Empathy and humanitarianism predict preferential moral responsiveness to in-groups and out-groups

ABSTRACT

The current research tests whether empathy—sharing others’ emotions—and humanitarianism—recognizing the moral worth of all people—each predict moral responsiveness toward others but in ways that favor in-groups and out-groups, respectively. In Studies 1 and 2, empathy and humanitarianism differentially predicted preferential moral concern for in-groups and out-groups. In Study 3, humanitarianism predicted lower in-group-targeted prosociality and greater out-group prosociality. In Study 4, empathy and humanitarianism predicted perceived moral obligation to in-groups and out-groups respectively. In Study 5, out-group obligation mediated between humanitarianism and allocations to out-group charities, and in-group obligation mediated between empathy and one of two in-group charities. In sum, empathy and humanitarianism are associated with preferential morality via group-based obligation, suggesting that morality could be extended by altering empathy, humanitarianism, or group processes.

KEYWORDS:

• Empathy
• humanitarianism
• morality; group processes
• prosociality
• values

View correction statement:

Correction

Data availability statement

The data described in this article are openly available in the Open Science Framework at http://bit.ly/2mLcXyu

Open Scholarship

This article has earned the Center for Open science badges for Open Data and Open Materials through Open Practices Disclosure. The data and materials are openly accessible at http://bit.ly/2mLcXyu

Notes

1. The general pattern and direction of results in all studies is the same when predictors (humanitarianism and empathy) are decomposed into the separate scales from which their items originate. The hypotheses were also supported when testing each of our hypotheses using the Empathic Concern (EC; Davis, 1980) subscale only, rather than the composite of measures. We tested hypotheses with the EC only because it is the most-cited measure of affective empathy from which we draw items, and so readers may want more information to compare our results with the wealth of other research using the EC. In Studies 1 and 2, EC-only analyses supported our hypotheses that empathy predicts in-group moral concern and humanitarianism predicts out-group moral concern. In Study 3, EC-only analyses supported our hypotheses that humanitarianism, but not empathy, predicts out-group charity. In Study 4, EC-only analyses supported our hypotheses that humanitarianism, but not empathy, predicts out-group moral obligation, and that empathy, but not humanitarianism, predicts in-group moral obligation. In Study 5, EC-only analyses supported our hypotheses that empathy predicts in-group charity votes via in-group obligation, and that humanitarianism predicts out-group charity votes via out-group obligation.

2. All studies were part of a large data collection and not all measured variables are reported here. Data from these and from reported measures are included in the publicly available materials and dataset posted on the Open Science Framework (https://osf.io/ygy5e/), where analysis syntax is also available. There is sufficient information for an independent researcher to reproduce the reported methodology and results.

3. We recognize that some readers may be more accustomed to seeing preferences represented using difference scores. Thus, we also computed the difference between in-group and out-group moral concern, and used that difference score as the outcome in a regression using empathy and humanitarianism as predictors. The overall model was significant, R² = .18, p < .001. As expected, empathy predicted greater preferential moral responsiveness for in-group targets relative to out-group targets, b = 0.47 (95% CI = 0.05, 0.89), β = .11, SE = .21, p = .027, and humanitarianism predicted less preferential moral responsiveness, b = −2.07 (95% CI = −2.48, −1.66), β = -.48, SE = .21, p < .001.

4. We recognize that some readers may be more accustomed to seeing preferences represented using difference scores. Thus, we also computed the difference between in-group and out-group moral concern, and used that difference score as the outcome in a regression using empathy and humanitarianism as predictors. The overall model was significant, R² = .12, p < .001. As expected, empathy predicted greater preferential moral responsiveness for in-group targets relative to out-group targets, b = 0.75 (95% CI = 0.40, 1.09), β = .17, SE = .18, p < .001, and humanitarianism predicted less preferential moral responsiveness, b = −1.69 (95% CI = −2.02, −1.35), β = -.39, SE = .17, p < .001.

5. We recognize that some readers may be more accustomed to seeing preferences represented using difference scores. Thus, we also computed the difference between in-group and out-group moral obligation, and used that difference score as the outcome in a regression using empathy and humanitarianism as predictors. The overall model was significant, R² = .07, p < .001. As expected, empathy predicted greater preferential moral obligation for in-group targets relative to out-group targets, b = 0.08 (95% CI = 0.01, 0.16), β = .10, SE = .04, p = .03, and humanitarianism predicted less preferential moral obligation, b = −0.27 (95% CI = −0.34, −0.18), β = -.31, SE = .04, p < .001.