This article appeared in a journal published by Elsevier. The attached copy is furnished to the author for internal non-commercial research and education use, including for instruction at the authors institution and sharing with colleagues. Other uses, including reproduction and distribution, or selling or licensing copies, or posting to personal, institutional or third party websites are prohibited. In most cases authors are permitted to post their version of the article (e.g. in Word or Tex form) to their personal website or institutional repository. Authors requiring further information regarding Elsevier’s archiving and manuscript policies are encouraged to visit: http://www.elsevier.com/authorsrights
Author's personal copy Journal of Economic Behavior & Organization 93 (2013) 400–407
Contents lists available at ScienceDirect
Journal of Economic Behavior & Organization journal homepage: www.elsevier.com/locate/jebo
Moral firmness Shaul Shalvi ∗,1 , David Leiser ∗,1 Ben-Gurion University of the Negev, Israel
a r t i c l e
i n f o
Article history: Received 3 March 2013 Accepted 8 March 2013 Available online 18 March 2013 Keywords: Judgment Decision making Ethics Religion Morality Temptation Dishonesty Flexibility Firmness Behavioral ethics Self-control
a b s t r a c t Firm moral judgment deems dishonest acts as categorically wrong, and considers any self-serving justification for them as further dishonesty. People, however, commonly use self-serving justifications in order to feel honest even as they behave dishonestly, indicating reduced moral firmness. We test variation in moral firmness by comparing a sample of religious and secular female students. Arguably, religious people’s upbringing and ongoing exposure to moral admonitions promote a firm moral approach which should translate into firmer moral judgments in adulthood. Results of a moral judgment experiment supported this proposition: Religious students judged lies more harshly than secular students, and were less influenced by the availability of self-serving justifications. A moral behavior experiment provided support to the notion that moral firmness in judgment may translate to moral firmness in behavior: whereas modest amount of lying was found among the secular students, no evidence for lying was observed among the religious student. Overall, we provide strong evidence for firm moral judgment among female religious students, and weaker evidence for firm moral behavior. We discuss the relation between firm moral judgment and behavior. © 2013 Elsevier B.V. All rights reserved.
1. Introduction People often refrain from lying, even when lying yields personal profit. One reason people avoid lying is their fear of being detected and punished (Becker, 1968). However, even when their anonymity is secured, people restrict how much they lie. This means that to some extent, people are genuinely lie averse (Gneezy, 2005; Lundquist et al., 2009; Fischbacher and Heusi, 2008; Cappelen et al., 2013), and that beyond their desire to appear honest to others, people have an intrinsic desire to being and feeling honest. But do some people hold a firmer moral approach regarding self-serving dishonesty than others? Do people holding such firm moral approach lie less? Put differently, do people who profess deontological views about dishonesty (i.e., lying is always wrong, no matter what self-serving excuses can be cited), lie less than people with more flexible views about dishonesty? These are the questions the current paper addresses. 1.1. Moral firmness In tempting situations, people restrict the amount of their lies allowing them to boost profit unethically while maintaining feeling honest (Mazar et al., 2008; see also Ariely, 2012; Bazerman and Tenbrunsel, 2011; Ploner and Regner, 2013). People seem to stretch the truth (Schweitzer and Hsee, 2002), exactly to the extent they manage to justify their lies (Shalvi et al., 2011a, 2012; Gino and Ariely, 2012; Wiltermuth, 2011). Shalvi et al. (2011a) engaged participants in a die rolling task in
∗ Corresponding authors at: Department of Psychology, Ben Gurion University of the Negev, Beer Sheva 84105, Israel. Tel.: +972 86472049. E-mail addresses:
[email protected] (S. Shalvi),
[email protected] (D. Leiser). 1 Both authors contributed equally. 0167-2681/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jebo.2013.03.014
Author's personal copy S. Shalvi, D. Leiser / Journal of Economic Behavior & Organization 93 (2013) 400–407
401
which they were asked to roll anonymously, report the outcome and gain money according to their reports (Fischbacher and Heusi, 2008). They found that participants who were instructed to roll multiple times but report the outcome of the first roll only, lied more than those who were (ceteris paribus) instructed to roll only once. The idea is that the high numbers appearing on the non-relevant for pay rolls (second roll, third roll, etc.) makes people feel justified to use these numbers to determine the amount of their lies. People seem to benefit from having information, even completely private and otherwise irrelevant information, that makes their lies feel justifiable, allowing them to profit dishonestly while feeling honest. Is all people’s morality equally flexible? Can we expect some people to hold firmer moral views on dishonesty, and judge any dishonest act as wrong? Will people holding such firm view consider justifiable information irrelevant and judge all lies with equal severity? These questions concern the moral stance. Assuming such people are found, do people professing firm moral views also lie less? Various factors may contribute to the development of firm moral beliefs. Such factors may include the cultural norms within one’s close family, friends or society at large. They may relate to strict upbringing, education or exposure to media sources advocating such approach. This suggests that people from all walks of life may vary on their level of moral firmness in their judgment as well as behavior. The current work is focused on (1) assessing variation in people’s moral firmness and (2) its relation to moral behavior. We thus focused on one sub-group likely to hold relatively high moral firmness, namely, religious female Jewish students, from whom lying is forbidden by their religious beliefs. These individuals are raised and socialized under the explicit and frequent treatment of morality and temptation in stern terms such as “Thou shalt not steal” (Exodus 20: 12) and “I hate and detest falsehood, your Torah I love” (Psalms 119: 163). The relationship between religious reminders and morality has been studied in recent years. Being exposed to implicit religious reminders such as unscrambling sentences with religiousness-related words led people to demonstrate higher levels of self-control (Rounding et al., 2012), generosity (Shariff and Norenzayan, 2007), and helping behaviors (Pichon et al., 2007). More explicit reminders, such as asking participants to recall the Ten Commandments led people to lie less compared to participants who recalled the last ten books they read (Mazar et al., 2008). Explaining this pattern of behavior, some have argued that exposure to religious reminders make people feel that some divine authority is watching them (Shariff and Norenzayan, 2007; and also Haley and Fessler, 2005; Bateson et al., 2006) and thus feel uncomfortable with behaving immorally or in an asocial way. Others have argued that religious reminders serve as moral cues making it more difficult to lie while maintaining an honest self-concept (Mazar et al., 2008). Recent work (Fischbacher and Utikal, 2011) provided further evidence that religious people, twelve nuns in this case, even lied in a disadvantageous way (over-reported lower numbers in the dice task) when this allowed them to signal to others that they would not steal. Given that religious reminders are associated with moral behavior and that religious people are raised in settings advocating moral firmness, we expect to find higher firmness in religious people’s moral judgment compared to secular people, which may additionally translate to more honest behavior. We test this prediction by comparing samples of students from two different campuses of the same Israeli university – one campus attended solely by religious students and another campus attended by “secular” students. The two groups do not divide into atheists vs. believers. The group we call “religious” lives in a segregated environment, and follow a way of life where religion is very dominant, while their education stressed religious ideals. The “secular” group is more varied, but is united by their being culturally part of the larger society; religion plays a modest role in their lives, if any. In a moral judgment experiment we test (1) whether religious students hold firmer moral judgments when it comes to assessing self-serving dishonest acts; specifically, we test whether these religious students will be less influenced than the secular ones by a context that may make the unethical behavior more acceptable. (2) We further test whether religious students predict that they would be less tempted to lie compared to secular students and whether this expectation varies as a function of that context. In a moral behavior experiment, we test whether firm moral judgment, expected among religious students, will also translate to resolutely moral behavior, that is, whether religious students will lie less compared to secular students.
2. Procedure overview Participants were all undergraduates attending an Israeli University. We recruited students from two different campuses, the main university campus and a campus attended by a special track followed by religious students. This track is designed for Jewish religious female students, encouraging them to acquire higher education to increase their chances of entering the job market. The unique track was ideal for the purpose of the current study for two main reasons. First, all students were religious Jews, a sub-group we predicted would display higher levels of moral firmness. Second, the students regularly participated in psychological experiments as part of their study requirements. As a result, compared to sampling religious participants in settings in which conducting experiments is unnatural (e.g., in religious higher education institutes or in proximity to a praying place), our sample included students in a natural setting where they were unlikely to feel or behave as representatives of their sub-group (religious Jews). Rather, our sampling method ensured that participants were acting upon their intrinsic moral beliefs.
Author's personal copy 402
S. Shalvi, D. Leiser / Journal of Economic Behavior & Organization 93 (2013) 400–407
2.1. Participants We sampled students to participate in two experiments: The first experiment included the die-under-cup behavioral task (Shalvi et al., 2011a,b, 2012, a variation of a task by Fischbacher and Heusi, 2008). Participants were 126 university students, of which 65 were following the special religious BA track and were recruited and tested at the religious campus, while the remaining 61 were following the regular BA track and were recruited and tested at the main university campus. The second experiment was conducted at least a week after the first and included hypothetical die-under-cup scenarios asking a sub set of these students (N = 87; of which 45 religious) to judge how unethical were different combinations of die rolls and reports conducted by other students. In both experiments participants were individually tested and received 10 Israeli Shekels (ILS) for participation. This sampling strategy (i.e., using a subset of the same participants in the judgment experiment) was used as we wished to obtain as many students as possible for the behavioral task. The sample of the behavioral experiment included the entire religious student group studying in their first year of studies, which precluded our ability to find additional religious students during the same academic year. In the interest of homogeneity, we followed the same sampling procedure for the secular students as well. We begin by describing the experiment about moral judgment, about which we had firmer predictions. We then report on the experiment on moral behavior, allowing us to assess whether moral firmness in judgment translates to moral firmness in behavior. We relied on aggregate level data to test our predictions. As the task we employed did not involve deception that would allow tracing individuals’ behavior (see similarly, Batson et al., 1997; Jiang, 2012; Schurr et al., 2012), we did not attempt (nor could we) to link individuals’ judgment to their behavior. 3. Moral judgment experiment To test variation in moral firmness in judgment, we employed hypothetical scenarios in which we described the dieunder-cup task and asked participants to evaluate reports of different die roll combinations allegedly performed by other participants (see Shalvi et al., 2011a, Experiments 3 and 4). Specifically, participants read a description of other people performing the die-under-cup task, and received information about: the outcome of the first roll (determining pay), the second roll (non-relevant for pay), and the reported outcome. We provided information about only one extra roll (the second), which is a sufficient manipulation to enable self-serving justifications (see Shalvi et al., 2011a, Exp. 3). Each participant evaluated eighteen combinations in total and was asked to answer two questions regarding each of the combinations: (1) to what extent was the report a lie? (1 = not at all to 6 = very much) and (2) what is the likelihood that, if you were in the same situation, you would report the same? (1 = low to 6 = high). Participants evaluated three types of combinations: (1) honest reports in which the person rolling reported the outcome observed on the first roll (e.g., 1st roll = 5, 2nd roll = 2, report = 5); (2) “justifiable” lies in which the other person reported rolling a higher outcome than the outcome observed on the first roll determining pay, but equal to the outcome appearing on the second roll which may serve as a justification for the lie (e.g., 1st roll = 2, 2nd roll = 5, report = 5); (3) unjustifiable lies in which the other person reported rolling a higher outcome than the outcome observed on the first roll determining pay, but critically not equal to the outcome appearing on the second roll, thus having no justification for the lie (e.g., 1st roll = 2, 2nd roll = 1, report = 5). We were able to test our predictions regarding moral firmness by comparing the mean ratings of the extent to which participants judged different reports to be lies, and the likelihood that they would “do the same”. Specifically, we predicted that the secular sample would replicate previous results showing that justifiable lies are evaluated as less of a lie compared to unjustifiable lies. Further, we predicted this effect to be attenuated in the religious sample, where we expected higher moral firmness and harsher judgments of all type of lies, with no distinction between “justifiable” vs. unjustifiable lies. Regarding the “secular” group, we expected them to replicate past findings and to assess a higher likelihood of behaving unethically and lying when evaluating justified, compared to unjustified lies (Shalvi et al., 2011a,b, Exp. 4). Further, regarding the “religious” participants, we explored whether their (expected) firm moral judgment would also lead them to claim a low likelihood to “do the same” and lie, independent of the type of lie evaluated. Or alternatively, whether religious participants, like the secular ones, would predict higher likelihood of being tempted to lie when evaluating “justifiable” lies. 3.1. Task Participants evaluated 6 honest reports, 4 justifiable lies, and 4 unjustifiable lies. For both justifiable and unjustifiable lies we held the magnitude of the lie constant at either 1 or 3 (i.e., reporting the first roll outcome +1 or +3). We included 6 honest reports (rather than 4) to avoid suspicion among participants regarding the low proportion of honesty in the total number of combinations they were evaluating. To further conceal the purpose of the study, we interspersed four filler combinations with different magnitudes of lies and involving lies lower than the outcome of the first roll. In total, participants thus evaluated 18 combinations in a 2 (Religion: Secular vs. Religious) × 3 (Report: Honest vs. Justifiable lie vs. Unjustifiable lie) design with the former as between-subjects and the latter as within-subjects factors (see Appendix A). Combinations were organized into six different random orders and participants were randomly assigned to complete one of these six versions of the task. The order of combinations had no effect on any of the reported results and is not discussed any further.
Author's personal copy S. Shalvi, D. Leiser / Journal of Economic Behavior & Organization 93 (2013) 400–407
403
6.4 6.2 6 5.8 Jusfied 5.6
Unjusfied
5.4 5.2 5 Religious
Secular
Fig. 1. Mean (standard error) of moral judgments as a function of justifications and religion (moral judgment experiment).
3.2. Results We computed indices for the extent to which participants judged honest reports, “justifiable” lies and unjustifiable lies to be “lies”. Similarly, we computed three indices for the extent to which participants reported that, had they been in the same situation as the one that they were evaluating, they were likely to report the same value reported by the person they were evaluating. Four participants (2 religious and 2 secular) were excluded from the sample as their responses were ±3SD or more from the sample mean on the moral judgment indices. 3.2.1. Does this count as a lie? A 2 (Religion: Secular vs. Religious) by 3 (Report: Honest vs. Justifiable vs. Unjustifiable) repeated measures ANOVA with the latter as within-subject factor, supported our predictions revealing a main effect for religion, F (1, 81) = 5.78, p < .02, 2 = .067, a main effect for type of report, F (2, 81) = 2785.33, p < .001, 2 = .972, and most importantly a significant interaction, F (2, 81) = 5.08, p < .01, 2 = .059. Simple effect analyses revealed that while religious (M = 1.04, SD = .18) and secular (M = 1.07, SD = .23) students did not differ in their judgment of honest reports, F (1, 81) = .39, p = .53, they did differ in their judgments of dishonest reports. Specifically, religious student evaluated justifiable lies more harshly (M = 5.72, SD = .60) than secular students (M = 5.30, SD = .92), F (1, 81) = 6.15, p < .02, 2 = .071. Further, religious student also evaluated unjustifiable lies more harshly (M = 5.87, SD = .33) than secular students (M = 5.63, SD = .65), F (1, 81) = 4.64, p < .05, 2 = .054. Furthermore, as can be seen in Fig. 1, replicating previous results (Shalvi et al., 2011a, Exp. 3) secular students greatly differed on their moral judgment of justifiable vs. unjustifiable lies, F (1, 39) = 15.84, p < .001, 2 = .289. Supporting out prediction, religious students differed in their moral judgment to a lesser extent, F (1, 39) = 5.07, p < .05, 2 = .108, as confirmed by the specific interaction contrast comparing justifiable vs. unjustifiable lies by religion was also significant at, F (1, 81) = 3.02, p < .05, 2 = .054 (one-tailed). 3.2.2. Moral firmness We further categorized participants into two groups: morally firm or morally flexible according to their tendency to dichotomize: in judging the different die roll reports across all combinations, did they use only the end points of the scale (i.e., firm, using only ‘1 [not at all a lie] for all honest combinations and ‘6 [very much a lie] for all dishonest combinations) or did they use other values available on the spectrum between 1 and 6 (i.e., flexible, using also other values). Providing further support to the moral firmness prediction, we observed more morally firm participants in the religious sample (30 of 43; 69.8%) compared with the secular sample (20 of 40; 50%), 2 (1) = 3.38, p = .066 (marginal). 3.2.3. Would I yield to temptation? We analyzed the answer to the question “would you do the same?” as a 2 (Religion: Secular vs. Religious) by 3 (Report: Honest vs. Justifiable vs. Unjustifiable) repeated measures ANOVA with the latter as within subject factor. This revealed a main effect for religion, F (1, 81) = 8.14, p < .01, 2 = .091, a main effect for type of report, F (2, 81) = 2044.95, p < .001, 2 = .962 and a significant interaction, F (2, 81) = 7.51, p < .01, 2 = .085. Simple effect analyses revealed that while religious (M = 5.96, SD = .18) and secular (M = 5.85, SD = .39) students understandably did not differ in their assessment of the likelihood they would “do the same” when it came to honest reports, F (1, 81) = 2.66, p = .11, they did differ in their assessment of likelihood to “do the same” when it came to lying. Specifically, evaluating justifiable lies, religious student reported that they were less likely to “do the same” and report such lies (M = 1.26, SD = .71) than secular students (M = 1.70, SD = .94), F (1, 81) = 5.81, p < .02, 2 = .067. Similarly, evaluating unjustifiable lies religious student reported that they were less likely to “do the same” and report such lies (M = 1.08, SD = .25) than secular students (M = 1.52, SD = .73), F (1, 81) = 14.19, p < .001, 2 = .149. Interestingly, both secular and religious student reported a higher likelihood of lying when evaluating justifiable compared to unjustifiable
Author's personal copy 404
S. Shalvi, D. Leiser / Journal of Economic Behavior & Organization 93 (2013) 400–407
100 90 80 70 60 50 40 30 20 10 0 1 to 3
4 to 6
Religious sample
1 to 3
4 to 6 Secular sample
Behavioral experiment: Reported die rolls Fig. 2. Percentage of participants reporting low (1–3) versus high (4–6) outcomes as a function of religiousness sample (moral behavior experiment). Note: The dashed line represents the honest distribution predicted by chance (50% for rolling a number ≤3 and 50% for rolling a number ≥4). The error bars represent the 95% confidence interval around the proportion.
lies. For secular student this effect was significant at, F (1, 39) = 4.94, p < .05, 2 = .112, while for the religious students the effect was significant at, F (1, 42) = 5.95, p < .02, 2 = .124. 4. Moral behavior experiment Results of the moral judgment experiment supported our prediction and showed that, when it comes to assessing selfserving dishonest acts “religious” students hold firmer moral judgments than “secular” students. Specifically, while for both secular and religious students we replicated previous results showing that “justified” lies are evaluated more leniently compared to unjustified lies, religious students were less influenced by the context of the lie, that is, whether it was “justifiable” or not. Interestingly, we found that both secular and religious students estimated that if confronted with such tempting situations, providing them with a justification to serve self-interest by lying, they will be more likely to lie that when no such justification is available. The moral behavior experiment to which we now turn was designed to test whether the firmer moral judgment held by religious students translates to firmer moral behavior or alternatively, whether their behavior will be in-line with their own assessment of the likelihood to yield to temptation, and lie to similar extents as secular students. 4.1. Task and procedure Participants rolled a die and earned money according to what they reported rolling. We used a simple payoff structure, participants earned the number they reported rolling times three in Israeli Shekels (1ILS ∼$.25). That is, participants who reported rolling 1 dot received 3ILS, those reporting 2 received 6ILS, etc. We ensured the privacy of the rolls by placing a paper cup upside-down over the die. We asked participants to shake the cup, roll the die, and peek through a small hole at the bottom of the cup to see the result. Participants were further instructed to roll the die two more times to verify the die’s fairness, but to report the outcome of the first roll only to determine their pay. The procedure guaranteed participants that we could not observe their first roll, even after they left the experiment (see a discussion on the pros and cons of such method in Gneezy et al., 2013). After completing the task, participants were debriefed and thanked for their participation. 4.2. Results While participants’ rolls were truly completely anonymous, we detected lying by comparing the observed distribution to the expected distribution of fair rolls (Fischbacher and Heusi, 2008). The distribution of reported outcomes was found to differ from the uniform distribution expected from a fair die roll, 2 (5) = 10.95, p = .052. Participants reported the higher (and more profitable) values appearing on the die (4, 5, and 6) more often (59.5%; 75 of 126) than the lower values (1, 2, and 3; 40.5%; 51 of 126), 2 (1) = 4.20, p < .05. A Mann–Whitney test was used to assess whether the religious and secular samples differed in the extent of dishonesty. The analysis revealed no significant difference between the samples, Z = −1.63, p = .103. Thus, while compared to the secular students, the religious students in the first experiment displayed higher moral firmness in judgment this did not translate to moral firmness in behavior. Notwithstanding the lack of significant difference between the two samples, we analyzed each distribution separately to assess the extent to which each student sample lied. This more fine-tuned approach revealed no statistical evidence of lying in the religious sample, as the distribution of reported outcomes did not differ from the uniform distribution expected from a fair die roll, 2 (5) = 7.65, p = .176. In contrast, in the secular sample, the distribution of reported outcomes did indicate lying, as outcomes differed from the uniform distribution, 2 (5) = 12.87, p < .05, see Fig. 2.
Author's personal copy S. Shalvi, D. Leiser / Journal of Economic Behavior & Organization 93 (2013) 400–407
405
5. General discussion Results of two experiments revealed that religious female students held firmer moral judgments compared to secular female students. Specifically, while both religious and secular students were influenced by contextual information that may be exploited to justify unethical behavior, judging dishonest acts leniently in such context compared to a context in which justifications were not available, religious students demonstrated less sensitivity to the contextual information and judged dishonesty categorically harsher. Interestingly, when asked to predict how likely they were to behave unethically and lie if they were to be placed in such tempting situations, both religious and secular students alike, expected to be tempted when justifications for self-serving dishonesty were available. The moral behavior experiment provided weak indication that religious students’ firmness in moral judgment translates into firmness in moral behavior. While no statistically significant differences in lying were obtained between the religious and secular samples, modest amount of lying was observed in the secular sample, but none in the religious sample. 5.1. Moral firmness in judgment and behavior The setting used in the current research was intentionally designed to be minimalistic. Students evaluated combinations of die rolls and reports representing different type of lies and honest scenarios. Using such minimalistic settings ensured that the observed difference on moral judgment between the different combinations can be attributed to the studied variables – namely, the availability of information (i.e., extra rolls) making some lies more “justified” than others, and the moderator, religiousness. In real life obviously, justifications may take various shapes and forms. For example, when determining if and how much to lie, people care about the consequences of their behavior not only for the self but also for others (Gneezy, 2005). They also anticipate feeling guilty about it and take this into consideration (Battigalli et al., 2013). Recent work revealed that people lie to benefit a good cause (e.g., a charity devoted to cancer research) even when provided with no additional contextual information to justify their immoral behavior (Lewis et al., 2012). Other work has revealed that when people feel empathy toward another person, they are more likely to lie and boost this other’s profit (Gino and Pierce, 2010). More generally, when another person benefits from one’s lies, people are more likely to lie, profiting both the other and themselves as they feel it is more legitimate (Gino et al., 2013; Wiltermuth, 2011). Given that the religious group in our study live in close and relatively segregated communities compared to secular people, could it be that they would legitimize bending moral rules in order to benefit people from their own community? Alternatively, would their firm commitment lead them to condemn lying even in such a context? And how would their moral judgment translate into behavior? Would settings allowing people to benefit their loved ones lead religious people to more lying, thus increasing the gap between their moral judgment and behavior? These questions seem promising avenues for future research to explore. The obtained results suggest that people vary in the way that contextual information influences their moral judgment. Haidt (2001, 2007) suggests that people form their moral judgment quickly and only later consider ways to reason and justify their original moral intuition. In settings involving a moral conflict, such as whether to push one person to her death to save the lives of several others, time pressure leads people to base their behavior on the emotional rejection of the act of pushing a person to her death (Suter and Hertwig, 2011). When facing a moral dilemma they need to judge, people’s automatic reaction leads them to distinguish between right and wrong, and only later modulate this judgment to the extent that reasoning leads them to adopt a different course of action. Temptation complicates matters. In tempting and anonymous settings, people’s immediate choice of behavior seems to be to serve their self-interest even when this involves cheating. When provided with more time, people adjust their behavior to what social and moral norms require and restrict the amount of their lies or avoid lying altogether (Gunia et al., 2012; Shalvi et al., 2012; but see Zhong, 2011). Do people holding firm moral beliefs, such as those identified in the current work, have the same automatic tendencies to serve their self-interest but reason their way out of behaving dishonestly by dint of deliberate moral scrutiny? Alternatively, could it be that some people’s socialization leads them to automatically behave in ways acceptable by society and will thus avoid cheating even when under time pressure? These questions remain for now, unanswered. 5.2. Limitations While one factor in our design was manipulated within participants (type of report: honest vs. justified lie vs. unjustified lie), allowing us to draw causal inferences about the role of justification in shaping ethical judgment, the other factor (religiousness) obviously could not. This cross-sectional feature of the design prevents making causal claims regarding religiousness leading people to more or less firmness in the moral judgment and/or behavior. Despite this limitation, the purpose of the current research was to demonstrate variation in moral firmness in judgment and behavior. Others before us (e.g., Rounding et al., 2012; Shariff and Norenzayan, 2007; Pichon et al., 2007; Mazar et al., 2008) demonstrated that religious reminders evoke moral behavior. In our study, we sampled two naturalistic groups varying on the level of religiousness. We predicted and obtained results supporting the idea that these groups would vary on their moral firmness in judgment, and explored whether differences in firm moral judgment also translate to firmness in behavior. To allow drawing causal conclusions about how religiousness shapes people’s firmness however, longitudinal studies should be conducted (e.g., by assessing moral judgment and behavior at different ages). Such studies may be especially fruitful if they assess how firmness emerges not only in one of the aspects, judgment or behavior, but in the interplay of both factors.
Author's personal copy 406
S. Shalvi, D. Leiser / Journal of Economic Behavior & Organization 93 (2013) 400–407
Another limitation concerns the specific groups we used, specifically, our religious sample. This group is composed of intensely religious female students, its members are highly committed to the very demanding requirements of the Jewish religion, are exposed from their youngest age to extensive religious teachings and admonitions, and also live in homogeneous communities. While this made this group highly suitable for testing our hypothesis, it remains to be seen to what extent the attitudes and behavior found in our sample generalize to less extremely religious groups, to other faiths, as well as to men. 5.3. Conclusion People vary in the firmness of their moral judgment: some judge all lies as categorically wrong, others adopt a more flexible approach exploiting contextual information when forming a moral judgment. We focused on two groups, expected to differ in moral firmness, religious and secular students. Clear experimental evidence indicates that religious students are firmer in their moral judgment. Weaker evidence suggests that such firm moral judgment translates to firmness in behavior. While secular students lied to modest degrees, we found no evidence that the religious students lied at all. Studying the development and interplay between firmness in moral judgment and behavior seems a promising avenue worth pursuing. Acknowledgment We thank Daniel Heiman for his superb on-going research assistance in collecting the data for this study. Appendix A. Die roll combinations evaluated in the moral judgment experiment Unjustified combinations: 1st roll = 3, 2nd roll = 2, report = 6 1st roll = 3, 2nd roll = 2, report = 4 1st roll = 2, 2nd roll = 1, report = 3 1st roll = 2, 2nd roll = 1, report = 5 Justified combinations: 1st roll = 3, 2nd roll = 4, report = 4 1st roll = 2, 2nd roll = 3, report = 3 1st roll = 3, 2nd roll = 6, report = 6 1st roll = 2, 2nd roll = 5, report = 5 Honest combinations: 1st roll = 5, 2nd roll = 3, report = 5 1st roll = 2, 2nd roll = 1, report = 2 1st roll = 4, 2nd roll = 3, report = 4 1st roll = 5, 2nd roll = 2, report = 5 1st roll = 6, 2nd roll = 2, report = 6 1st roll = 3, 2nd roll = 2, report = 3 Fillers: 1st roll = 1, 2nd roll = 2, report = 6 1st roll = 1, 2nd roll = 1, report = 5 1st roll = 6, 2nd roll = 1, report = 3 1st roll = 4, 2nd roll = 1, report = 5
References Ariely, D., 2012. The honest truth about dishonesty: how we lie to everyone, especially ourselves. HarperCollins, New York. Bateson, M., Nettle, D., Roberts, G., 2006. Cues of being watched enhance cooperation in a real-world setting. Biological Letters 2, 412–414. Battigalli, P., Charness, G., Dufwenberg, M., 2013. Deception: the role of guilt. Journal of Economic Behavior & Organization 93, 227–232. Batson, C.D., Kobrynowicz, D., Dinnerstein, J.L., Kampf, H.C., Wilson, A.D., 1997. In a very different voice: unmasking moral hypocrisy. Journal of Personality and Social Psychology 72, 1335–1348. Bazerman, M.H., Tenbrunsel, A.E., 2011. Blind Spots: Why We Fail to Do What’s Right and What to Do about It. Princeton University Press, Princeton. Becker, G.S., 1968. Crime and punishment: an economic approach. Journal of Political Economy 76, 169. Cappelen, A.W., Sorensen, E.O., Tungodden, B., 2013. When do we lie? Journal of Economic Behavior & Organization 93, 258–265. Fischbacher, U., Heusi, F., 2008. Lies in Disguise, An Experimental Study on Cheating. TWI Working Paper 40, Thurgau Institute of Economics, University of Konstanz. Fischbacher, U., Utikal, V., 2011. Disadvantageous Lies. TWI Research Paper 71, Thurgau Institute of Economics, University of Konstanz. Gino, F., Ariely, D., 2012. The darks side of creativity: original thinkers can be more dishonest. Journal of Personality and Social Psychology 102, 445–459. Gino, F., Ayal, S., Ariely, D., 2013. Self-serving altruism? The lure of unethical action that benefits others. Journal of Economic Behavior & Organization 93, 285–292. Gino, F., Pierce, L., 2010. Lying to level the playing field: why people may dishonestly help or hurt others to create equity. Journal of Business Ethics 95, 89–103. Gneezy, U., 2005. Deception: the role of consequences. The American Economic Review 95, 384–394. Gneezy, U., Rockenbach, B., Serra-Garcia, M., 2013. Measuring lying aversion. Journal of Economic Behavior & Organization 93, 293–300. Gunia, B.C., Wang, L., Huang, L., Wang, J., Murnighan, J.K., 2012. Contemplation and conversation: subtle influences on moral decision making. Academy of Management Journal 55, 13–33. Haidt, J., 2001. The emotional dog and its rational tail: a social intuitionist approach to moral judgment. Psychological Review 108, 814–834. Haidt, J., 2007. The new synthesis in moral psychology. Science 316, 998–1002. Haley, K.J., Fessler, D.M.T., 2005. Nobody’s watching? Subtle cues affect generosity in an anonymous economic game. Evolution and Human Behavior 26, 245–256.
Author's personal copy S. Shalvi, D. Leiser / Journal of Economic Behavior & Organization 93 (2013) 400–407
407
Jiang, T., 2012. The Mind Game: Invisible Cheating and Inferable Intentions. TIBER working paper. Lewis, A., Bardis, A., Flint, C., Mason, C., Smith, N., Tickle, C., Zinser, J., 2012. Drawing the line somewhere: an experimental study of moral compromise. Journal of Economic Psychology 33, 718–725. Lundquist, T., Ellingson, T., Gribbe, E., Johannesson, M., 2009. The aversion to lying. Journal of Economic Behavior & Organization 70, 81–92. Mazar, N., Amir, O., Ariely, D., 2008. The dishonesty of honest people: a theory of self-concept maintenance. Journal of Marketing Research 45, 633–644. Pichon, I., Boccato, G., Sarogiou, V., 2007. Nonconscious influences of religion on prosociality: a priming study. European Journal of Social Psychology 36, 1032–1045. Ploner, M., Regner, T., 2013. Self-image and moral balancing. Journal of Economic Behavior & Organization 93, 374–383. Rounding, K., Lee, A., Jacobson, J.A., Ji, L.J., 2012. Religion replenishes self-control. Psychological Science, http://dx.doi.org/10.1177/0956797611431987. Schweitzer, M.E., Hsee, C.K., 2002. Stretching the truth: elastic justification and motivated communication of uncertain information. The Journal of Risk and Uncertainty 25, 185–201. Schurr, A., Ritov, I., Kareev, Y., Avrahami, J., 2012. Is that the answer you had in mind? The effect of perspective on unethical behavior. Judgment and Decision Making 7 (6), 679–688. Shalvi, S., Dana, J., Handgraaf, M.J.J., De Dreu, C.K.W., 2011a. Justified ethicality: observing desired counterfactuals modifies ethical perceptions and behavior. Organizational Behavior and Human Decision Processes 115, 181–190. Shalvi, S., Eldar, O., Bereby-Meyer, Y., 2012. Honesty requires time (and lack of justifications). Psychological Science 23, 1264–1270. Shalvi, S., Handgraaf, M.J.J., De Dreu, C.K.W., 2011b. Ethical maneuvering: why people avoid both major and minor lies. British Journal of Management 22, s16–s27. Shariff, A., Norenzayan, A., 2007. God is watching you: priming god concepts increases prosocial behavior in an anonymous economic game. Psychological Science 18, 803–809. Suter, R.S., Hertwig, R., 2011. Time and moral judgment. Cognition 119, 454–458. Wiltermuth, S.S., 2011. Cheating more when the spoils are split. Organizational Behavior and Human Decision Processes 115, 157–168. Zhong, C.B., 2011. The ethical dangers of deliberative decision making. Administrative Science Quarterly 56, 1–25.
