Unpackaging gender differences in justifying morally debatable behaviors around the world: The role of personal religiosity and society’s socialization priorities for its children

Abstract

Women generally report greater religiosity and justify morally debatable behaviors less than men. This study examined if personal religiosity mediates the relationship of gender and justification of different types of morally debatable behaviors across societies with diverse religious heritages. We also explored how a society’s endorsement of preferred qualities in the socialization of children would moderate the links between personal religiosity and justification of morally debatable types of behavior. Using the World Values Survey Wave 7 data (47 societies; 66,992 respondents), we identified three types of justifiable behaviors, namely, behaviors threatening human life and family values, dishonest-illegal, and interpersonally violent behaviors. Controlling for age and education, women scored higher in personal religiosity and justified dishonest-illegal and interpersonally violent behaviors less than men, but behaviors threatening human life and family values more than men. Personal religiosity only partially mediated the link between gender and the justification of behaviors threatening human life and family values, indicating that factors other than personal religiosity account for gender differences in justifying types of morally defensible behavior. The linkage strengths of personal religiosity to behaviors threatening human life and family values and dishonest-illegal behaviors were moderated by societal endorsement of preferred qualities in the socialization of children. Thus, both gender and the socialization of a society’s human capital impact on how different types of morally debatable behaviors are justified around the world. Possible explanations for these results are offered and future research directions suggested.

Keywords

Gender morally debatable behaviors religiosity societal endorsement of preferred qualities in the socialization of children

A contemporary society inherits the legacy of its heritage religion as that religious tradition is shaped by eco-cultural factors acting across its historical development. This legacy will be used by members of that society to judge the behavior of members inside and outside their own group as right and proper. An individual’s personal religiosity then indicates how strongly he or she has incorporated their cultural group’s religious legacy of beliefs, values, and practices as transmitted through the socialization experiences characterizing their society. Given each society’s religious heritage and contemporary circumstances, these socialization experiences will differ across societies. This developmental context characterizing each society will shape the moral code that its members apply to their daily living with others.

Gender, religiosity, and justification of morally debatable behaviors

Personal religiosity has been measured in various ways by psychologists researching in different societies, each with its different religious heritage. As has been consistently demonstrated through the results of these many studies, “The greater religiosity of women must be one of the oldest, and clearest, findings in the psychology of religion, and should be considered one of the universals in human religious behavior” (Beit-Hallahmi, 2004, p. 120).

A prominent gender gap also exists in endorsing moral codes, as has recently been shown by Atari et al. (2020) using moral foundations theory (MFT) (Graham et al., 2013). MFT conceptualizes moral concerns as underpinned by the five basic values of Care, Fairness, Purity, Loyalty, and Authority. Antari et al. showed that women consistently scored higher on the MFT components of moral judgment tapping Care, Fairness, and Purity as reported by 336,691 citizens of 67 countries. Similarly, a consistent gender difference was revealed in a multi-cultural study of the willingness to justify some of the “morally debatable behaviors” originally proposed by Harding and Phillips (1986). Vauclair and Fischer (2011) used representative samples of more than 266,000 citizens from 84 nations and territories of the World Values Survey, showing that women were judging “dishonest-illegal” behaviors, like avoiding fares on public transport or accepting a bribe, less strictly than men, but judging “personal-sexual” behaviors, like abortion or suicide, more strictly than men. Gender differences in moral codes thus appear to be robust around the world regardless of how they may be measured. But, how to explain these gender differences?

Integrating results across many, mostly monocultural studies, Saroglou (2019) pointed out that one feature of greater female religiosity around the world is women’s tendency to evaluate various human behaviors differently than men. In their extensive multi-cultural study, Vauclair and Fischer (2011) had found that women reported consistently higher levels of religiosity than men. These researchers did not, however, establish links between an individual’s personal religiosity and their justification of personal-sexual and dishonest-illegal behaviors. Given the pancultural role of religion in guiding moral beliefs and the consistent gender differences in measures of morality, we propose that the gender differences in these two types of morally debatable behaviors may be unpacked through the mediating agency of personal religiosity. Thus, we hypothesize that the following:

Hypothesis 1 (H1). The link between gender and the justification for different types of morally debatable behaviors is mediated by personal religiosity (i.e. women are less likely to justify these behaviors due to their higher levels of personal religiosity).

The impact of the endorsement of preferred qualities in the socialization of children

All social systems face the requirement to induct their new members and train them to function adequately within that system’s established patterns for coordinating their members’ behaviors to promote the group’s survival and flourishing in its current ecological and inter-group circumstances (Berry, 2017). Whatever the type of ongoing human group, those new members who adapt to its membership demands will replace older members in time. So, as children, they must be socialized to comply with and function within the society’s cultural system developed across its history (Bronfenbrenner & Ceci, 1994).

Group norms for right conduct in playing roles within the society will crystallize, often in the form of laws and requirements for membership, becoming its moral code. Members of the society will be socialized to valorize those character qualities in its members which are regarded as conducing to compliance with that desired code of conduct, and hence promote the continuing viability of their social group (Miller et al., 2019).

At the level of societies, those valorized qualities of character in new members, that is, children, have been measured by the World Values Survey. Using Wave 5 data, Bond and Lun (2014) identified and validated two dimensions of the endorsement of preferred qualities in the socialization of children. These two dimensions were identified as Self-Directedness versus Other-Directedness and Civility versus Practicality. Self-Directedness versus Other-Directedness reflects a societal preference for training character qualities like responsibility and independence versus those of obedience and religious faith; Civility versus Practicality, by character qualities like respect and tolerance of others versus hard work. Fifty-five nations and territories from the Wave 5 data set were then arrayed across these two dimensions, each location specifying the societal context under which citizens emphasized different priorities for developing their society’s human capital.

We hypothesize that socializing children to become self-directed in their orientation toward social life as opposed to other-directed would reflect a code for right behavior that accords greater freedom to the individual rather than to established authority and group dictates in evaluating matters pertaining to an individual’s life. We thus anticipate that citizens of self-directed societies would justify personal-sexual behaviors more strongly than would citizens of other-directed societies:

Hypothesis 2 (H2). The justification of personal-sexual behaviors will be stronger in self-directed societies than in other-directed societies.

Within self-directed societies, religious persons have been socialized to express their personal moral code for personal-sexual behaviors derived from their society’s religious heritage, whatever that heritage may be. Given that personal-sexual behaviors have traditionally been sanctioned in all religious heritages (Saroglou, 2019), religious persons would be more reluctant than non-religious persons to justify these personal-sexual behaviors. However, within other-directed societies, members are more uniformly compliant with society’s moral code regardless of their personal religiosity. Thus, the difference between religious and non-religious persons in justifying personal-sexual behaviors would be weaker:

Hypothesis 2a (H2a). The greater a society’s self-directedness, the greater the difference between its religious and non-religious members in their justification of personal-sexual behaviors.

However, socializing children to become civil as opposed to becoming practical in their orientation toward social life would reflect a code for right behavior geared toward protecting the community against its exploitation by members more focused on their self-seeking for material advantage. Practical societies, however, provide fewer constraints against such acquisitiveness (Bond & Lun, 2014). We thus anticipate that citizens of practical societies would be less likely to justify dishonest-illegal behaviors than citizens of civil societies:

Hypothesis 3 (H3). The justification of dishonest-illegal behaviors will be weaker in civil societies than in practical societies.

Within civil societies, both religious and non-religious persons have been socialized to condemn dishonest-illegal behaviors because they disturb community life and its egalitarian dynamic (Bond & Lun, 2014). All religious heritages endorse a similar dynamic of in-group protecting and promoting that is compatible with condemning dishonest-illegal behavior (Saroglou, 2019). So, the difference between religious and non-religious persons in condemning dishonest-illegal behaviors should be smaller in civil societies; in practical societies, members are socialized to accept self-seeking as less anti-normative. Thus, the difference between religious and non-religious persons in justifying dishonest-illegal behaviors would be greater in practical societies:

Hypothesis 3a (H3a). The greater a society’s civility, the less the difference between its religious and non-religious members in their justification of dishonest-illegal behaviors.

So, we argue that these two dimensions of the endorsement of preferred qualities in the socialization of children will moderate the links of religiosity to the justification of these two types of morally debatable behavior, personal-sexual and dishonest-illegal. How strongly morally debatable behaviors in one society would be justified by its religious and non-religious members would be different and moderated depending on how that society socialized their human capital for right acting as acceptable members of their society. Figure 1 presents the conceptual model tested in this study.

Figure 1.

The multilevel, moderated mediation model tested in this study.

Method

Participants

We used the latest wave of the World Values Survey, Wave 7 (WVS 7; Haerpfer et al., 2020), released in July 2020, to test our research hypotheses. The data were collected from representative samples of citizens from 49 countries/territories between 2017 and 2020 and are openly accessible online (http://www.worldvaluessurvey.org/wversus.jsp). We removed two countries (Guatemala and Tunisia) from our possible sample due to their non-inclusion of the question about the endorsement of preferred qualities in the socialization of children, as explained in the following section. The final data set consisted of 66,992 participants (M_age = 42.77 years, SD = 16.26 years; 53% female participants, 47% male participants) from 47 countries/territories.

Measures

Level-1 (individual-level) variables

Religiosity—measured as the relative centrality of religion

Participants were asked to report how important each of the six major domains (i.e. family, friends, leisure time, politics, work, and religion) was in their lives using a five-point Likert-type scale (1 = not at all important, 5 = very important). We assessed the importance of religion for the person relative to the summed importance the person attached to these six major domains of living, terming each person’s ratio as “the relative centrality of religion.”

Justification of morally debatable behaviors

Participants were asked to indicate how justifiable the following behaviors are to them on a 10-point scale (1 = never justifiable; 10 = always justifiable): (1) claiming government benefits to which you are not entitled; (2) avoiding a fare on public transport; (3) stealing property; (4) cheating on taxes; (5) someone accepting a bribe in the course of their duties; (6) homosexuality; (7) prostitution; abortion; (8) divorce; (9) sex before marriage; (10) suicide; (11) euthanasia; (12) for a man to beat his wife; (13) parents beating children; (14) violence against other people; (15) terrorism as a political, ideological or religious mean; (16) having casual sex; (17) political violence; and (18) death penalty.

The following behavior items were deleted for various reasons: (1) We deleted the item, “death penalty” because it is not personal behavior. (2) Some countries did not collect data for certain items. In order to include all 47 countries or territory samples in the factor analysis, we made a trade-off by deleting items lacking data in at least one country or territory. The following items were accordingly deleted: “avoiding a fare on public transport” (data lacking from Jordan and Lebanon); “homosexuality” (data lacking from Egypt and Tajikistan); “prostitution” (data lacking from Egypt, Iran, Iraq, Jordan, and Lebanon); “sex before marriage” (data lacking from Egypt and Iraq); “terrorism as a political, ideological or religious mean” (data lacking from Turkey); “having casual sex” (data lacking from Egypt, Iran, Iraq, and Jordan); and “political violence” (data lacking from Turkey). It will be noted that these countries are Muslim-heritage societies, so we deleted these seven items to ensure that these Muslim societies, often overlooked in multi-cultural studies, were included in our multi-cultural study; the generalizability of our results would thereby be improved.

With the remaining 11 items, we used multilevel factor analysis (MLFA) to identify different types of morally debatable behaviors at the individual-level. MLFA can simultaneously model within- and between-country level effects by using individual-level data only, without the need of creating separate measures to capture the level-specific data variabilities. MLFA outputs factor loadings for both within- and between-country levels, but only the former was used in this study. Detailed information about MLFA is documented in Dunn et al. (2015).

We tested four MLFA models varying from 1 to 4 factors at the within-country level, setting the between-country level, factor number consistent with the within-country level ones. Following previous work (Dunn et al., 2015), we determined the overall model fit using the following criteria: (1) comparative fit index (CFI) value greater than 0.95; (2) root mean square error of approximation (RMSEA) and standardized root mean squared residual (SRMR) (at both within and between levels) values below 0.10. Note that a non-significant Chi-square test result was deemed unnecessary for evaluating model fit in this study due to its tendency to be significant given such a large sample size (Dunn et al., 2015).

Results indicated that both the three- and four-factor solutions showed adequate model fit; we chose the three-factor solution for theoretical parsimony and conceptual overlap with Vauclair and Fischer’s (2011) results. The geomin rotated factor loadings of the three-factor solution are presented in Table 1. We named the factor grouped by four items including “claiming government benefits to which you are not entitled,” “stealing property,” “cheating on taxes,” and “someone accepting a bribe in the course of their duties” as justification of dishonest-illegal behaviors; the factor grouped by three items including “for a man to beat his wife,” “parents beating children,” and “violence against other people,” as justification of interpersonally violent behaviors; the factor grouped by four items including “abortion,” “divorce,” “suicide,” and “euthanasia,” as justification of behaviors threatening human life and family values. The factor scores for these three types of justifications were calculated by weighted averaging of item scores within each factor according to their item loadings, with a higher score indicating a higher willingness to justify that type of morally debatable behavior. Cronbach’s alphas of these three factors are 0.762, 0.717, and 0.792, respectively.

Table 1.

Individual-level factor structure of justification of morally questionable behaviors.

Items	Justification of dishonest-illegal behaviors	Justification of interpersonally violent behaviors	Justification of behaviors threatening human life and family values
Justifiable: Claiming government benefits to which you are not entitled	0.463
Justifiable: Stealing property	0.673
Justifiable: Cheating on taxes	0.717
Justifiable: Someone accepting a bribe in the course of their duties	0.675
Justifiable: Abortion			0.714
Justifiable: Divorce			0.652
Justifiable: Suicide			0.430
Justifiable: Euthanasia			0.547
Justifiable: For a man to beat his wife		0.816
Justifiable: Parents beating children		0.657
Justifiable: Violence against other people		0.736

Only factors with factor loading ⩾0.40 were presented; model-fit information: Chi-square = 3054.77; df = 50; p < 0.001; CFI = 0.987; RMSEA = 0.030; SRMR_within = 0.015; SRMR_between = 0.030. Demographics: Participants’ age, sex (0 = male; 1 = female), and education (1 = lower, 2 = middle, and 3 = higher level) were also recorded from the WVS 7 data set.

Level-2 (national-level) variables

The endorsement of preferred qualities in the socialization of children

Participants were asked to select up to five qualities of character that they regarded as important for children to learn at home from this list of 11 qualities: (1) good manners, (2) independence, (3) hard work, (4) feeling of responsibility, (5) imagination, (6) tolerance and respect for other people, (7) thrift, saving money and things, (8) determination, perseverance, (9) religious faith, (10) not being selfish, and (11) obedience.

We extracted factors for the endorsement of preferred qualities in the socialization of children following the previous practice of Bond and Lun (2014). We calculated each country’s/territory’s percentages of respondents selecting each of the 11 qualities as input for factor analysis. At this stage and as mentioned above, the data of Guatemala and Tunisia were excluded because more than 25% of participants in each country did not follow the question’s instructions (i.e. participants selected either no quality or more than five qualities).

Considering the conceptual frameworks suggested by different factor solutions, prior research, and item-factor loadings (⩾0.40) (Table 2), we selected a two-factor solution with varimax rotation. Consistent with previous work (Bond & Lun, 2014), we label these factors as “Self-Directedness versus Other-Directedness” and “Civility versus Practicality.” The Self-Directedness versus Other-Directedness dimension included the qualities, “independence,” “thrift, saving money and things,” “determination, perseverance,” “imagination,” “obedience” (negatively loaded), and “religious faith” (negatively loaded). The Civility versus Practicality dimension included the qualities, “tolerance and respect for other people,” “good manners,” and “hard work” (negatively loaded).

Table 2.

Factor structure of the endorsement of preferred qualities in the socialization of children.

Items	Self-Directedness versus Other-Directedness Factor variance = 36%	Civility versus PracticalityFactor variance = 18%
Obedience	–0.829
Independence	0.825
Thrift, saving money and things	0.792
Religious faith	–0.778
Determination, perseverance	0.631
Feeing of responsibility	0.621	0.475
Imagination	0.604
Unselfishness
Tolerance, respect for other people		0.797
Good manners		0.779
Hard work		–0.567

Only factors with factor loading ⩾0.40 are presented.

Referring to former studies that focusing on subjective social indicators, we aggregated Level-1 variable across participants to index Level-2 variable (Almakaeva et al., 2018). For each country/territory, factor scores were calculated by weighted averaging item scores within each factor according to their item loadings, with a higher score indicating the relative importance of Self-Directedness and Civility. The item “feeling of responsibility” was dropped from calculating either factor due to its high loadings on both factors. Figure 2 presents the two-dimensional graph for the endorsement of preferred qualities in the socialization of children across 47 countries/territories.

Figure 2.

The two-dimensional graph of the endorsement of preferred qualities in the socialization of children across 47 nations/territories.

Statistical analyses

To examine whether a citizen’s religiosity (Level-1 factor) mediates the relationship between their gender (Level-1 factor) and justification of morally debatable behaviors (Level-1 factor), and whether the link between religiosity and justification of morally debatable behaviors are moderated by the endorsement of preferred qualities in the socialization of children (Level-2 factors), we conducted unconflated multilevel moderated mediation analyses (1-1-1 multilevel mediation model with Level-2 moderators) (Hayes & Rockwood, 2020; Zhang et al., 2009). Both Self-Directedness versus Other-Directedness and Civility versus Practicality were simultaneously introduced into each model for modeling the two-dimensional nature of the endorsement of preferred qualities in the socialization of children. The effect of age and education upon religiosity and justification of morally debatable behaviors were controlled in both levels of the models. All models were estimated using the Bayesian method with a maximum number of iterations of 10,000, using the software Mplus. We set the cut-off p-value as 0.01 in this study. Listwise deletion was used for handling missing data.

Results

Table 3 shows the descriptive statistics for the key variables studied. Table 4 shows the summary of the multilevel analysis conducted.

Table 3.

Descriptions of key variables and demographics of participants.

Variables	N	Mean (SD)	Range	ICC
Individual-level
Religiosity	65,089	0.16 (0.04)	0.07 to 0.33	0.39
Justification of dishonest-illegal behaviors	65,249	5.46 (4.09)	2.53 to 25.28	0.13
Justification of interpersonally violent behaviors	67,072	4.76 (3.63)	2.21 to 22.09	0.12
Justification of behaviors threatening human life and family values	63,979	8.25 (5.34)	2.34 to 23.43	0.31
Age	66,707	42.77 (16.26)	16 to 103	0.12
Gender	66,957	0.53 (0.50)	0 to 1	0.01
Education	66,496	1.97 (0.81)	1 to 3	0.19
Country-level
Self-Directedness versus Other-Directedness	47	0.34 (0.53)	–0.80 to 1.43
Civility versus Practicality	47	0.80 (0.24)	–0.10 to 1.17

For gender, 0 = male and 1 = female. SD: standard deviation; ICC: intraclass correlation coefficient.

Table 4.

Multilevel estimates for the 1-1-1 moderated mediation models.

	Justification of dishonest-illegal behaviors			Justification of interpersonally violent behaviors			Justification of behaviors threatening human life and family values
	Estimate (Posterior SD)	p	95% CI	Estimate (Posterior SD)	p	95% CI	Estimate (Posterior SD)	p	95% CI
Within-country level
Religiosity → JOMDB	−1.89 (1.22)	0.058	(−4.31, 0.51)	1.08 (1.00)	0.133	(−0.88, 3.05)	–25.73 (3.14)	<0.001	(−31.95, −19.60)
Gender → JOMDB	–0.20 (0.03)	<0.001	(−0.26, −0.14)	–0.34 (0.03)	<0.001	(−0.39, −0.29)	0.19 (0.03)	<0.001	(0.12, 0.25)
Gender → Religiosity	0.01 (0.000)	<0.001	(0.01, 0.01)	0.01 (0.000)	<0.001	(0.01, 0.01)	0.01 (0.000)	<0.001	(0.01, 0.01)
Indirect effect	–0.01 (0.01)	0.058	(−0.02, 0.003)	0.01 (0.01)	0.133	(−0.01, 0.02)	–0.14 (0.02)	<0.001	(−0.18, −0.11)
Cross-level interaction
S-D/O-D × Religiosity → JOMDB	5.25 (2.44)	0.018	(0.37, 10.05)	1.76 (1.95)	0.187	(−2.18, 5.47)	–17.81 (6.05)	0.002	(−29.84, −5.93)
C-P × Religiosity → JOMDB	14.39 (5.26)	0.004	(3.81, 24.71)	3.99 (4.21)	0.171	(−4.04, 12.44)	–17.67 (13.39)	0.090	(−43.90, 8.42)

C/P: Civility versus Practicality; JOMDB: justification of morally debatable behaviors; S-D/O-D: Self-Directedness versus Other-Directedness. SD: standard deviation; CI: confidence interval.

The effect of age and education upon religiosity and JOMDB were controlled in both levels of the models; these effects were not presented for simplicity. For gender, 0 = male and 1 = female.

The mediation analyses results

Justification of dishonest-illegal behaviors

Respondents’ gender was positively correlated with religiosity (estimate = 0.01, p < 0.001, 95% confidence interval (CI) = (0.01, 0.01)); religiosity was not correlated with justification of dishonest-illegal behaviors (p = 0.058); gender was negatively correlated with the justification of dishonest-illegal behaviors (estimate = −0.20, p < 0.001, 95% CI = (−0.26, −0.14)). The indirect effect of religiosity was non-significant (p = 0.058).

Justification of interpersonally violent behaviors

Respondents’ gender was positively correlated with religiosity (estimate = 0.01, p < 0.001, 95% CI = (0.01, 0.01)); religiosity was not correlated with justification of interpersonally violent behaviors (p = 0.133); gender was negatively correlated with justification of interpersonally violent behaviors (estimate = −0.34, p < 0.001, 95% CI = (−0.39, −0.29)). The indirect effect of religiosity was non-significant (p = 0.133).

Justification of behaviors threatening human life and family values

Respondents’ gender was positively correlated with religiosity (estimate = 0.01, p < 0.001, 95% CI = (0.01, 0.01)); religiosity was negatively correlated with justification of behaviors threatening human life and family values (estimate = −25.73, p < 0.001, 95% CI = (−31.95, −19.60)); gender was positively correlated with the justification of behaviors threatening human life and family values (estimate = 0.19, p < 0.001, 95% CI = (0.12, 0.25)). The indirect effect of religiosity was negative (estimate = −0.14, p < 0.001, 95% CI = (−0.18, −0.11)).

The moderation effects of the endorsement of preferred qualities in the socialization of children

Justification of dishonest-illegal behaviors

Civility versus Practicality moderated the correlation between religiosity and justification of dishonest-illegal behaviors (estimate = 8.38, p ⩽ 0.001, 95% CI = (4.72, 12.05)): their relation was non-significant in societies prioritizing Civility (p = 0.198) but negative in societies prioritizing Practicality (estimate = −5.18, p = 0.002, 95% CI = (−8.62, −1.75)) (Figure 3(a)). The indirect effect of religiosity was moderated accordingly (Supplementary Table 1).]

Figure 3.

Illustrations of the cross-level moderating effect of the endorsement of preferred qualities in the socialization of children on the relationship between religiosity and justification of morally debatable behaviors.

Justification of interpersonally violent behaviors

Neither Self-Directedness versus Other-Directedness nor Civility versus Practicality moderated the correlation between religiosity and justification of interpersonally violent behaviors (ps > 0.171).

Justification of behaviors threatening human life and family values

Self-Directedness versus Other-Directedness moderated the correlation between religiosity and justification of behaviors threatening human life and family values (estimate = −17.81, p = 0.002, 95% CI = (−29.84, −5.93)): their negative relation was stronger in societies prioritizing Self-Directedness (estimate = −35.21, p < 0.001, 95% CI = (−43.94, −26.41)) compared with societies prioritizing Other-Directedness (estimate = −16.38, p < 0.001, 95% CI = (−25.18, −7.62)) (Figure 3(b)). The indirect effect of religiosity was moderated accordingly (Supplementary Table 1).

Discussion

This study was designed to test a moderated mediation model predicting the justification given by persons to different types of morally justifiable behavior. In that model, personal religiosity was hypothesized to mediate gender effects on personal-sexual and dishonest-illegal behaviors; the linkage strength of personal religiosity to personal-sexual and dishonest-illegal behaviors was hypothesized to be moderated by the endorsement of preferred qualities in the socialization of children, namely, a cross-level moderation effect. Given the 47-society data set available to test this model, we found that three, not two, types of morally justifiable behavior best described the justifiability ratings made by our 66,992 respondents to the 11 behaviors they rated: dishonest-illegal behaviors (parallel with dishonest-illegal behaviors in Vauclair & Fischer, 2011); interpersonally violent behaviors (adding a distinct type of morally debatable behavior to those found in Vauclair & Fischer, 2011), and behaviors threatening human life and family values, a type of morally debatable behavior overlapping somewhat with the “personal-sexual behavior” identified by Vauclair and Fischer (2011). The hypotheses we initially offered are thus evaluated below with respect to these three types of morally justifiable behavior.

Gender

Across all 47 nations and territories, women justified behaviors threatening human life and family values more, but dishonest-illegal behavior and interpersonally violent behavior less than men did. Similar gender differences have been found across societal cultures in much prior research conducted in and across many societies, for example, in gender preferences for mate characteristics, in the provision of nurturance, in personality profiles on the Big Five, and so forth (see Best & Puzio, 2019, for a summary). These culture-general gender differences have typically been explained by the differential biology of males and females and its associated gender-role requirements across human history into the present (e.g. Lippa, 2009).

So, greater female justification of behaviors threatening human life and family values in the present study may be explained as arising because these behaviors, such as abortion and divorce, are closely related to women’s growing contemporary rights to exercise control over their bodies (Beaumont, 2007); lesser female justification of dishonest-illegal behaviors as arising because women are less likely to confront social sanctions for misbehavior in the public arena compared to men, so that women judge civic exploitation more strictly than men; lesser female justification of interpersonally violent behaviors as arising because the traditional female role of nurturer is carried into contemporary societies by women who remain biologically the bearers of the next generation and primary caretakers of the family; women continue to be the communal nurturers of life tasked with sanctioning interpersonal harmful action in the family and in society more widely.

Previous research has repeatedly shown that women report higher personal religiosity than men in both monocultural and multi-cultural studies using various operationalizations of personal religiosity (Beit-Hallahmi, 2004; Saroglou, 2019). In this study, we measured personal religiosity by asking respondents about the importance of religion in their lives relative to the summed importance of six fundamental domains of living. Using this measure, women across these many societies also reported greater personal religiosity than men.

However, the higher personal religiosity of women mediated their greater justification of behaviors threatening human life and family values, but not their greater justification of dishonest-illegal behavior or interpersonally violent behavior. Although women justify these three types of behavior at different levels than do men, only women’s justification of behaviors threatening human life and family values can be attributed to the higher relative importance of religion in their lives compared to men’s. So, women’s different levels of justification of dishonest-illegal and interpersonally violent behavior compared to men must then be a consequence of some other characteristic of women that differs from men across societies, as, for example, has been found for the Big Five personality traits of agreeableness and neuroticism (McCrae, 2009). Given that religiosity is only moderately tapped by the Big Five (Saroglou, 2010), it is then plausible that women’s greater personal religiosity and its associated traditions of gender-specific responsibilities for family life are mediating their greater justification of behaviors threatening human life and family values than was found for men. Other pan-societal gender differences (Schmitt et al., 2017) may also be hypothesized to test and explain these gender differences in the justifiability of these three types of morally debatable behaviors.

Societal endorsement of preferred qualities in the socialization of children as cross-level moderator

Societal endorsement of preferred qualities in the socialization of children were assessed by the percentage of a representative sample from a nation or territory’s population selecting certain preferred qualities in training their children. As in other research (e.g. Jing & Bond, 2015), two bi-polar dimensions were identified in this research: Self-Directedness versus Other-directedness and Civility versus Practicality. These societal-level factors distinguished societies in this study as shown in Figure 3, and moderated the links found in this study between personal religiosity and two of the three types of morally justifiable behaviors.

Across all societies, personal religiosity strongly and negatively predicted the justification of behaviors threatening human life and family values, confirming Hypothesis 2. However, as shown in Figure 3(b), where societies socializing children for higher self-directedness, religious persons justified behaviors threatening human life and family values more negatively than did non-religious persons in other-directed societies, confirming Hypothesis 2a. We believe that societies socializing children for self-directedness and its societally held value of autonomy (Bond & Lun, 2014) confer greater authority upon their citizens to control their personal affairs, like their behaviors threatening human life and family values. So, religious persons in self-directed societies are more likely than those in other-directed societies to support the typical rejection of behaviors threatening human life and family values found in all societies.

As reported in Table 4 and shown in Figure 3(a), persons reporting higher personal religiosity reported a stronger negative link with the justification of dishonest-illegal behavior in societies higher in civility than did those persons in societies higher in practicality. In societies socializing children for civility and its societally held value of egalitarianism, dishonest-illegal behaviors are no more strongly justified by religious citizens than by non-religious citizens, failing to confirm Hypothesis 3. However, as Hypothesis 3a proposed, in societies higher in civility, community norms for justly managing the commons have been socialized for all citizens and associated egalitarian values are widespread (Bond & Lun, 2014). In consequence, we believe that the communitarian orientation associated with religiosity everywhere is less needed in civil societies to support the condemnation of dishonest-illegal behaviors. In practical societies, however, the lack of cooperative social norms is compensated by the communitarian ethos promoted by their more religious members by sanctioning dishonest-illegal behaviors.

These cross-level moderations of religiosity’s relationship to the justification of these two types of morally debatable behaviors are revealing and suggestive. The confusing literature on religiosity and moral codes, each measured in many ways across many types of cultural groups, may arise in part because societal differences across these cultural groups have not been considered as a factor in determining each cultural group’s nomological network on religiosity. Given the results of the present study, the meanings of religiosity as a personality factor shaping cognitive, emotional, and behavioral processes appear to vary by how a society endorses the character qualities in training its children.

No cross-level moderating effects were found for the relationship of religiosity to the justification of interpersonally violent behaviors. So, is the connection between personal religiosity and injunctions against violence therefore a “universal,” that is, an “invariant” of the human condition? Does its invariance across cultural systems arise because all members of all societies are socialized to condemn violence regardless of other societal endorsement of preferred qualities in the socialization of children? Or, do other societal factors, such as inequality (Wilkinson et al., 1998), moderate the individual-level association between personal religiosity and the justification of interpersonal violence? Further research seems warranted.

Limitations and future research

Greater representation of countries from Africa and Central/Latin America would improve the generalizability of our results and is welcome in future waves of the World Values Survey. Understandably however, working with wide-ranging surveys across many cultural-linguistic groups requires many compromises by survey creators and subsequent adjustments by multi-cultural psychologists who are intent on finding the “kernel of truth” about societal culture as an important context for human behavior (Smith & Bond, 2019). We must apply due diligence and work with whatever data are available and trustworthy.

Future research on religiosity, however that concept is measured, must return to the roots of a religious person’s religious heritage. Religions are various and varied in their ontological, theological, societal, and moral teachings as well as their daily practices. Different religions emerged at different periods in history in different geographical regions, and those that survived were later transmitted across the planet with varying levels of adoption. Across time and the changing ecology of those societies adopting a particular religion, each religion adapted its ideology and practices to meet the requirements of local circumstances. Through that process, some religions were more successful at retaining their purity than others, so the imprint of that success will be reflected in the details of the religiosity absorbed by its contemporary legatees (see, for example, Cohen & Hill, 2007). Future cross-cultural work on morality across personal religiosity (and gender) must explore a more fine-grained study of religiosity’s nomological network across societies with different religious heritages, just as it has with values (Georgas et al., 2004).

Conclusion

We explored the justification of three types of morally debatable behaviors, namely, behaviors threatening human life and family values, dishonest-illegal behaviors, and interpersonally violent behaviors, across representative samples of persons in 47 nations and territories. Across these societal cultures, women justified dishonest-illegal and interpersonally violent behaviors less than men, but behaviors threatening human life and family values more than men. Personal religiosity mediated the gender difference in condoning behaviors threatening human life and family values.

These 47 societies varied in their endorsement of preferred qualities in the socialization of children. As found in prior cross-cultural research (Bond & Lun, 2014), these endorsements were shown to vary across two dimensions in this study, Self-Directedness versus Other-Directedness and Civility versus Practicality. The linkages of personal religiosity to the endorsement of the moral codes for behaviors threatening human life and family values and dishonest-illegal behaviors were enhanced by these socialization priorities for children characterizing each society.

We propose that this “nuanced” effect by the societal-cultural context of socialization can help explain variable results for the associations of personal religiosity found in other studies for many outcomes other than moral judgment, for example, in studies of interpersonal trust (Jing & Bond, 2015). Exploring societal-level factors as moderators of individual processes enables researchers to distinguish culture-general from culture-specific results with greater precision (Smith & Bond, 2019). We encourage researchers to include societally variable factors involved in member induction into cultural groups of any type in future cross-cultural studies to demonstrate how cultural context impacts the psychological processes of its individual members.

Supplemental Material

sj-docx-1-prj-10.1177_00846724231197239 – Supplemental material for Unpackaging gender differences in justifying morally debatable behaviors around the world: The role of personal religiosity and society’s socialization priorities for its children

Supplemental material, sj-docx-1-prj-10.1177_00846724231197239 for Unpackaging gender differences in justifying morally debatable behaviors around the world: The role of personal religiosity and society’s socialization priorities for its children by Michael Harris Bond and Xiaobin Lou in Archive for the Psychology of Religion/Archiv Für Religionpsychologie

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.

Ethics statement

This study used data from the open data set of the World Values Survey. Thus, IRB approval and informed consent were not required.

Funding

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

ORCID iD

Xiaobin Lou

Supplemental material

Supplemental material for this article is available online.

References

Almakaeva

Welzel

Ponarin

. (2018). Human empowerment and trust in strangers: The multilevel evidence. Social Indicators Research, 139, 923–962. https://doi.org/10.1007/s11205-017-1724-z

Atari

Lai

M. H.

Dehghani

. (2020). Sex differences in moral judgements across 67 countries. Proceedings of the Royal Society B, 287(1937), Article 20201201. https://doi.org/10.1098/rspb.2020.1201

Beaumont

. (2007). Moral dilemmas and women’s rights: The attitude of the Mothers’ Union and Catholic Women’s League to divorce, birth control and abortion in England, 1928 - 1939. Women’s History Review, 16(4), 463–448. https://doi.org/10.1080/09612020701445677

Beit-Hallahmi

. (2004). Religion, religiosity, and gender. In Ember

C. R.

Ember

(Eds.), Encyclopedia of sex and gender: Men and women in the world’s cultures (pp. 117–127). Springer.

Berry

J. W

. (2017). Theories and models of acculturation. In Schwartz

S. J.

Unger

J. B.

(Eds.), Oxford handbook of acculturation and health (pp. 15–28). Oxford University Press.

Best

D. L.

Puzio

A. R

. (2019). Gender and culture. In Matsumoto

Hwang

H. C.

(Eds.), The handbook of culture and psychology (pp. 235–291). Oxford University Press. https://doi.org/10.1093/oso/9780190679743.003.0009

Bond

M. H.

Lun

M.-C

. (2014). Citizen-making: The role of national goals for socializing children. Social Science Research, 44, 75–85. https://doi.org/10.1016/j.ssresearch.2013.11.002

Bronfenbrenner

Ceci

S. J

. (1994). Nature-nurture reconceptualized in developmental perspective: A bioecological model. Psychological Review, 101, 568–586. https://doi.org/10.1037/0033-295x.101.4.568

Cohen

A. B.

Hill

P. C

. (2007). Religion as culture: Religious individualism and collectivism among American Catholics, Jews, and Protestants. Journal of Personality, 75, 709–742. https://doi.org/10.1111/j.1467-6494.2007.00454.x

10.

Dunn

E. C.

Masyn

K. E.

Johnston

W. R.

Subramanian

S. V

. (2015). Modeling contextual effects using individual-level data and without aggregation: An illustration of multilevel factor analysis (MLFA) with collective efficacy. Population Health Metrics, 13(1), 1–11. https://doi.org/10.1186/s12963-015-0045-1

11.

Georgas

van de Vijver

F. J. R.

Berry

J. W

. (2004). The ecocultural framework, ecosocial indices, and psychological variables in cross-cultural research. Journal of Cross-Cultural Psychology, 35, 74–96. https://doi.org/10.1177/0022022103260459

12.

Graham

Haidt

Koleva

Motyl

Iyer

Wojcik

S. P.

Ditto

P. H

. (2013). Moral foundations theory: The pragmatic validity of moral pluralism. Advances in Experimental Social Psychology, 47, 55–130. https://doi.org/10.1016/B978-0-12-407236-7.00002-4

13.

Haerpfer

Inglehart

Moreno

Welzel

Kizilova

Diez-Medrano

Lagos

Norris

Ponarin

Puranen

, et al. (2020). World Values Survey: Round seven—country-pooled datafile. JD Systems Institute & WVSA Secretariat. https://doi.org/10.14281/18241.1

14.

Harding

Phillips

. (1986). Contrasting values in Western Europe: Unity, diversity, and change. Macmillan.

15.

Hayes

A. F.

Rockwood

N. J

. (2020). Conditional process analysis: Concepts, computation, and advances in the modeling of the contingencies of mechanisms. American Behavioral Scientist, 64(1), 19–54. https://doi.org/10.1177/0002764219859633

16.

Jing

Bond

M. H

. (2015). Sources for trusting most people: How national goals for socializing children promote the contributions made by trust of the in-group and the out-group to non-specific trust. Journal of Cross-Cultural Psychology, 46(2), 191–210.

17.

Lippa

R. A

. (2009). Sex differences in sex drive, sociosexuality, and height across 53 nations: Testing evolutionary and social structural theories. Archives of Sexual Behavior, 38(5), 631–651. https://doi.org/10.1007/s10508-007-9242-8

18.

McCrae

R. R

. (2009). The five-factor model of personality traits: Consensus and controversy. In Corr

P. J.

Matthews

(Eds.), The Cambridge handbook of personality psychology (pp. 148–161). Cambridge University Press.

19.

Miller

J. G.

Wice

Goyal

. (2019). Culture and moral development. In Matsumoto

Hwang

H. C.

(Eds.), Oxford handbook of culture and psychology (2nd ed., pp. 318–360). Oxford University Press.

20.

Saroglou

. (2010). Religiousness as a cultural adaptation of basic traits: A five-factor model perspective. Personality and Social Psychology Review, 14(1), 108–125. https://doi.org/10.1177/1088868309352322

21.

Saroglou

. (2019). Religion and related morality across cultures. In Matsumoto

Hwang

H. C.

(Eds.), The handbook of culture and psychology (2nd ed., pp. 732–793). Oxford University Press.

22.

Schmitt

D. P.

Long

A. E.

McPhearson

O’Brien

Remmert

Shah

S. H

. (2017). Personality and gender differences in global perspective. International Journal of Psychology, 52(S1), 45–56. https://doi.org/10.1002/ijop.12265

23.

Smith

P. B.

Bond

M. H

. (2019). Cultures and persons: Characterizing national and other types of cultural difference can also aid our understanding and prediction of individual variability. Frontiers in Psychology, 10, Article 2689. https://doi.org/10.3389/fpsyg.2019.02689

24.

Vauclair

C. M.

Fischer

. (2011). Do cultural values predict individual’s moral attitudes? A cross-cultural multilevel approach. European Journal of Social Psychology, 41, 645–657. https://doi.org/10.1002/ejsp.794

25.

Wilkinson

R. G.

Kawachi

Kennedy

B. P

. (1998). Mortality, the social environment, crime and violence. Sociology of Health & Illness, 20(5), 578–597. https://doi.org/10.1111/1467-9566.00120

26.

Zhang

Zyphur

M. J.

Preacher

K. J

. (2009). Testing multilevel mediation using hierarchical linear models: Problems and solutions. Organizational Research Methods, 12(4), 695–719.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.11 MB