of the outcomes of such acts (e.g., students who witness or engage in academic cheating but defer to observe the professor’s (re)action to such unethical behavior). Consequences arising from an individual’s actions can be either physical, emotional, or cognitive (Greene, Nystrom, Engell, Darley, & Cohen, 2004; Greene, Sommerville, Nystrom, Darly, & Cohen, 2001). Well-known consequentialist theories include (a) egoism and (b) utilitarianism. In ethical egoism, the agent focuses on satisfying his or her self-interest. In utilitarianism, the agent is concerned with maximizing the greatest good for the greatest number of people based on the importance of one’s conditional duty, or actual moral obligation, to act virtuously in every situation (Ross & Stratton-Lake, 2002).

Nonconsequentialism, on the other hand, refers to rule/duty-based ethical theories, which stress the importance of rules and rights in moral judgment and moral decision making. Three well-known nonconsequentialist theories include (a) deontology (especially, Kantian theory on moral philosophy), (b) formalism, and (c) contractarian. Deontology is based on the comparison of actions with existing deontological norms to ensure adherence with existing norms (Hunt & Vitell, 1986, 2006; Rendtorff, 2015; Vitell & Hunt, 2015; Wiltermuth, Bennett, & Pierce, 2013). Kantian deontology argues in favor of dutifully adhering to rules, irrespective of the consequence—good or bad. When describing formalist, Wiltermuth et al. (2013, p. 281) succinctly state that,

People with formalist ethical predispositions focus on rules or principles of behavior when determining the ethicality of actions and ignore consequences of the actions except insofar as those consequences affect which rules or principles govern the behavior (e.g., Kant, 1785/1994).

This account showcases the conceptual similarity between formalism and deontological moral philosophy. Contractarian nonconsequentialism supports adherence to broad social principles (as opposed to personal or organizational principles) in guiding individual behavior (N. E. Landrum, 2001).

The unique differences between these moral philosophical foundations of decision making, moral judgments, and moral behaviors are governed by dual consequentialist and nonconsequentialist processes, such as deontology versus teleology, cognition versus social conventions, utilitarianism versus formalism, and liberalism versus conservativism (Greene et al., 2001; Greene et al., 2004; Hunt & Vitell, 2006; Rendtorff, 2015; Vitell & Hunt, 2015; Wiltermuth et al., 2013). It seems that the use of normative theories of moral behavior may provide three benefits to reasoning by business students. First, they provide students with a vision of acceptable moral behavior. Second, an understanding of these ethical theories helps students recognize the importance of making ethical decisions when facing ethical dilemmas (e.g., to engage in academic cheating). Third, students can understand and appreciate the role of organizational culture in mitigating peer pressure (Dalton & Crosby, 2010; Ferrell et al., 2013) to engage in academic cheating.

Integrated Moral Conviction Theory

Moral conviction, which refers to a “strong absolute belief that something is right or wrong, moral or immoral” (Skitka, 2002, p. 453), stems from social domain theory (Turiel, 2006). Social domain theory suggests that individuals combine three different domains of social knowledge in a coordinated manner: (a) psychological (from which they derive their individual preferences), (b) societal (from which they derive their perceptions of different conventions), and (c) moral (from which they derive their perceptions of goodness and fairness; Nucci, 2001; Turiel, 2002). Based on this theory, one’s moral conviction varies by the extent to which one combines and coordinates these domains and by giving primacy to a specific domain. In this way, moral conviction is “high” among those that tend to moralize issues by emphasizing moral elements and is “low” among those that tend to be more pragmatic and frame issues by emphasizing preferential or conventional elements (Richardson, Mulvey, & Killen, 2012).

The integrated theory of moral conviction articulates moral conviction as an essential attitude about morality, amorality, and immorality (for a comprehensive review, see Skitka, 2010). This theory posits that individual attitudes held with strong moral conviction differ from strong nonmoral attitudes because strong individual moral convictions are “experienced as a unique combination of factual belief, compelling motive, and justification for action” (Skitka, 2010, p. 270). This means that individuals (e.g., business students) vary not only in the extent to which they combine and coordinate their three social domains but also in the extent to which (a) they believe that it should both be held by and apply to everyone, (b) they see those convictions as facts, and (c) they experience strong emotions when moralizing the issue considered (Skitka et al., 2015). When a student makes a moral or nonmoral appraisal of academic honesty as an issue, this appraisal activates his or her emotional response that influences his or her cognitive processing of moral disengagement. This means that the influence of emotion on moral judgment is tempered by the related moral disengagement.

Students’ moral disengagement manifests as a result of the extent to which they moralize the issue of academic dishonesty as wrong, which causes them to vary in their propensity to use mechanisms of moral disengagement to make unethical decisions (Haidt, Rosenberg, & Hom, 2003). In other words, moral tolerance toward or propensity for moral disengagement and unethical decision making about academic dishonesty is more likely to be encountered among those business students with nonmoral attitudes about academic dishonesty than among those with strong moral convictions about academic dishonesty. Hence, we propose

Hypothesis 1: The relationship between students’ moral conviction and their unethical decision making is mediated by their propensity morally disengage.

Integrating Moral Conviction Theory and Hunt–Vitell Theory

Research investigating moral judgment and unethical decision making has a long tradition (Craft, 2012), with early researchers focusing their attention on rational–descriptive models of moral or ethical judgment developed by Kohlberg (1973) and Rest (1986). However, the Hunt and Vitell (1986, 1993, 2006) positivist model posits that a decision maker’s moral judgment is based on cognitive processing schemes that simultaneously incorporate deontological and teleological ethical philosophies. The main assumption of this model is that the decision maker typically includes both nonconsequentialist and consequentialist evaluations into his or her moral judgment.

For the deontological evaluation, after recognizing an ethical problem, the individual considers appropriate behavior resulting from solutions in the consideration set of the moral dilemma. Individuals evaluate and compare the possible alternatives within the consideration set by applying deontological norms arising from personal values and beliefs regarding appropriate behavior. According to Hunt and Vitell (1986), behavioral norms manifest as behavioral rules associated with abstract concepts (e.g., honesty, justice, peace) as well as more concrete issues (e.g., product safety, truth-in-advertising, consumer privacy). As such, this portion of the process is inherently rule-based, and thus represents in general a nonconsequentialist approach, and, in particular, a formalist evaluation style.

In contrast, teleological evaluations are aimed at assessing the desirability of the perceived consequences on selecting a particular alternative from the consideration set (Murphy & Laczniak, 1981; Novicevic, Buckley, Harvey, & Fung, 2008). Hunt and Vitell (1986) argue that, “the overall result of the teleological evaluation will be beliefs about the relative goodness versus badness brought about by each alternative as perceived by the individual” (p. 9), making this portion of the process consequentialist in nature. They also claim that, in theory, individuals might rely solely on one process or the other (i.e., consequentialism vs. nonconsequentialism), but in reality, such situations will be extremely rare. In addition, the authors do not indicate any individual differences or situational characteristics which may influence individuals to rely strictly on either deontological or teleological moral evaluations.

The Hunt–Vitell model has accumulated significant empirical support for the model’s main hypothesis that deontological and teleological evaluations are influential to ethical judgment (Hunt & Vitell, 2006). However, we know little about the psychological mechanisms that affect these two moral philosophical evaluations and their interplay. To that end, neuroscience research supports the inclusion of emotion in research on unethical decision making (Greene et al., 2001; Greene et al., 2004), and by integrating this field with the extant Hunt–Vitell theory, we expect greater insight into the complex nature of cognitive evaluation processing during ethical dilemmas. Decision makers’ emotional responses (or the lack thereof) to a moral dilemma about a given issue frames their preference for either a deontological or a consequentialist evaluative path to (un)ethical judgment. While Hunt and Vitell (1993), who do not include emotions into their model, argue that an either–or dichotomy among the processes is relatively rare, the dual-process model does allow that the presence of a highly charged emotional issue may assert undue influence toward one cognitive process over the other.

For that reason, moral judgment is often a function of emotion as individuals rely on reasoning and cognition only if there is a need to justify their moral judgments (Greene, 2007). Indeed, recent work in this area indicates that neither cognition nor emotion or intuition can independently fully explain the process of individual moral judgments (Dedeke, 2015; Greene, 2007; Greene et al., 2001; Greene et al., 2004; Schwartz, 2015). The dual-process theory of moral judgment posits that “automatic emotional responses and more controlled cognitive responses play crucial and, in some cases, mutually competitive roles” (Greene, Morelli, Lowenberg, Nystrom, & Cohen, 2008, p. 1145) in individuals’ formation of moral judgments. Accordingly, cognition is the primary driver of teleological, utilitarian moral judgments made by individuals; while emotion is the primary driver of deontological, formalist moral judgments made by individuals (Greene, 2007). Emotion, however, has its role within each judgment process, by serving different purposes. In particular, within consequentialist judgments, emotion serves as a quantifiable commodity to be included in any calculus of outcomes, while in nonconsequentialist judgments emotion serves as a warning against decisions that violate deeply held personal values thus resisting the urge to morally disengage and behave unethically (Greene, 2007). Based on this, we anticipate that students holding strong moral convictions on the issue of academic cheating make deontological, formalist evaluations of cheating via their emotional response and their propensity to morally disengage will be reduced. Therefore, we propose

Hypothesis 2: The relationship between student moral conviction and their propensity to use mechanisms of moral disengagement is mediated by student formalist (deontological) ethical philosophy.

Integrating Moral Conviction Theory and the Social Cognitive Conceptualization of Moral Identity

Moral identity was originally defined by Hart, Atkins, and Ford (1998) as “a commitment to one’s sense of self to lines of action that promote or protect the welfare of others” (p. 515). Since their work, the conceptualization of moral identity has shifted from the character-based idea of being the “perfect” moral person to that of holding the cognitive self-schema. Specifically, Aquino and Reed (2002) argue that although moral identity is characterized by explicit moral attributes, it “is also amenable to a distinct mental image of what a moral person is likely to think, feel, and do” (p. 1424). In other words, individuals using a social referent to define their moral identity take the traits of that individual or group and incorporate them into their own self-schema (Bing et al., 2012; Reed, 2002).

The cognitive schema conceptualization of moral identity proposed by Blasi (1984) and operationalized by Aquino and Reed (2002) stems from several assertions made about moral identity. The first assertion is that differences exist between individuals’ moral identities (Blasi, 1984). Specifically, individuals emphasize different attributes as being central to their identity, such as being trustworthy or kind, although there may be several moral attributes which are more commonly held among different people. The second assertion made by Aquino and Reed is that moral identity varies to the extent that individuals hold that identity at their core. Hart et al. (1998) further asserted that strength of identities held by individuals is malleable and may change, suggesting the possibility that the moral identity itself may change if individuals experience large shifts in attributes they hold as central to their core. Finally, Aquino, Reed, Thau, and Freeman (2007) asserted that individuals with strong moral identities, those who hold their moral identity as more important in their self-definition, will likely be more cognizant of others when decisions are made, particularly when such decisions are detrimental to those others.

Moral identity is theoretically grounded in social cognitive theory of moral agency (Bandura, 2001). A social cognitive conceptualization of moral identity posits that its salience and central moral importance in one’s self-schema are the primary factors that facilitate its activation (Aquino, Freeman, Reed, Lim, & Felps, 2009; Vitell et al., 2009) and suppress its influence on moral disengagement and unethical decision making (McFerran, Aquino, & Duffy, 2010). This process of activation and suppression can be also influenced by the individual’s (i.e., students’) internal moral or nonmoral framing of an external issue, such as academic dishonesty, because this issue will to a varying degree prime students’ moral schema for propensity to use mechanisms of moral disengagement in their unethical decision-making process. In effect, students who are high in moral identity are less likely to rationalize academic dishonesty and refrain from unethical decision making because they will strive to sustain a consistency between their moral selves and their decisions. Therefore, we propose,

Hypothesis 3: The relationship between student moral conviction and their propensity to use mechanisms of moral disengagement is mediated by student moral identity.

Social cognitive theory has also introduced the concept of moral disengagement, which posits that individuals use self-regulating processes to govern their own thoughts and behaviors (Bandura, 1999). Moral disengagement is described as the deactivation of the self-regulatory processes on which an individual typically relies to avoid actions that are unethical and which impose intraindividual restrictions in the form of internalized moral standards (Detert, Treviño, & Sweitzer, 2008). The deactivation of these self-regulatory mechanisms leads to the deactivation of self-monitoring processes that are the self-regulatory processes rooted in social cognitive theory (Bandura, 1999). Based on this theory, the elimination of the self-monitoring process allows individuals to bypass the negative emotions which typically accompany one’s unethical actions. Individuals eliminate self-monitoring and deactivate self-regulatory processes by using personal devices or mechanisms of moral disengagement, such as moral justification, euphemistic labeling, advantageous comparison, displacement of responsibility, diffusion of responsibility, disregarding or distorting the consequences, dehumanization, and attribution of blame (Moore, 2015). As the deactivation of self-regulatory mechanisms leads to moral disengagement, which increases the likelihood that individuals decide to deviate from moral standards (Detert et al., 2008), we propose,

Hypothesis 4: Student propensity to use mechanisms of moral disengagement is positively related to unethical decision making.

Method

Data Collection and Sample Characteristics

Data were collected from undergraduate business students enrolled in four different required junior classes taught in the business school of a midsized university located in the Southeastern United States (the majority of students, as is typical in public universities of this size in this geographic region, are from the United States). Survey completion was voluntary and extra credit was offered for participation. For those who chose to participate, time was given during class to complete the surveys. Completed surveys were received from 311 respondents out of a potential 352 (88% response rate), with the remaining business students either being absent on the day of data collection or choosing not to participate.

Measures

Moral Conviction

Moral conviction was measured with a “single item face valid measure of moral conviction” (Skitka & Bauman, 2008, p. 36; see also Brandt & Wetherell, 2012 for similar operationalization) developed by Skitka et al. (2005). Respondents were asked to read a statement and indicate on a 5-point Likert-type scale the extent to which an issue is connected to their core beliefs on cheating. In this study, the item read, “How much are your feelings about academic cheating connected to your core moral beliefs?” The scale ranged from 1 (not at all) to 5 (very much). The selection of the single-item measure was based on Skitka’s early use of the single-item measure. In addition, Skitka and Morgan (2014) warn against measures that will potentially confound moral conviction with other constructs and suggest using measures where items

(1) explicitly assess moral content, (2) do not introduce confounds that capture either the things moral conviction should theoretically predict or other dimensions of attitude strength and (3) avoid using attitude strength indices or other variables as proxies for moral conviction. (p. 98)

Skitka (2010) explained that to avoid the above issues she and her colleagues “have generally used the single-item and face valid measure of moral conviction (p. 271). Recent methodological work has challenged the assumption that single-item measures are inherently unreliable, and Wanous and Hudy (2001) have reaffirmed that a minimum estimate of .70 is a reasonable choice. This is conservative estimate and due to attenuating statistical relationships, could make it more difficult to detect significant relationships. However, by being conservative, we are more confident that the relationships we are able to detect are accurate depictions of true relationships within the population. Post hoc, we conducted a second study where we collected data from a sample of 168 business students. To ensure the single-item moral conviction measure (Scale 1) was not significantly different from the four-item measure (Scale 2), we assessed the polychoric correlation between Scale 1 and Scale 2 (see Test of Hypothesized Structural Model section for results).

Moral Identity

Moral Identity was measured with a scale developed by Aquino and Reed (2002). Participants were provided a list of moral characteristics or traits (e.g., being caring, compassionate, fair, friendly, generous, helpful, hardworking, honest, and kind), and were asked to visualize a person with all of these characteristics, and then to imagine how that person would think, feel, and act. Respondents were then presented with 10 statements designed to determine how important it is for the respondent to be like the person they have visualized. Participants rated their agreement with the statements on a scale from 1 (strongly disagree) to 7 (strongly agree). Scale alpha was .80.

Moral Philosophy (Formalist and Utilitarian)²

Moral philosophies were measured using the character traits scale developed by Brady and Wheeler (1996) in their measure of ethical viewpoints. Seven items of character traits scales represented a utilitarian moral philosophy. These seven items were “innovative, resourceful, effective, influential, results-oriented, productive, and a winner.” Six items represented a formalist moral philosophy. These were “principled, dependable, trustworthy, honest, noted for integrity, and law abiding.” Responses ranged from 1 (not important to me at all) to 7 (very important to me). Scale alpha was .86 for the consequentialist items and .85 for the formalist items.

Moral Disengagement

We used the scale developed by Detert et al. (2008) that assesses eight mechanisms of moral disengagement: moral justification, euphemistic labeling, advantageous comparison, displacement of responsibility, diffusion of responsibility, distortion of consequences, attribution of blame, and dehumanization. Each mechanism was assessed using three items, resulting in a scale totaling 24 items. Respondents were asked to assess each statement on a 5-point scale ranging from 1 (strongly disagree) to 5 (strongly agree). Scale alpha was .89.

Unethical Decision Making

Unethical decision making was assessed using the “cheat–lie–steal” scale developed by Detert et al. (2008). With this scale, participants were asked to indicate how often they had engaged in 13 different cheating, lying, and stealing behaviors, indicating their response on a 5-point scale ranging from 1 (never) to 5 (many times). For our analysis, we used the six items specifically addressing cheating behaviors. Scale alpha was .88.

Analytical Procedures

Table 1 presents the means, standard deviations, and correlations of the variables used in the study. For all analyses, we relied on two statistical software packages. We used SPSS, version 20, to generate summary data for the variables, bivariate correlations, missing data analysis, and alpha estimates, and Mplus version 7 (Muthén & Muthén, 1998-2012) for confirmatory factor analysis (CFA) and structural equation modeling (SEM) in which we analyzed covariance matrices using maximum-likelihood estimation, and used the expectation maximization algorithm to address missing data (Roth, Switzer, & Switzer, 1999). We used four indices to assess how well our models fit the data (Hu & Bentler, 1999; Williams, Vandenberg, & Edwards, 2009): chi-square, root mean square error of approximation (RMSEA), comparative fit index (CFI), and standardized root mean square residual (SRMR). RMSEA values of .06 or less, in conjunction with CFI values of .95 or greater, were considered indicative of good fit (Hu & Bentler, 1999). Models were considered to have adequate fit if they met the less stringent, but traditionally accepted, values of .90 or greater for CFI, and values less than .08 for RMSEA. We also included SRMR because it has been identified as the index that is most sensitive to misspecified factor covariances or latent structures (Hu & Bentler, 1999). For SRMR, values less than.10 are acceptable, with values less than .08 being preferred.

Table 1.

Means, Standard Deviations, and Correlations.

	M	SD	1	2	3	4	5	6
1. Moral Conviction	3.72	1.02	—
2. Moral Identity	5.88	1.05	.28	.80
3. Consequentialism	5.90	0.90	.17	.15	.86
4. Formalism	6.17	0.78	.35	.37	.63	.85
5. Moral Disengagement	2.60	0.75	−.27	−.32	−.17	−.33	.89
6. Unethical Decision Making	2.14	0.90	−.22	−.19	−.17	−.35	.39	.88

Note. N = 311. Alphas appear in italics on the diagonal. All correlations are significant at p < .01.

Data Diagnostics

Prior to testing our hypothesized model, we conducted several diagnostic examinations related to our data. First, we examined the degree of missing data. Of the respondents, 278 (89%) provided complete responses, and another 28 (9%) missed only a single response among the 49 potential items. Overall, among 15,239 potential data points (i.e., 311 × 49), there were 102 missing entries (<1% missing), with the percentage of missing values for each item ranging from 0.3% to 1%. Using the missing data procedure in SPSS, we obtained a nonsignificant value for Little’s test (χ² = 1731.88, degrees of freedom [df] = 1643, p = .06), indicating the assumption of missing completely at random was met, and subsequently, that our data were suitable for imputation procedures such as expectation maximization (R. J. Little & Rubin, 2001). A fundamental assumption of SEM is a joint distribution of variables that is multivariate normal (Ullman, 2006; West, Finch, & Curran, 1995), thus, we examined multivariate normality in our data using the Mardia’s test which assesses multivariate kurtosis (Mardia, 1970). We obtained a Mardia’s PK value of 1.14, where values less than 3 are acceptable. Finally, due to the self-report nature of our data, we followed published recommendations regarding common method variance (CMV; Arbaugh & Hwang, 2012; Podsakoff, MacKenzie, Lee, & Podsakoff, 2003; Sullivan, Baruch, & Schepmyer, 2010). We used Harman’s (1976) single (one)-factor test, which involves testing an unrotated exploratory factor analysis on all the items and determining whether a single factor, or one factor that accounts for greater than 50% of extracted variance, emerges (Fuller, Simmering, Atinc, Atinc, & Babin, 2016; Podsakoff, MacKenzie, & Podsakoff, 2012). Using SPSS, we subjected the covariance matrix to principal axis factoring, and obtained favorable results indicating that CMV would not significantly bias our results. Specifically, with 22 factors having eigenvalues greater than 1 being extracted. With the first explaining just over 19% of the variance, this suggests that there is not a single method factor that accounts for the majority of the variance.

Confirmatory Factor Analysis

Our next step was to test the measurement model using CFA. There are well-documented problems associated with obtaining acceptable fit indices when the number of individual items approaches 50 or more (see Ding, Velicer, & Harlow, 1995; Kenny & McCoach, 2003), and item parceling has been advocated as a remedy for these problems (see Bandalos, 2002; Williams & O’Boyle, 2008). Because parceling tends to mask deviations from normality, multivariate normality should be verified prior to parceling (see Data Diagnostics section). Parceling is appropriate for continuous rather than coarsely categorized variables (Bandalos, 2002; T. D. Little, Cunningham, Shahar, & Widaman, 2002). All of our response scales used 5- or 7-point response anchors, and as such can be considered sufficiently continuous (Bollen & Barb, 1981).

We followed a random parceling strategy to reduce each scale to a smaller number of indicators. To ensure that each latent variable was independently just identified (see Williams & O’Boyle, 2008), we created three parcels for each factor. The two exceptions to this strategy were the measures for moral conviction and moral disengagement. For the single-item moral conviction variable, we set the lambda (loading) path to 1 and theta (error) term to 1 minus the square root of reliability times the variance of the observed measure. Because of the nature of this construct as a single-item (Wanous & Hudy, 2001), individual-level data (Wanous, Reichers, & Hudy, 1997), we used a conservative estimate of .70 for reliability. For moral disengagement, for each of the eight dimensions, we averaged the three items to obtain a single value for that dimension. We then used these eight values as indicators of the multidimensional moral disengagement latent variable.

The path diagram in Figure 2 depicts the results from our CFA. Fit for the model was good: χ² = 316.39 (df = 175), p < .001, CFI = .95, RMSEA = .05, SRMR = .05. Additionally, all item loadings were significant at p < .001. Latent covariances ranged from −.22 between unethical decision making and utilitarian moral philosophy to .72 between formal and utilitarian forms of moral philosophy. Given recent work indicating that these two philosophies jointly contribute to cognitive processing during moral dilemmas, this high covariance was not unexpected and should not be deemed problematic nor indicative of low discriminant validity (Conway & Gawronski, 2013).

Figure 2.

Measurement model: Confirmatory factor analysis.

Test of Hypothesized Structural Model

Following the favorable results obtained via CFA, we tested our hypothesized network of relationships. We chose SEM to test our hypotheses because it allows for simultaneous estimation of multiple direct and indirect paths of influence along with assessing model fit. We included a path of covariation between the disturbance (error) terms of formalist and utilitarian moral philosophies. The error terms depict the amount of variance not accounted for by the model, and the covariance path connecting these residual variances accounts for the possibility that the variables share a common source of variance not accounted for by our model (McDonald & Ho, 2002). Such paths have previously been used to account for the influence of a factor (e.g., personality trait, beliefs, or attitudes about a target object) that is not included in the model, but may jointly influence two other factors in the model. In relation to the dual-process theory, researchers have noted that these two forms of judgment do not exist orthogonally within individuals, it is rare for any individual to solely rely on one over the other, and instead both forms are jointly used in decision making (Conway & Gawronski, 2013; Hunt & Vitell, 1986, 1993). The degree to which they are jointly used varies across individuals, and extant research has not provided a list of personal characteristics to explain the degree of overlap. Allowing their residuals to covary is a methodological technique we used to recognize that unmeasured individual attributes account for residual shared variance between these two factors that is not explained by our predictor (moral conviction). The full structural model and standardized path estimates are shown in Figure 2. Fit was acceptable, χ² = 345.17 (df = 177), CFI = .94, RMSEA = .06, SRMR = .07, in the full structural model shown in Figure 3.

Figure 3.

Full hypothesized model.

Hypothesis 1 stated that the link between individuals’ moral conviction and subsequent unethical decision making is mediated by their use of moral disengagement mechanisms. As shown in Table 1, there is a significant, negative bivariate correlation (r = −.21, p < .001) between moral conviction and unethical decision making. In the full structural model, results show that moral conviction is not a significant predictor of unethical decision making (β = −.04, p = .68), thus indicating that the network of intervening variables fully accounts for the influence of moral conviction on unethical decision making. To establish a mediating effect, two other relationships must be verified. First, moral conviction must significantly predict moral disengagement, and moral disengagement must subsequently predict unethical decision making. As shown in the model, both portions of this relationship were significant. Moral conviction was negatively related to moral disengagement (β = −.19, p = .04), and moral disengagement was significantly predictive of unethical decision making (β = .33, p < .001). Notably, the significance of the second portion of this mediated effect supports Hypothesis 4 which proposed a significant impact of moral disengagement on unethical decision making. Furthermore, the indirect effect, calculated as the product of the coefficients from these two paths, was significant (−.06, p = .05).

Given the significant direct and intervening influence of moral disengagement on unethical decision making, Hypotheses 2 and 3 posited indirect mechanisms to explain additional ways in which moral conviction influences moral disengagement. Hypothesis 2 proposed that the relationship between moral conviction and use of moral disengagement mechanisms would be mediated by students’ ethical philosophies (i.e., formal and consequential). Hypothesis 3 predicted a significant mediated relationship between moral conviction and moral disengagement in which the intervening mechanism was moral identity.

Although moral conviction significantly predicted both moral philosophies, neither moral philosophy significantly predicted moral disengagement mechanisms. Thus, Hypothesis 2 was not supported. It is worth noting that the influence of formalism on moral disengagement was in the expected direction (β = −.22, p = .06), but fell just short of statistical significance at the level of p ≤ .05. Likewise, the indirect effect of moral conviction on moral disengagement conveyed through formalism was in the anticipated direction, but again was not significant (−.11, p = .07). Moving to Hypothesis 3, moral conviction significantly predicted moral identity (β = .44, p < .001), while moral identity significantly predicted moral disengagement (β = −.23, p = .002). Additionally, the indirect effect was significant (−.10, p = .004), thus this hypothesis was supported. When added to the direct effect of moral conviction on moral disengagement, these three indirect paths result in a total combined effect of −.40 (p < .001).

Although not explicitly hypothesized, the model contained three additional paths to more fully explicate the underlying theoretical mechanisms constituting the integrated model. Specifically, direct paths to unethical decision making from moral identity, utilitarian, and formalism were tested. Of these three, only the path from formalism (β = −.33, p = .01) was significant, and it also served to convey a significant indirect effect of moral conviction to unethical decision making (−.15, p = .01). Taking into account all of the potential intervening paths in the full model, the total effect of moral conviction on unethical decision making which comprises both direct and indirect effects was −.29 (p < .001). A post hoc model test in which all nonsignificant paths from the full model were removed resulted in the following fit: χ² = 353.76 (df = 181), p < 001, CFI = .94, RMSEA = .06, SRMR = .07. Thus, the more parsimonious reduced model can be substituted without significantly detracting from fit (Δχ² = 8.59, Δdf = 4).

We then conducted a post hoc study, where we collected a sample of 164, to determine the reliability of the one item measure of moral conviction. When assessing reliability estimate for a single-item scale researchers use the “reliability value from another study or similar measure, if available” (Petrescu, 2013, p. 114; MacKenzie, 2001). For Skitka et al.’s (2005) single-item moral conviction scale used in Study 1, we draw from Morgan’s (2011) findings where, across three studies, the reliability for a single-item moral conviction scale ranged from 0.93 to 0.99. In Study 2, we used Skitka and Morgan’s (2014) four-item MC scale due to added advantages multi-item scales offer. For instance, multi-item scales provide increased reliability and validity of measurements, assessment of nonconcrete or abstract constructs, and calculation of measurement error and Cronbach’s alpha (Petrescu, 2013). Scale alpha (.94) for the four-item moral conviction scale from our post hoc analysis falls within Morgan’s (2011) Cronbach alpha range. The authors conducted additional post hoc analyses (N = 168) to ensure a robust study.

Initially, we calculated the grand mean of the four-item moral conviction scale (Field, 2006). Next, we calculated Levene’s test for equality of variance, comparing the mean for the single-item moral conviction scale (Scale 1) with the grand mean of the four-item moral conviction scale (Scale 2). The statistic mean (M = 3.87 vs. 3.78) and standard deviation (SD = 1.06 vs. 1.01) for Scales 1 and 2, respectively, are not significantly different (see Table 1). We also assessed the difference in mean of the two scales (cf. Detert et al., 2008). Since the p value for Levene’s test of homogeneity of variance (p = .877) and the test for equality of means (p = .45) are not statistically significant, we assume the variance for Scales 1 and 2 are same (Bryman & Cramer, 2005). Last, we assessed the polychoric correlation between Scales 1 and 2 (Holgado-Tello, Chacón-Moscoso, Barbero-García, & Vila-Abad, 2010). The polychoric correlation coefficient of 0.859 between Scales 1 and 2 was statistically significant at p < .001. This indicates both scales are reliable measures of a student’s moral conviction. See Tables 2 and 3 for details of these tests.

Table 2.

Study 2 Group Statistics.

	Group	N	M	SD	SE M
Morcon	One-item scale	168	3.869	1.06417	0.0821
Morcon	Four-item scale	168	3.7827	1.01248	0.07811

Note. Morcon = Moral Conviction; SE = standard error.

Table 3.

Two-Way Independent Sample t Test and Leven’s Test of Homogeneity of Variance.

	Levene’s Test		t	df	t Test for equality of means			95% Confidence interval
	F	Sig.	t	df	Sig. (two-tailed)	Mean difference	SE difference	Lower	Upper
Morcon
Equal variances assumed	0.22	0.64	0.76	334	0.45	0.09	0.11	−0.14	0.31
Equal variances not assumed			0.76	333.2	0.45	0.09	0.11	−0.14	0.31

Note. Morcon = Moral Conviction; df = degrees of freedom; SE = standard error.

Discussion

Theoretical Contributions

The results of this study provide empirical support for the proposed integrated moral conviction theory of student cheating. We found that moral conviction directly drives student moral disengagement and subsequent unethical decision making about academic dishonesty among business students, not only as an antecedent of ethical philosophy and moral identity. In addition, this study provides empirical support for the dual-process model of the Hunt–Vitell’s positivist theory, as we find that when students face an ethical dilemma, formalism (i.e., students’ adherence to rules, as contained in the students’ handbook or course syllabus) but not utilitarian (i.e., students’ focus on the rightness or goodness of the consequences of actions) moral philosophies predicted that they would morally disengage—with a behavioral intention to avoid repercussions of their unethical acts. Moreover, this study advances the social cognitive conceptualization of moral identity as a predictor of moral disengagement and subsequent unethical decision making, as we found that it was influenced by the students’ moral conviction, or how each student individually frames or moralizes the issue of academic dishonesty.

Practical Implications for Teaching and Learning Practice

The empirical support found in this study for our proposed integrated moral conviction theory of student cheating has specific implications for teaching and learning practice. Specifically, the main practical challenge for educators is how to influence student moral conviction as the primary factor that attenuates student moral disengagement and subsequent unethical decision making about cheating. Teachers might employ students with strong moral convictions toward academic dishonesty to create an ethical environment that frames cheating as a moral issue through use of an honor code, procedures, and social influence.

Rahman, Hussean, and Esa (2016) found that the most effective method of eliminating cheating behaviors is through the adoption of a code of ethics. Additionally, a stated honor code (Jones, 2011) in the course materials will increase the effectiveness of an institutional code of ethics by relating it to the specific course. While evidence exists that having an honor code does reduce cheating behaviors (McCabe et al., 2002), making such code explicit through verbal announcement has a more profound effect on students’ propensity to cheat (Bing et al., 2012). We argue that through explicitly stating the honor code and relating the honor code to the integrity of the institution and the individual that students will be more likely to moralize the issue of breaking the honor code.

Teachers may also influence students’ tendency to moralize the issue of cheating through involving the students in the discussion of the honor code as well as the processes and procedures utilized when the teacher becomes aware of academic dishonesty. In a classroom experiment, Rosile (2007), who had caught students cheating and dismissed them from class, found evidence that instructors influence student framing of cheating behaviors through explicitly explaining the impact that cheating has on the different stakeholders. Such impact includes doubt being cast on all students’ performance, harm to reputation of school, and therefore students’ degrees, raised questions of all grades legitimacy, and further temptation to cheat in the future. After explaining the effect to stakeholders of cheating explaining why students were no longer in the class, Rosile opened discussion up to the remainder of the class. Just over 40% of the comments made in the open discussion were emotional in nature, which indicates a moral framing of the issue of academic cheating. Emotions, in the form of guilt and shame, have been shown to result to norm-violating behavior, such as cheating (van Kleef, Wanders, Stamkou, & Homan, 2015), however Rosile’s study provides evidence of students’ emotional engagement related to cheating and how such engagement can be developed through peer influence. Similarly, McCabe, Butterfield, and Treviño (2006) found that cheating is significantly related to peers’ cheating behaviors as well as inversely related to the perception that they will be reported by a peer. In this same study, they found no significant relationship between cheating and perceived severity of penalties, indicating that students are influenced primarily by other students’ behaviors and reactions with respect to cheating. We argue that teachers can use open discussion of the honor code and cheating to allow those students who do have strong moral convictions pertaining to academic dishonesty to influence how their peer students moralize the issue of cheating behaviors.

Study Limitations and Future Research Directions

Future studies investigating student unethical decision making about academic dishonesty should include the measure of moral conviction as a predictor of academic dishonesty, as we found that the relationship between moral conviction and moral disengagement is partially mediated by formalism on one hand (i.e., nonconsequentialist moral philosophy), and moral identity on the other hand. In particular, the identification of a direct negative relationship between student moral convictions’ and unethical decision making should motivate researchers to identify the antecedents of moral conviction, and thus fine-tune predictions of unethical decision making about student academic dishonesty.

As it is the case with most studies (e.g., Peng, Van Dyne, & Oh, 2014), the present study is not without limitations. The study’s first limitation is the cross-sectional nature of data used. Researchers (e.g., Podsakoff et al., 2003, 2012; Scandura & Williams, 2000; Sullivan et al., 2010) argue that cross-sectional data are susceptible to CMV. Thus, to assess the presence of CMV in the study, the authors apply Harman’s single-factor test. Results show that a small percentage variance is explained by CMV. In the future, researchers examining honest and dishonest behaviors among business students should use other empirical techniques to assess CMV such as Lindell and Whitney’s (2001) marker variable technique or Brown’s common factor analysis method (Fuller et al., 2016).

A second limitation from this study is that we incentivized participation in this study.³ Some researchers (e.g., R. E. Landrum & Chastain, 1995; Norcross, Dooley, & Stevenson, 1993) decry giving students extra credit as incentives for participating in educational research, due in part to the findings that “offering voluntary research for extra credit affects the generalizability of research conducted” (Padilla-Walker, Thompson, Zamboanga, & Schmersal, 2005, p. 150). Furthermore, there is a long-held belief among scholars (e.g., Sieber, 1999; Sieber & Saks, 1989) that by offering students the opportunity to earn extra class credit, researchers (i.e., professors) may be tacitly coercing students into participating in research. To mitigate this limitation, the authors carried out several procedures in line with existing ethical research standards. First, the authors ensured that the research procedures and incentive provided were approved by the university’s institutional review board (see also Padilla-Walker et al., 2005). This ensures that the authors did not coerce the participating business students nor did these students feel compelled to complete the survey reports. Second, to foster an educational environment conducive for learning and students’ interest in academic-based scientific research, students were given the opportunity to voluntarily participate in the research (Norcross et al., 1993). Third, not only did students express their desire to participate in the study without receiving extra credit the authors provided students alternative avenues to earn extra credit. Majority of the students preferred the survey option.

The last limitation of this study is that all data were collected only from a single source, that is, from undergraduate business students attending a midsized university located in the southeast area of the United States. By collecting data from a single source, we are constrained from concluding that causality exists among the various hypothesized relationships in the conceptual model (Campbell & Fiske, 1959). Furthermore, we acknowledge that this way of collecting data might strictly limit the generalizability that this study provides (Podsakoff & Organ, 1986). Future studies would better contribute to our theory testing if they replicated this study in different settings, particularly in settings with a broader geographical distribution. Findings similar to those in this study replicated in these settings would more appropriately inform educators concerned with the issue of academic dishonesty.

Conclusion

Moral conviction that academic dishonesty (i.e., in student terms, academic cheating) is unethical behavior is important to students’ moral compass, yet research of this construct has been neglected in past studies of teaching and learning. Internalizing this moral compass is particularly relevant to business students because they tend to give priority to expediency over equity and “stigmatize the goodness” (Giacalone & Promislo, 2013, p. 86) of academic honesty by framing it as a nonmoral pragmatic issue. This problem needs to be addressed by business educators preparing students to enter today’s knowledge-rich workplace because on being hired, students face an ongoing series of ethical dilemmas regarding how to reconcile: (a) organizational pressures to behave as organizational stewards that share knowledge with others and (b) inertia of their learned skill to behave as self-focused agents that prefer to hoard knowledge and “cut corners” when making moral choices.

By nurturing student moral conviction about academic (dis)honesty, we develop an essential aspect of ethics education in business schools. Moral conviction is an important factor in reducing student moral disengagement and subsequent unethical decision making, both directly and indirectly, through their moral identity and beliefs in moral rules. Business schools should work toward instilling the importance of nurturing moral conviction by promoting ethics of care and compassion in their instructional deliveries.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

References

Aquino

Freeman

Reed

II Lim

V. K.

Felps

(2009). Testing a social-cognitive model of moral behavior: The interactive influence of situations and moral identity centrality. Journal of Personality and Social Psychology, 97, 123-141.

Aquino

Reed

II . (2002). The self-importance of moral identity. Journal of Personality and Social Psychology, 83, 1423-1440.

Aquino

Reed

II Thau

Freeman

(2007). A grotesque and dark beauty: How moral identity and mechanisms of moral disengagement influence cognitive and emotional reactions to war. Journal of Experimental Social Psychology, 43, 385-392.

Arbaugh

J. B.

Hwang

(2012). Uses of multivariate analytical techniques in online and blended business education: An assessment of current practice and recommendations for future research. Journal of Management Education, 37, 229-260.

Ashby

(1950). Teleology and deontology in ethics. Journal of Philosophy, 47, 765-773.

Bandalos

D. L.

(2002). The effects of item parceling on goodness-of-fit and parameter estimate bias in structural equation modeling. Structural Equation Modeling, 9, 78-102.

Bandura

(1999). Moral disengagement in the perpetration of inhumanities. Personality and Social Psychology Review, 3, 193-209.

Bandura

(2001). Social cognitive theory: An agentic perspective. Annual Review of Psychology, 52, 1-26.

Bandura

Caprara

G. V.

Barbaranelli

Pastorelli

Regalia

(2001). Sociocognitive self-regulatory mechanisms governing transgressive behavior. Journal of Personality and Social Psychology, 80, 125-135.

10.

Berker

(2013). Epistemic teleology and the separateness of propositions. Philosophical Review, 122, 337-393.

11.

Bing

Davison

Vitell

Ammeter

Garner

Novicevic

(2012). An experimental investigation of an interactive model of academic cheating among business school students. Academy of Management Learning & Education, 11, 28-48.

12.

Blasi

(1984). Moral identity: Its role in moral functioning. In Kurtines

Gewirtz

(Eds.), Morality, moral behavior and moral development (pp. 128-139). New York, NY: Wiley.

13.

Bollen

K. A.

Barb

K. H.

(1981). Pearson’s R and coarsely categorized measures. American Sociological Review, 46, 232-239.

14.

Brady

F. N.

Wheeler

G. E.

(1996). An empirical study of ethical predispositions. Journal of Business Ethics, 15, 927-941.

15.

Brandt

M. J.

Wetherell

G. A.

(2012). What attitudes are moral attitudes? The case of attitude heritability. Social Psychological & Personality Science, 3, 172-179.

16.

Bryman

Cramer

(2005). Quantitative data analysis with SPSS 12 and 13: A guide for social scientists. New York, NY: Psychology Press.

17.

Buckley

M. R.

Fedor

D. B.

Marvin

D. S.

(1994). Ethical considerations in the recruiting process: A preliminary investigation and identification of research opportunities. Human Resource Management Review, 4, 35-50.

18.

Campbell

D. T.

Fiske

(1959). Convergent and discriminant validation by the multitrait-multimethod matrix. Psychological Bulletin, 56, 81-105.

19.

Chugh

Bazerman

M. H.

Banaji

M. R.

(2005). Bounded ethicality as a psychological barrier to recognizing conflicts of interest. In Moore

Loewenstein

Cain

Bazerman

M. H.

(Eds.), Conflicts of interest: Challenges and solutions in business, law, medicine, and public policy (pp. 74-95). Cambridge, England: Cambridge University Press.

20.

Conway

Gawronski

(2013). Deontological and utilitarian inclinations in moral decision making: A process dissociation approach. Journal of Personality and Social Psychology, 104, 216-235.

21.

Craft

J. L.

(2012). A review of the empirical ethical decision-making literature: 2004–2011. Journal of Business Ethics, 117, 221. doi:10.1007/s10551-012-1518-9

22.

Dalton

J. C.

(2015). Making moral mistakes: What ethical failure can teach students about life and leadership. New Directions for Student Leadership, 2015(146), 71-79.

23.

Dalton

J. C.

Crosby

P. C.

(2010). College peer culture: Taming the “monster within the gates.” Journal of College & Character, 11(4), 1-8.

24.

Dedeke

(2015). A cognitive–intuitionist model of Moral Judgement. Journal of Business Ethics, 126, 437-457.

25.

Detert

J. R.

Treviño

L. K.

Sweitzer

V. L.

(2008). Moral disengagement in ethical decision making: A study of antecedents and outcomes. Journal of Applied Psychology, 93, 374-391.

26.

Ding

Velicer

W. F.

Harlow

L. L.

(1995). Effects of estimation methods, number of indicators per factor, and improper solutions on structural equation modeling fit indices. Structural Equation Modeling, 2, 119-144.

27.

Duffy

M. K.

Aquino

Tepper

B. J.

O’Leary-Kelly

(2005, August). Moral disengagement and social identification: When does being similar result in harm doing? Paper presented at the Academy of Management Meeting, Honolulu, Hawaii.

28.

Ferrell

O. C.

Crittenden

V. L.

Ferrell

Crittenden

W. F.

(2013). Theoretical development in ethical marketing decision making. AMS Review, 3, 51-60.

29.

Field

(2006). Discovering statistics using SPSS (2nd ed.). Thousand Oaks, CA: Sage.

30.

Fuller

C. M.

Simmering

M. J.

Atinc

Babin

B. J.

(2016). Common methods variance detection in business research. Journal of Business Research, 69, 3192-3198.

31.

Gholami

Tirri

(2012). Caring teaching as a moral practice: An exploratory study on perceived dimensions of caring teaching. Education Research International, 2012. doi:10.1155/2012/954274

32.

Giacalone

R. A.

Promislo

M. D.

(2013). Broken when entering: The stigmatization of goodness and business ethics education. Academy of Management Learning & Education, 12, 86-101.

33.

Greene

J. D.

(2007). The secret joke of Kant’s soul. In Sinnott-Armstrong

(Ed.), Moral psychology, Vol. 3: The neuroscience of morality: Emotion, brain disorders, and development (pp. 35-80). Cambridge: MIT Press.

34.

Greene

J. D.

Morelli

S. A.

Lowenberg

Nystrom

L. E.

Cohen

J. D.

(2008). Cognitive load selectively interferes with utilitarian moral judgment. Cognition, 107, 1144-1154.

35.

Greene

J. D.

Nystrom

L. E.

Engell

A. D.

Darley

J. M.

Cohen

J. D.

(2004). The neural bases of cognitive conflict and control in moral judgment. Neuron, 44, 389-400.

36.

Greene

J. D.

Sommerville

R. B.

Nystrom

L. E.

Darly

J. M.

Cohen

J. D.

(2001). An fMRI investigation of emotional engagement in moral judgment. Science, 293, 2105-2108.

37.

Haidt

(2001). The emotional dog and its rational tail: A social intuitionist approach to moral judgment. Psychological Review, 108, 814-834.

38.

Haidt

Rosenberg

Hom

(2003). Differentiating diversities: Moral diversity is not like other kinds. Journal of Applied Social Psychology, 33, 1-36.

39.

Hardy

(2006). Identity, reasoning, and emotion: An empirical comparison of three sources of moral motivation. Motivation and Emotion, 30, 205-213.

40.

Harman

H. H.

(1976). Modern factor analysis. Chicago, IL: University of Chicago Press.

41.

Hart

Atkins

Ford

(1998). Urban America as a context for the development of moral identity in adolescence. Journal of Social Issues, 54, 513-530.

42.

Holgado-Tello

F. P.

Chacón-Moscoso

Barbero-García

Vila-Abad

(2010). Polychoric versus Pearson correlations in exploratory and confirmatory factor analysis of ordinal variables. Quality & Quantity, 44, 153-166.

43.

Bentler

P. M.

(1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6, 1-55.

44.

Huang

Y. H.

(2004). Is symmetrical communication ethical and effective? Journal of Business Ethics, 53, 333-352.

45.

Hunt

S. D.

Vitell

(1986). A general theory of marketing ethics. Journal of Macromarketing, 6(1), 5-16.

46.

Hunt

S. D.

Vitell

S. J.

(1993). The general theory of marketing ethics: A retrospective and revision. In Smith

N. C.

Quelch

J. A.

(Eds.), Ethics in marketing (pp. 775-784). Homewood, IL: Irwin.

47.

Hunt

S. D.

Vitell

S. J.

(2006). The general theory of marketing ethics: A revision and three questions. Journal of Macromarketing, 26, 143-153.

48.

Jones

D. L.

(2011). Academic dishonesty: Are more students cheating? Business Communication Quarterly, 74, 141-150.

49.

Kenny

D. A.

McCoach

D. B.

(2003). Effect of the number of variables on measure of fit in structural equation modeling. Structural Equation Modeling, 10, 333-351.

50.

Kern

M. C.

Chugh

(2009). Bounded ethicality: The perils of loss framing. Psychological Science, 20, 378-384.

51.

Kohlberg

(1973). Stages and aging in moral development: Some speculations. The Gerontologist, 13, 497-502.

52.

Landrum

N. E.

(2001). My friend Morgan: An exercise in ethics. Journal of Management Education, 25, 606-616.

53.

Landrum

R. E.

Chastain

(1995). Experiment spot-checks: A method for assessing the educational value of undergraduate participation in research. IRB: Ethics & Human Research, 17(4), 4-6.

54.

Lindell

M. K.

Whitney

D. J.

(2001). Accounting for common method variance in cross-sectional research designs. Journal of applied psychology, 86, 114-121.

55.

Little

R. J. A.

Rubin

D. B.

(2002). Statistical analysis with missing data (2nd ed.). Hoboken, NJ: Wiley.

56.

Little

T. D.

Cunningham

W. A.

Shahar

Widaman

K. F.

(2002). To parcel or not to parcel: Exploring the question, weighing the merits. Structural Equation Modeling, 9, 151-173.

57.

Longenecker

Ludwig

(1990). Ethical dilemmas in performance appraisal revisited. Journal of Business Ethics, 9, 961-969.

58.

MacKenzie

S. B.

(2001). Opportunities for improving consumer research through latent variable structural equation modeling. Journal of Consumer Research, 28, 159-166.

59.

Mardia

K. V.

(1970). Measures of multivariate skewness and kurtosis with applications. Biometrika, 57, 519-530.

60.

McCabe

D. L.

Butterfield

K. D.

Treviño

L. K.

(2006). Academic dishonesty in graduate business programs: Prevalence, causes, and proposed action. Academy of Management Learning & Education, 5, 294-305.

61.

McCabe

D. L.

Trevino

L. K.

(1995). Cheating among business students: A challenge for business leaders and educators. Journal of Management Education, 19, 205-218.

62.

McCabe

D. L.

Trevino

L. K.

Butterfield

K. D.

(2002). Honor codes and other contextual influences on academic integrity: A replication and extension to modified honor code settings. Research in higher Education, 43, 357-378.

63.

McDonald

R. P.

M. H. R.

(2002). Principles and practice in reporting structural equation analyses. Psychological Methods, 7, 64-82.

64.

McFerran

Aquino

Duffy

M. K.

(2010). Individual predictors of the commitment to integrity: The role of personality and moral identity. Business Ethics Quarterly, 20, 35-56.

65.

Moore

(2015). Moral disengagement. Current Opinion in Psychology, 6, 199-204.

66.

Morgan

G. S.

(2011). Toward a model of morally motivated behavior: Investigating mediators of the moral conviction-action link (Doctoral dissertation). University of Chicago, Chicago, IL.

67.

Murphy

P. E.

Laczniak

G. R.

(1981). Marketing ethics: A review with implications for managers, educators and researchers. In Enis

B. M.

Kenneth

J. R.

(Eds.), Review of marketing (pp. 251-266). Chicago, IL: American Marketing Association.

68.

Muthén

L. K.

Muthén

B. O.

(1998-2012). Mplus user’s guide (7th ed.). Los Angeles, CA: Muthén & Muthén.

69.

Norcross

J. C.

Dooley

H. S.

Stevenson

J. F.

(1993). Faculty use and justification of extra credit: No middle ground? Teaching of Psychology, 20, 240-242.

70.

Novicevic

M. M.

Buckley

M. R.

Harvey

M. G.

Fung

(2008). Self-evaluation bias of social comparisons in ethical decision making: The impact of accountability. Journal of Applied Social Psychology, 38, 1061-1091.

71.

Nucci

L. P.

(2001). Education in the moral domain. Cambridge, England: Cambridge University Press.

72.

Nye

Plunkett

(2015). Non-Consequentialism Demystified. Philosophers’ Imprint, 15, 1-28.

73.

Padilla-Walker

L. M.

Thompson

R. A.

Zamboanga

B. L.

Schmersal

L. A.

(2005). Extra credit as incentive for voluntary research participation. Teaching of Psychology, 32, 150-153.

74.

Peng

A. C.

Van Dyne

(2014). The influence of motivational cultural intelligence on cultural effectiveness based on study abroad: The moderating role of participant’s cultural identity. Journal of Management Education, 39, 572-596.

75.

Petrescu

(2013). Marketing research using single-item indicators in structural equation models. Journal of Marketing Analytics, 1, 99-117.

76.

Podsakoff

P. M.

MacKenzie

S. B.

Lee

J. Y.

Podsakoff

N. P.

(2003). Common method biases in behavioral research: A critical review of the literature and recommended remedies. Journal of Applied Psychology, 88, 879-903.

77.

Podsakoff

P. M.

MacKenzie

S. B.

Podsakoff

N. P.

(2012). Sources of method bias in social science research and recommendations on how to control it. Annual Review of Psychology, 63, 539-569.

78.

Podsakoff

P. M.

Organ

D. W.

(1986). Self-reports in organizational research: Problems and prospects. Journal of Management, 12, 531-544.

79.

Rahman

N. A. A.

Hussein

Esa

M. M.

(2016). Academic Dishonesty Among Business Students: Cheating Acts and Proposed Ways to Reduce Cheating Behavior. In 7th International Conference on University Learning and Teaching (InCULT 2014) Proceedings (pp. 211-219). Singapore: Springer.

80.

Reed

(2002). Social identity as a useful perspective for self-concept based consumer research. Psychology & Marketing, 19, 235-266.

81.

Rendtorff

J. D.

(2015). Case studies, ethics, philosophy, and liberal learning for the management profession. Journal of Management Education, 39, 36-55.

82.

Rest

J. R.

(1986). Moral development: Advances in research and theory. New York, NY: Praeger.

83.

Richardson

C. B.

Mulvey

K. L.

Killen

(2012). Extending social domain theory with a process-based account of moral judgments. Human Development, 55, 4-25.

84.

Rosile

G. A.

(2007). Cheating: Making it a teachable moment. Journal of Management Education, 31, 582-613.

85.

Ross

W. D.

Stratton-Lake

(2002). The right and the good. Oxford, England: Oxford University Press.

86.

Roth

P. L.

Switzer

F. S.

Switzer

D. M.

(1999). Missing data in multiple item scales: A Monte Carlo analysis of missing data techniques. Organizational Research Methods, 2, 211-232.

87.

Scandura

T. A.

Williams

E. A.

(2000). Research methodology in management: Current practices, trends and implications for future research. Academy of Management Journal, 43, 1246-1264.

88.

Schwartz

M. S.

(2015). Ethical decision-making theory: An integrated approach. Journal of Business Ethics, 139, 755-776.

89.

Sieber

J. E.

(1999). What makes a subject pool (un)ethical? In Chastain

Landrum

R. E.

(Eds.), Protecting human subjects: Departmental subject pools and institutional review boards (pp. 43-64). Washington, DC: American Psychological Association.

90.

Sieber

J. E.

Saks

M. J.

(1989). A census of subject pool characteristics and policies. American Psychologist, 44, 1053-1061.

91.

Skitka

L. J.

(2002). Do the means always justify the ends or do the ends sometimes justify the means? A value protection model of justice. Personality and Social Psychology Bulletin, 28, 452-461.

92.

Skitka

L. J.

(2010). The psychology of moral conviction. Social & Personality Psychology Compass, 4, 267-281.

93.

Skitka

L. J.

Bauman

C. W.

(2008). Moral conviction and political engagement. Political Psychology, 29, 29-54.

94.

Skitka

L. J.

Bauman

C. W.

Lytle

B. L.

(2009). The limits of legitimacy: Moral and religious convictions as constraints on deference to authority. Journal of Personality and Social Psychology, 97, 567-578.

95.

Skitka

L. J.

Bauman

C. W.

Sargis

E. G.

(2005). Moral conviction: Another contributor to attitude strength or something more? Journal of Personality and Social Psychology, 88, 895-917.

96.

Skitka

L. J.

Morgan

G. S.

(2014). The social and political implications of moral conviction. Political Psychology, 35, 95-110.

97.

Skitka

L. J.

Washburn

A. N.

Carsel

T. S.

(2015). The psychological foundations and consequences of moral conviction. Current Opinion in Psychology, 6, 41-44.

98.

Sullivan

S. E.

Baruch

Schepmyer

(2010). The why, what, and how of reviewer education: A human capital approach. Journal of Management Education, 34, 393-429.

99.

Turiel

(2002). The culture of morality: Social development, context, and conflict. Cambridge, England: Cambridge University Press.

100.

Turiel

(2006). Thought, emotions, and social interactional processes in moral development. In Melanie

Judith

S. G.

(Eds.), Handbook of moral development (pp. 7-35). Mahwah, NJ: Lawrence Erlbaum.

101.

Ullman

J. B.

(2006). Structural equation modeling: Reviewing the basics and moving forward. Journal of Personality Assessment, 87, 35-50.

102.

van Kleef

G. A.

Wanders

Stamkou

Homan

A. C

. (2015). The social dynamics of breaking the rules: Antecedents and consequences of norm-violating behavior. Current Opinion in Psychology, 6, 25-31.

103.

Vitell

Bing

Davison

Ammeter

Garner

Novicevic

(2009). Religiosity and moral identity: The mediating role of self-control. Journal of Business Ethics, 88, 601-613.

104.

Vitell

S. J.

Hunt

S. D.

(2015). The general theory of marketing ethics: The consumer ethics and intentions issues. In Nill

(Ed.), Handbook on ethics and marketing (pp. 15-37). Northampton, MA: Edward Elgar.

105.

Wanous

J. P.

Hudy

M. J.

(2001). Single-item reliability: A replication and extension. Organizational Research Methods, 4, 361-375.

106.

Wanous

J. P.

Reichers

A. E.

Hudy

M. J.

(1997). Overall job satisfaction: How good are single-item measures? Journal of Applied Psychology, 82, 247-252.

107.

Weaver

G. R.

(2006). Virtue in organizations: Moral identity as a foundation for moral agency. Organizational Studies, 27, 341-368.

108.

West

S. G.

Finch

J. F.

Curran

P. J.

(1995). Structural equation models with nonnormal variables: Problems and remedies. In Hoyle

R. H.

(Ed.), Structural equation modeling: Concepts, issues, and applications (pp. 56-75). Thousand Oaks, CA: Sage.

109.

Williams

L. J.

O’Boyle

E. H.

(2008). Measurement models for linking latent variables with indicators: A review of human resource management research using parcels. Human Resource Management Review, 18, 233-242.

110.

Williams

L. J.

Vandenberg

R. J.

Edwards

J. R.

(2009). Structural equation modeling in management research: A guide for improved analysis. Academy of Management Annals, 3, 543-604.

111.

Wiltermuth

S. S.

Bennett

V. M.

Pierce

(2013). Doing as they would do: How the perceived ethical preferences of third-party beneficiaries impact ethical decision-making. Organizational Behavior and Human Decision Processes, 122, 280-290.

112.

Woodbine

G. F.

Amirthalingam

(2013). Cheat and you lose! Don’t cheat and you lose! Reflections and analysis of accounting student data. Journal of Academic Ethics, 11, 311-327.

Integrated Moral Conviction Theory of Student Cheating: An Empirical Test

Abstract

Keywords

Introduction

Theoretical Foundations

Integration of Moral Philosophies on Moral Judgment 1

Integrated Moral Conviction Theory

Integrating Moral Conviction Theory and Hunt–Vitell Theory

Integrating Moral Conviction Theory and the Social Cognitive Conceptualization of Moral Identity

Method

Data Collection and Sample Characteristics

Measures

Moral Conviction

Moral Identity

Moral Philosophy (Formalist and Utilitarian) 2

Moral Disengagement

Unethical Decision Making

Analytical Procedures

Data Diagnostics

Confirmatory Factor Analysis

Test of Hypothesized Structural Model

Discussion

Theoretical Contributions

Practical Implications for Teaching and Learning Practice

Study Limitations and Future Research Directions

Conclusion

Footnotes

Declaration of Conflicting Interests

Funding

Notes

References

Integration of Moral Philosophies on Moral Judgment¹

Moral Philosophy (Formalist and Utilitarian)²