Contingencies of self-worth and the strength of deontological and utilitarian inclinations

ABSTRACT

Many studies have tried to understand what makes people adopt deontological or utilitarian inclinations when forming moral judgments or making moral choices. The present research examined the impact of motivational factors: contingencies of people’s self-worth on the strength of deontological and utilitarian inclinations. Study 1 found positive correlations between two contingencies of self-worth: basing one’s self-worth on being a virtuous, moral person (Virtue) and outperforming others (Competition), and the strength of deontological inclinations and utilitarian inclinations respectively. Studies 2 and 3 found that increasing saliency of Virtue as a source of one’s self-worth selectively increased the strength of deontological inclinations, leaving utilitarian inclinations unchanged, while the opposite was true for increasing saliency of Competition as a source of self-worth and the strength of utilitarian inclinations: such priming increased utilitarian inclinations while deontological inclinations remained unchanged.

KEYWORDS:

• contingencies of self-worth
• deontological judgment
• utilitarian judgment
• moral self
• process dissociation procedure

Data availability statement

The data described in this article are openly available in the Open Science Framework at https://osf.io/b7d2t/

Open scholarship

  

This article has earned the Center for Open Science badges for Open Data, Open Materials and Preregistered. The data and materials are openly accessible at https://osf.io/b7d2t/

Notes

1. The scale consists of two – public (Symbolization) and private (Internalization) – aspects. Symbolization taps the degree to which moral traits are reflected in the respondent’s actions in the world, while internalization refers to the degree to which these traits are central to person’s self-concept.

2. This method enables deontological and utilitarian inclinations to be calculated both conventionally – as a single dimension with deontological inclinations representing one end of a continuum and utilitarian inclinations representing the other end (thus, the stronger one’s deontological inclinations are, the weaker their utilitarian inclinations are, and vice versa) – and as independent parameters (deontological and utilitarian inclinations measured that way are uncorrelated with each other but correlated with conventionally measured inclinations, see: Conway & Gawronski, 2013).

3. The Ariadna survey panel is a Polish research panel with over 80 000 registered Polish responders aged 15–65.

4. They were compensated according to the compensation scheme employed by Ariadna’s administrators, namely with points that they could later exchange for rewards in panel’s catalog.

5. Conway and Gawronski (2013) battery of dilemmas consists of 10 basic dilemmas presented in two versions: congruent and incongruent. Incongruent dilemmas pit deontological concerns about causing harm against utilitarian concerns about maximizing overall good consequences. In congruent dilemmas, harmful action is unappealing from deontological perspective and at the same time not justified (or very hard to justify) on utilitarian grounds. Choices in incongruent dilemmas set serve as an index of the relative, bipolar measure of deontological/utilitarian judgment. In other, words, bipolar traditional score of deontological vs. utilitarian judgment is a proportion of “inappropriate” responses on 10 incongruent dilemmas This measure is relative because stronger endorsement of deontological judgment (more choices of solutions that align with deontological principle in the dilemma set) reflect at the same time weaker endorsement of utilitarian judgment.

6. Several studies have reported significant gender differences in strength of deontological inclinations, and such a difference was presently observed. There was a significant difference in the strength of both types of deontological inclinations: measured conventionally p =.038, and with the process dissociation method (deontological PD parameter) p =.007, with females being significantly more deontological than males. There was no gender difference in the strength of utilitarian inclinations (p =.662), but women also had higher scores than men on Empathic Concern (p =.031). The gender difference in cognitive reflection was only marginally non-significant (p =.090), with men having slightly higher CRT scores. No significant gender differences were observed in the contingencies of self-worth (all ps >.345).

7. Additional analysis showed that gender was a significant predictor of conventionally measured deontological inclinations, but the effect was fully mediated by empathy.

8. In this study and a previous study (Byrd & Conway, 2019), Cognitive reflection was uncorrelated with the Deontological PD Parameter. Therefore it was not included in the analysis.

9. In a manipulation check study (a separate study, part of another project), 54 participants (not taking part in Study 2) completed three tasks. First, they completed Scales Measuring Agency and Communion, by Wojciszke and Szlendak (2010), measuring their self-perceptions on two basic dimensions of social perception (agency and communion), then either Virtue (n = 25) or Competition (n = 29) manipulation task, and finally three subscales from Contingencies of Self-Worth Scale: Virtue, Competition and Approval from Others (Crocker & Wolfe, 2001). In the Virtue/Competition manipulation task, participants were asked to recall three situations from their life when they: (1) had chosen to act in accordance with their moral standards instead of taking a less ethical, easier way (Virtue priming) or; (2) had performed better than other people (Competition priming). After recalling the three situations, they were asked to describe one of them more thoroughly, so somebody reading the description could sense how they felt in that situation. Nine participants were excluded for not following the instructions of the priming task (6 in Virtue and 3 in Competition primed group). Not following the instructions meant writing “I don’t remember/I don’t know”, providing general reflections without any reference to their personal experience, describing situations of luck (winning a lottery) or group activities without showing how individual contribution of the participants was responsible for the final results.

10. The exclusion criteria were similar to those employed in the manipulation check study. Thus, failing to complete the task activating salience of Virtue or Competition as a source of self-worth meant writing “I don’t remember/I don’t know”, writing general reflections instead of describing particular events/situations, providing incomprehensible descriptions or listing and describing situations of luck or group successful activities.

11. The value of p was nearly significant, with participants in the Virtue manipulation group being slightly older (M = 25.86 years, SD = 7.17) than in the Competition manipulation group (M = 22.24 years, SD = 5.87) and Control group (M = 23.23 years, SD = 5.99). However participants in neither the Virtue nor Competition manipulation group differed significantly from the Control group, t(62) = 1.60, p =.11, and t(67) =.70, p =.49, respectively.

12. The Answeo survey panel is a newly established Polish research panel with over 7 000 registered Polish participants.

13. In order to obtain small size effects (f = 0.20) with 95% power in 2 × 3 repeated measures, within-between interaction ANOVA, (alpha =.05, correlations among repeated measures =.1, non-sphericity correction = 1) required sample would be 177. Thus, conservatively the final sample was set at 255.

14. The exclusion criteria were as follows: (1) similarly to Study 2, not conforming with an instruction to provide a description of experiences in the tasks increasing the saliency of Virtue or Competition as a source of self-worth; (2) not providing a correct response to the pseudo-dilemma that served as an attention check question; (3) spending less than one minute on the manipulation task. Additionally, the catch question from MFQ was placed among Contingencies of Self-Worth subscale items (only one person failed this question). All the participants that were excluded failed the pseudo-dilemma attention check question. Among these, five also failed to provide a satisfactory description of their experience during the manipulation task, one also failed the MFQ attention check question. Responses times were not registered due to an error in LimeSurvey settings, thus exclusion was based solely on two criteria described above: responses to attention check questions and not responding properly on the task increasing the saliency of Virtue or Competition as a source of self-worth.

15. They received equivalent of $ 0.80 for completing the study and $ 2 after assessment of their response quality. Information about “response quality assessment” aimed at encouraging participants to read all the materials before responding, to avoid distractors while answering, and to put some effort to provide the description of their experience in the task increasing saliency of Virtue or Competition as a source of one’s self-worth. They were informed that if they did not provide any examples of their experiences (answering “don’t remember” or “don’t know”), or did not describe one of their experiences, they would be not paid additional money.

16. This consisted of 20 dilemmas from Conway and Gawronski (2013) and one pseudo-dilemma at the end of the set, as an attention check question.

17. There was a small difference in wording in the instructions in Study 2 and 3, aimed at increasing participants involvement in this task, keeping in mind financial motivation for participation in the study, which was not the case in Study 2. In Study 2 they were asked to list 3 instances of virtuous or successful behavior. In Study 3 they were asked to list 2–3 examples of such behavior but were also asked to leave the study if they cannot remember such situations since not responding to all tasks in the study excludes their results from planned analyses. Additionally, they were asked to try to provide at least 4 sentences when describing their experience in detail (part 2 of the priming task).

18. Actually, in the case of PD Deontology, no significant correlation was found between this parameter and MI Internalization, while PD Utilitarian was positively correlated with MI Internalization in a sample including all participants, but this relationship ceased to be significant for the reduced sample. However, contrary to other studies, scores on the MI Internalization and MI Symbolization scales were not significantly correlated for either the whole or the reduced sample.

19. I am grateful for the Reviewer for bringing my attention to this way of thinking.

Additional information

Notes on contributors

Anna Macko

Anna Macko is an adjunct in the Department of Economic Psychology and a researcher at the Center for Economic Psychology and Decision Sciences at Kozminski University. Her research is mainly focused on ethical judgment and decision making, risk, and factors influencing decisions to trust and responses to trust violations.