Understanding Self-Respect and Its Relationship to Self-Esteem

Abstract

The concept of self-respect has received little attention in the psychological literature and is not clearly distinguished from self-esteem. The present research sought to empirically investigate the bases of self-respect by manipulating adherence to morals together with interpersonal appraisals (IAs), or task-related competence, in hypothetical scenarios (Studies 1a and 1b) and a situation participants relived (Studies 2 and 3). Participants’ levels of state self-respect and self-esteem were measured. Studies 1 to 3 found main effects of adherence to morals on self-respect, with self-respect mediating the effect of adherence to morals on self-esteem, but little support for competence and IAs directly influencing self-respect. Self-respect uniquely contributed to anticipated/felt self-esteem alongside competence or IAs. The pattern of results supports the conceptualization of self-respect as a component of self-esteem associated with morally principled conduct, distinct from performance and social self-esteem. The findings have implications for our understanding of self-esteem and moral behavior.

Keywords

self-respect self-esteem moral self competence interpersonal appraisals

Although we are all familiar with the term “self-respect,” the concept has received much less attention in the psychological literature than self-esteem (Kristjansson, 2007; Roland & Foxx, 2003). The two concepts are not clearly distinguished, with the terms often used interchangeably in popular culture, or self-respect being used to operationalize global self-esteem (e.g., Kernis, 2003; Marsh & O’Neill, 1984; Rosenberg, 1965). For example, Rosenberg, Schooler, Schoenbach, and Rosenberg (1995) write that the “central feature of global self-esteem appears to be self-acceptance or self-respect” and that “competence is only one factor contributing to such feelings” (p. 144). However, an examination of the philosophical and psychological literature suggests self-respect and self-esteem are rather distinct concepts and underlines the need for self-respect to take a more central place in psychological research.

Following from the interpersonal respect literature, it is logical to conceptualize self-respect as an attitude of respect for oneself, that is, a positive evaluation of oneself as valuable and deserving of respect, and, therefore, a specific self-evaluation, as opposed to a generalized attitude of favourableness or unfavourableness toward the self as a whole, that is, global self-esteem (Rosenberg et al., 1995). Unlike liking, the attitude of respect is directed toward someone who possesses attributes that command recognition and appreciation, regardless of personal affinities and needs (Van Quaquebeke & Eckloff, 2010). Respect can be paid to someone on the basis of (a) their social position—often referred to as “status respect,” (b) their inherent worth as a human being (Prestwich & Lalljee, 2009)—often referred to as “unconditional respect,” or (c) admirable personal qualities of a moral, principled, or honorable nature (e.g., moral integrity, concern for others’ welfare, working to the limits of ability and mental toughness) and achievements, intellectual talents, and skills (Clucas & St Claire, 2017; Frei & Shaver, 2002; Hamilton & Fallot, 1974; Prestwich & Lalljee, 2009)—often referred to as “achieved respect.” Similarly, the self-respect literature has distinguished between recognition and appraisal self-respect. The former is based on Kantian ideas and views self-respect as derived purely from an appreciation of personhood as a rational, autonomous, equal, and moral agent (see Kristjansson, 2007; Renger, 2017). The latter is based on Aristotelian ideas and views self-respect as earned and merit based, involving a positive appraisal or evaluation of oneself as worthy of honor and the “dutiful regard of oneself and others,” based on conformity to valued standards of conduct, moral integrity, and excellence of character (Dillon, 2010; Kristjansson, 2007; Kumashiro, Finkel, & Rusbult, 2002, p. 1016; Roland & Foxx, 2003). These two notions could exist alongside each other, in a similar manner to the attitudes of unconditional and achieved interpersonal respect (Clucas & St Claire, 2017).

This article concentrates on appraisal self-respect, which is more similar to self-esteem by its stronger evaluative and subjective nature and focus on individual merits, yet is not synonymous with it. Like self-esteem, self-respect is linked to a positive self-appraisal as worthy, but appears to be derived specifically from a favorable appraisal of one’s character and conduct, often from a moral viewpoint, as opposed to other qualities (Dillon, 2010; Roland & Foxx, 2003). Therefore, self-respect may not always be high in an individual with high self-esteem (Roland & Foxx, 2003) and is likely to exist alongside other well-known non–moral-specific self-evaluations such as performance, social approval, and physical appearance (Harter, 1999; Marsh, Craven, & Martin, 2013), which all contribute to global self-esteem. Self-respect has been conceptualized both as a trait, with some people enjoying higher self-respect across time and situations than others (Kristjansson, 2007; Kumashiro et al., 2002), and a state, fluctuating in response to temporary experiences (Luchies, Finkel, McNulty, & Kumashiro, 2010).

Although self-respect has only rarely been presented as a distinct self-evaluation attached to a given domain in the psychological literature, there is some wider recognition that self-esteem involves a sense of self-respect. For instance, the Rosenberg (1965) Self-Esteem Scale (RSES) includes an item on self-respect. Also, Crocker, Luhtanen, Cooper, and Bouvrette (2003) measure the extent to which self-esteem is contingent on moral/ethical conduct and use the term self-respect (instead of self-esteem) in some of the measure’s items.

Moreover, although moral/ethical conduct is recognized as a valued human strength cross culturally (Dahlsgaard, Peterson, & Seligman, 2005) and as a basis for self-esteem (see Crocker et al., 2003), moral self-esteem lacks a consistent operationalization in trait and state multidimensional self-esteem measures, which sometimes assess moral character and behavior along with religiosity (e.g., Tennessee Self-Concept Scale; Fitts, 1964), honesty and trustworthiness more specifically (e.g., Self-Description Questionnaire III; Marsh & O’Neil, 1984), or feelings of having engaged in morally acceptable behavior (e.g., Multidimensional Self-Esteem Inventory; O’Brien & Epstein, 1998). These disparate measures could actually be tapping on a broader self-evaluation as a principled, moral, and honorable person worthy of high regard, that is, self-respect (Kumashiro et al., 2002). Self-respect could, therefore, be an important pathway by which morally principled behavior relates to self-esteem.

Little research has sought to understand how best to conceptualize self-respect. Two notable exceptions provide indirect support for self-respect being attached to a self-conception as moral and honorable. Kumashiro et al. (2002) showed that trait self-respect (defined as perceiving oneself as a principled person worthy of honor and high regard) predicted prorelationship behavior in marital relationships and well-being independently from self-esteem, thus also supporting self-respect and self-esteem being distinct constructs, although the self-respect measure did not undergo a thorough validation process. Luchies et al. (2010) demonstrated an increase in a single-item measure of state self-respect when forgiving a partner who made amends, supposedly the right or honorable thing to do. In addition, Renger (2017) showed recognition self-respect to be distinct from self-esteem. However, the idea that self-respect is related to moral integrity has never been tested directly, and it is unclear whether self-respect has the same bases as interpersonal respect. More research is needed to identify the bases of self-respect, including its relationship with other major specific self-evaluations such as task-related competence and social approval.

Task-related competence is often considered a facet of self-esteem (Harter, 1999; Heatherton & Polivy, 1991). However, it is unclear whether task-related competence also influences self-respect. In addition to qualities relating to moral integrity and principled behavior, admirable qualities such as being inspiring and possessing talents, intellectual qualities, and skills (Frei & Shaver, 2002; Hamilton & Fallot, 1974; Prestwich & Lalljee, 2009) have been shown to influence respect for others. This raises the question of whether task-related competence influences self-respect in addition to principled, honorable, and moral behavior.

Similarly, the relationship between self-respect and interpersonal appraisals (IAs) has not been explored. Much literature has emphasized that self-esteem is influenced by social approval and acceptance (Coopersmith, 1967; Leary, Haupt, Strausser, & Chokel, 1998). However, an individual’s self-respect appears to be based on his or her appraisal of his or her own conduct and character and, thus, appears to be primarily internally driven. Moreover, research has shown that while respect is more strongly linked to intellectual attributes, such as skill, intelligence, and commitment, liking is more strongly linked to social qualities, such as warmth and popularity (Fiske, Cuddy, Glick, & Xu, 2002; Hamilton & Fallot, 1974; Prestwich & Lalljee, 2009). As a result, IAs might be expected to influence self-liking rather than self-respect.

The dearth of research on self-respect has hindered a solid understanding of the nature of the construct. Yet, support that self-respect is a component of self-esteem based on moral integrity would further suggest self-respect is likely to have important implications for moral behavior and social relationships in ways that self-esteem more globally may not (see Baumeister, Campbell, Krueger, & Vohs, 2003; Kumashiro et al., 2002; Roland & Foxx, 2003), and its more controllable nature may make it more amenable to intervention.

Four experimental studies were conducted to further our psychological understanding of self-respect by investigating the influence of adherence to moral principles, IAs, and task-related competence on anticipated or experienced self-respect and self-esteem, and exploring how these concepts differ from and relate to each other. The following was hypothesized:

Hypothesis 1: Adherence to morals will influence self-respect directly.

Hypothesis 2: Self-respect will mediate the effect of adherence to morals on self-esteem.

Hypothesis 3: IAs will influence self-esteem rather than self-respect.

Hypothesis 4: Task-related competence will influence self-esteem (its effect on self-respect will be explored).

Studies 1a and 1b investigated the influence of adherence to morals together with academic competence (Study 1a) or IAs (Study 1b) on anticipated state self-respect and self-esteem in hypothetical scenarios. Study 2 investigated participants’ levels of state self-respect and self-esteem when reliving a situation in their life in which they had (or had not) compromised on their morals and performed well (or badly) academically. Study 3 was similar to Study 2 but focused on more broadly defined task-related competence.

Studies 1a and 1b

Participants completed an online questionnaire, which instructed them to imagine themselves into one of four scenarios (assigned at pseudo-random by rotating questionnaire versions approximately every 10 participants) manipulating adherence to morals with IAs (Study 1a) or with academic competence (Study 1b), creating a 2 × 2 between-subjects factorial design in each study. Participants completed self-respect (SR) and self-esteem (SE) measures at baseline and while imagining themselves in the scenario.

Study 1a: Method

Participants

One hundred fifty-nine U.K. university students were recruited through the psychology department online research participation system (RPS). Sample characteristics are shown in Table 1. All studies were approved by the Psychology Department University Ethics Committee. Participation was voluntary and anonymous.

Table 1.

Sample Characteristics.

	Study 1a	Study 1b	Study 2	Study 3
N	159	148	178	204
M_age (SD)	22.03 (7.09)	24.67 (8.64)	23.20 (6.18)	35.41 (10.88)
Gender (%)
Male	18	64	66	74
Female	140	83	107	128
Unreported	1	1	3	2
Ethnicity
White		103	149	175
White British	146
White Other	2
Asian	3	18	17	15
Black	3	14	2	6
Other	4	11	9	6
Unreported	1	2	1	2

Materials and procedure

Participants were asked to imagine that they belonged to a prestigious student society and they either lied (without society peers knowing) or did not lie about their background to fit in, rejecting past friends, going against or abiding to their moral code and standards they cared about (low adherence to morals [LM] and high adherence to morals [HM], respectively—see Supplemental Appendix A for verbatim). Following the procedure of Leary et al. (1998), participants were then asked to imagine they had the chance to find out exactly how the society members had rated them on the extent to which they would be willing to engage with them in four social activities (such as inviting them if they were planning a party). The ratings were either high (positive IA) or low (negative IA; see Supplemental Appendix A). The four scenarios were matched on all characteristics, except the manipulated dimensions.

Manipulation checks

Participants indicated (a) how accepted by peers (IA ratings) and (b) how principled and moral they felt (moral ratings) on 1 (very rejected) to 10 (very accepted) and 1 (very immoral) to 10 (very moral) numeric rating scales.

Outcome measures

A three-item scale was used to measure state SR in the scenario. Items were responded to on 7-point Likert-type scales (1 = strongly disagree, 7 = strongly agree) and included “I feel I have a high level of self-respect,” “I have a lot of respect for myself,” and “I wish I could have more respect for myself” (α = .89). In all studies, parallel analysis using the fa.parallel function in the psych R package (Revelle, 2018)¹ supported the SR items representing a single factor with factor loadings from .51 to .99 across studies. The single item validated by Robins, Hendin, and Trzesniewski (2001), “I have high self-esteem” (rated on a Likert-type scale of 1 = not very true of me, 7 = very true of me), was used to measure state SE in the scenario.

Baseline measures

Trait SE was measured using the 10-item 4-point Likert-type RSES (Rosenberg, 1965; 1 = strongly disagree, 4 = strongly agree).² Trait SR was measured using the eight-item trait Self-Respect Scale (SRS), which showed good construct validity and internal reliability in past research (Clucas & Wilkinson, 2017), and α = .86 in the present sample. Examples of items are “I will stick to my principles even if asked to do otherwise” and “I see my behaviour as dignified” (rated on a Likert-type scale from 1 = strongly disagree to 7 = strongly agree, developed to represent attributes contributing to a self-conception as moral, principled, and honorable based on existing literature; see Kristjansson, 2007; Kumashiro et al., 2002; Luchies et al., 2010).

State SE was also measured using Heatherton and Polivy’s (1991) 20-item Likert-type State Self-Esteem Scale (SSES; 1 = not at all, 5 = extremely, α = .93) and state SR using the items “I have a lot of respect for myself” (1 = strongly disagree, 7 = strongly agree) and “I wish I could have more respect for myself” (1 = strongly disagree, 4 = strongly agree), whose standardized scores were averaged into one index, Spearman–Brown coefficient (r_SB) = .60.³ Scores were averaged for multi-item outcome measures, manipulation checks, and baseline SR measures. Total scores were otherwise computed.

Analytical approach

A 2 × 2 multivariate analysis of variance (MANOVA) was used given the strong correlation between anticipated scenario SR and SE (ρ = .70).⁴ Despite being high, a correlation of .70 indicates that 51% (1 – .70²) of the variance in state SR and SE remains unshared (Studies 2 and 3 further support the measures as tapping two strongly related but distinct factors in line with our conceptualization of SR as a specific self-evaluation as worthy of respect). A high correlation between the measures was expected given a reciprocal relationship between specific self-evaluations and global SE (Rosenberg et al., 1995). As a component of global SE, high state SR is expected to boost global state SE, whereas positive affectivity associated with high global state SE is likely to inflate global state appraisals of SR (Sedikides, Wildschut, Arndt, & Routledge, 2006). In view of this reciprocal relationship, analyses investigating influences on SR examined whether the results were accounted for by global SE feelings while in the scenario, and vice versa, using analysis of covariance (ANCOVA). This was done in all studies. Baseline state SR and SE were also adjusted for by being added as covariates in MANOVA and in anticipated SR and SE ANCOVAs, respectively, because they related to the outcome measures. Correlations between measures for all studies can be found in Supplemental Appendix D.

Study 1a: Results and Discussion

Preliminary analyses

There was no difference between the conditions in participants’ age, sex, and ethnicity, and SE or SR (trait or state).

Manipulation checks

Table 2 presents descriptive statistics for manipulation check and outcome measures. The 2 × 2 analyses of variance (ANOVAs) confirmed participants rated themselves more morally principled in the HM versus LM condition, F(1, 149) = 87.81, p < .001 $(η_{p}^{2} = . 37)$ , and felt more accepted by peers in the positive versus negative IA condition, F(1, 149) = 301.10, p < .001 $(η_{p}^{2} = . 67)$ .

Table 2.

Effects of Adherence to Morals and IA on Self-Respect and Self-Esteem Ratings (and Moral and IA Ratings).

Morals	IA	Moral ratings			IA ratings			Self-respect			Self-esteem
Morals	IA	M	SD	N	M	SD	N	M	SD	N	M	SD	N
High	Positive	8.20	1.86	41	8.50	2.28	42	5.63	1.15	41	5.02	1.60	42
	Negative	5.38	2.27	37	2.69	2.38	36	4.37	1.27	36	3.33	1.59	36
	Total	6.86	2.49	78	5.82	3.72	78	5.04	1.36	77	4.24	1.80	78
Low	Positive	4.43	2.32	37	8.70	1.65	37	3.66	1.36	36	4.03	1.61	37
	Negative	2.66	2.11	38	2.58	2.10	38	2.77	1.69	38	2.82	1.71	38
	Total	3.53	2.37	75	5.60	3.61	75	3.20	1.59	74	3.41	1.76	75
Total	Positive	6.41	2.81	78	8.59	2.00	79	4.71	1.59	77	4.56	1.67	79
	Negative	4.00	2.57	75	2.64	2.22	74	3.55	1.69	74	3.07	1.66	74
	Total	5.23	2.94	153	5.71	3.65	153	4.14	1.74	151	3.84	1.82	153

Note. IA = interpersonal appraisal.

Higher moral ratings were also found in the positive versus negative IA condition, F(1, 149) = 44.04, p < .001 $(η_{p}^{2} = . 23)$ , with no interaction between factors. Because of this finding, results are also presented for the effect of IA after controlling for moral ratings in anticipated SR and SE ANCOVAs. This finding raises the possibility that social (dis)approval by peers can intensify self-perceptions of (im)moral and (un)principled behavior.

Main results

A 2 × 2 multivariate analysis of covariance (MANCOVA) showed a significant multivariate main effect of adherence to morals, F(2, 127) = 27.85, p < .001, Wilks’s λ = .70, $η_{p}^{2} = . 31$ , and IA, F(2, 127) = 19.23, p < .001, Wilks’s λ = .77, $η_{p}^{2} = . 23$ , and no interaction.

Adherence to morals

Univariate tests showed a significant effect of adherence to morals on SR, F(1, 128) = 52.73, p < .001 $(η_{p}^{2} = . 29)$ , CI_95% = [1.222, 2.137], and SE, F(1, 128) = 9.17, p = .003 $(η_{p}^{2} = . 07)$ , CI_95% = [0.262, 1.249], in the expected direction. The effect on SR remained after controlling for anticipated SE in the scenario, F(1, 143) = 57.08, p < .001 $(η_{p}^{2} = . 29)$ , CI_95% = [1.009, 1.723]. Hypothesis 1 that adherence to morals will influence self-respect was, therefore, supported. Moreover, the effect of adherence to morals on SE was reversed and no longer significant after controlling for SR, F(1, 129) = 2.24, p = .137 $(η_{p}^{2} = . 02)$ , CI_95% = [–0.807, 0.112].

A mediation analysis using regression (see Table 3), followed by bootstrap procedures based on 5,000 random samples with replacement from the full sample to construct bias-corrected confidence intervals (MacKinnon, Fairchild, & Fritz, 2007), adjusting for the IA manipulation and baseline state SR and SE, showed the size of the indirect effect (1.12) to differ significantly from zero (0.74, 1.58), indicating that anticipated SR mediated the effect of adherence to morals on anticipated SE, thus supporting Hypothesis 2.

Table 3.

Hierarchical Regression Analysis: Test of Mediation.

	Anticipated self-respect			Anticipated self-esteem
	Step 1			Step 2
	B	t	B	t	B	t
Adherence to morals	1.67	7.29***	0.76	3.05**	−0.36	−1.53
Anticipated self-respect					0.66	8.84***
Interpersonal appraisals	1.15	4.93***	1.52	6.05***	0.76	3.50***
Baseline state self-esteem	0.02	1.91	0.04	3.80***	0.03	3.27**
Baseline state self-respect	0.30	1.73	0.27	1.45	0.07	0.48

Note. Adding anticipated self-respect increased the amount of variance explained for self-esteem from R² = .45 to R² = .64.

p < .05. **p < .01. ***p < .001.

IAs

Univariate tests showed a significant effect of IA on SE, F(1, 128) = 36.57, p < .001 $(η_{p}^{2} = . 22)$ , CI_95% = [0.682, 1.607], in the expected direction. Hypothesis 3 that IA will influence self-esteem was, therefore, supported. This effect remained significant after controlling for moral ratings, F(1, 131) = 10.70, p = .001 $(η_{p}^{2} = . 08)$ , CI_95% = [0.349, 1.417], or anticipated SR, F(1, 129) = 11.61, p = .001 $(η_{p}^{2} = . 08)$ , CI_95% = [0.303, 1.144], alongside anticipated SR, F(1, 129) = 81.31, p < .001 $(η_{p}^{2} = . 39)$ . This pattern of results suggests that both SR and IA uniquely influence SE.

A significant effect on SR was also found, F(1, 128) = 23.99, p < .001 $(η_{p}^{2} = . 16)$ , CI_95% = [0.682, 1.607], with higher anticipated SR in the positive versus negative IA condition. However, the effect became small after controlling for moral ratings, F(1, 143) = 4.83, p = .030 $(η_{p}^{2} = . 03)$ , CI_95% = [0.051, 0.964], and nonsignificant after controlling for anticipated SE, F(1, 143) = 3.02, p = .071 $(η_{p}^{2} = . 02)$ , CI_95% = [–0.030, 0.745] (p = .298, $η_{p}^{2} = . 01$ , when controlling for both variables). The effect of IA on SR was, therefore, explained by differences in feelings of being morally principled and anticipated SE, as opposed to being directly influenced by feelings of acceptance or rejection.

Study 1b: Method

Study 1b explored the relationship between SR and another important basis for SE: academic competence, to further explore differences in the bases for SR and SE. The study also aimed to replicate the effect of adherence to morals on participants’ anticipated SR and SE using a different moral behavior (cheating as opposed to lying) and a more detailed scenario to encourage stronger immersion in the scenario.

Participants

One hundred seventy-nine English-speaking Western university or preuniversity students⁵ (all aged above 18 years) were recruited through Prolific Academic (https://www.prolific.co/)—an international online participant recruitment platform, receiving £1.50 each. Students were purposely recruited to facilitate more “real world” responses to the academic competence manipulation (see Evans et al., 2015). Thirty students (16.8%) failed end of questionnaire scenario comprehension questions or did not engage with the questionnaire (see Supplemental Appendix A for details of exclusion for nonengagement for all studies), leaving 148 participants (see Table 1 for sample characteristics and Supplemental Appendix B for additional sample details).

Materials and procedure

Scenarios were piloted with several university students for clarity, credibility, and realism and modified accordingly. Participants were asked to imagine they had recently started a new course at university they really wanted to succeed in. According to the scenarios (see Supplemental Appendix A for verbatim), they (a) were confident they would do well, and were already achieving high grades or (b) found the course extremely difficult and were already struggling to achieve decent grades (high competence [HC] and low competence [LC] conditions, respectively). Because of unforeseen circumstances, they had to leave one of their assignments to the very last minute, and a friend from another university kindly lends them his or her essay to give them a bit of a head start. They either (a) decide to submit a close copy of their friend’s work (without them knowing), although it goes against their principles, or (b) decide not to, because it goes against their principles (LM and HM conditions, respectively). Following the incident, they do very well (or struggle) on remaining assignments and their overall performance was very good (or poor; HC and LC conditions, respectively).

Manipulation checks

Participants indicated (1a) how principled and moral they were feeling on a scale from 1 (very immoral) to 10 (very moral) and (1b) how high their moral status was on a scale from 1 (very low) to 10 (very high; r_SB = .96). They also indicated (2a) how academically competent they were feeling from 1 (very incompetent) to 10 (very competent) and (2b) how confident they were in their academic abilities from 1 (not confident) to 10 (very confident; r_SB = .97).

Outcome and baseline measures

These were the same as in Study 1a with similar internal reliabilities, except that the baseline state SR measure was the same as the outcome measure but with the instruction to respond at that moment (α = .81).

Comprehension checks

Questions were included to check participants had understood and read fully the scenario given its length (see Supplemental Appendix A).

Subsidiary measures

The Contingency of Self-Worth Virtue subscale (Crocker et al., 2003) was administered to explore whether the relationship of morally principled behavior with SR differed according to level of virtue-contingent SE. This was of interest because theory on contingencies of self-worth (Crocker et al., 2003) would predict a stronger effect of morally principled behavior on SE for people with higher (vs. lower) virtue-contingent SE, but morally principled behavior is theorized to be more defining of SR than SE (fuller detail is provided in Supplemental Appendix C).

Analytical approach

A 2 × 2 MANOVA was used in Study 1a because anticipated scenarios SR and SE were strongly correlated (ρ = .79). The correlation was high, still 38% of the variance in state SR and SE was available to be uniquely explained for each. A similarly high correlation was also observed between baseline state performance SE (an established facet) and RSE (RSES global SE; see Supplemental Appendix D Table 2a for correlations between all measures). Baseline state SR was added as a covariate in MANOVAs and experienced SR ANCOVAs but not baseline state SE because only the first related to the outcome.

Study 1b: Results and Discussion

Preliminary analyses

The conditions did not differ in terms of participants’ demographic characteristics, SE or SR (trait or state). There were more White (vs. non-White) participants in the HM condition, χ²(1,147) = 10.16, p = .001, but ethnicity did not affect any of the baseline or outcome variables.

Manipulation checks

Table 4 presents descriptive statistics for manipulation check and outcome measures.

Table 4.

Effects of Adherence to Morals and Academic Competence on Self-Respect and Self-Esteem Ratings (and Moral and Competence Ratings).

Morals	Competence	Moral ratings			Competence ratings			Self-respect			Self-esteem
Morals	Competence	M	SD	N	M	SD	N	M	SD	N	M	SD	N
High	High	8.76	1.31	37	8.22	1.59	37	5.46	1.00	36	5.69	1.41	36
	Low	6.99	2.14	37	2.71	1.64	37	4.24	1.44	37	3.75	1.59	36
	Total	7.87	1.97	74	5.46	3.20	74	4.85	1.38	73	4.72	1.79	72
Low	High	3.27	2.10	37	5.18	2.35	37	3.00	1.16	36	3.19	1.47	37
	Low	2.59	1.55	37	2.08	1.25	37	2.41	1.16	37	2.19	1.31	37
	Total	2.93	1.86	74	3.63	2.44	74	2.70	1.19	73	2.69	1.47	74
Total	High	6.01	3.26	74	6.70	2.51	74	4.23	1.64	72	4.42	1.91	73
	Low	4.79	2.88	74	2.40	1.48	74	3.32	1.59	74	2.96	1.65	73
	Total	5.40	3.13	148	4.55	2.98	148	3.77	1.67	146	3.69	1.92	146

Moral ratings

A 2 × 2 ANOVA confirmed that participants felt more morally principled in the HM versus LM condition, F(1, 144) = 275.81, p < .001 $(η_{p}^{2} = . 66)$ .

Higher moral ratings were also found in the HC versus LC condition, F(1, 144) = 16.91, p < .001 $(η_{p}^{2} = . 11)$ . In view of this, results are also presented for the effect of academic competence after controlling for moral ratings in anticipated SR and SE ANCOVAs.

Academic competence ratings

A 2 × 2 ANOVA confirmed that participants felt more academically competent in the HC versus LC condition, F(1, 144) = 208.80, p < .001 $(η_{p}^{2} = . 59)$ .

Higher competence ratings were also found in the HM versus LM condition, F(1, 144) = 38.75, p < .001 $(η_{p}^{2} = . 21)$ . However, a significant adherence to morals × academic competence interaction, F(1, 144) = 16.35, p < .001 $(η_{p}^{2} = . 10)$ , indicated when followed up, a significant effect of adherence to morals in the HC condition only, p < .001. The effect of academic competence was significant in the HM, p < .001, and LM conditions, p < .001.

Main results

A 2 × 2 MANCOVA showed a significant multivariate main effect of adherence to morals, F(2, 139) = 68.60, p < .001, Wilks’s λ = .50, $η_{p}^{2} = . 50$ , and academic competence, F(2, 139) = 21.56, p < .001, Wilks’s λ = .76, $η_{p}^{2} = . 24$ , and no interaction.

Adherence to morals

Univariate tests showed a significant effect of adherence to morals on anticipated SR, F(1, 140) = 129.87, p < .001 $(η_{p}^{2} = . 48)$ , CI_95% = [1.803, 2.560], and SE, F(1, 140) = 82.29, p < .001 $(η_{p}^{2} = . 37)$ , CI_95% = [1.642, 2.558], in the expected direction. The effect on anticipated SR remained after controlling for anticipated SE, F(1, 139) = 34.95, p < .001 $(η_{p}^{2} = . 20)$ , CI_95% = [0.768, 1.539], and also competence ratings, F(1, 138) = 32.16, p < .001 $(η_{p}^{2} = . 19)$ , CI_95% = [0.723, 1.498]. Hypothesis 1 that adherence to morals will influence self-respect was, therefore, supported. Moreover, the effect of adherence to morals on SE was no longer significant after adjusting for anticipated SR, F(1, 140) = 2.79, p = .097 $(η_{p}^{2} = . 02)$ , CI_95% = [–0.078, 0.924], and became even weaker after also adjusting for academic competence ratings, F(1, 139) = 1.32, p = .253 $(η_{p}^{2} = . 01)$ , CI_95% = [–0.201, 0.758].

A mediation analysis, adjusting for baseline state SR and academic competence manipulation supported anticipated SR to mediate the effect of adherence to morals on anticipated SE (indirect effect = 1.59, CI_95% = [1.15, 2.10] > 0), and Hypothesis 2 (see Table 5 for the regression coefficients).

Table 5.

Hierarchical Regression Analysis: Test of Mediation.

	Anticipated self-respect			Anticipated self-esteem
	Step 1			Step 2
	B	T	B	t	B	t
Adherence to morals	2.18	11.33***	2.10	8.97***	0.51	1.95
Anticipated self-respect					0.73	8.90***
Academic competence	0.94	4.91***	1.49	6.36***	0.80	3.93***
Baseline state self-respect	0.26	3.74***	0.31	3.70***	0.12	1.73

Note. Adding anticipated self-respect increased the amount of variance explained for self-esteem from R² = .48 to R² = .67. The pattern of results was similar after controlling for competence ratings with an indirect effect of 1.00 (0.61, 1.42).

p < .05. **p < .01. ***p < .001.

Academic competence

Univariate tests showed a significant effect of academic competence on anticipated SE, F(1, 140) = 41.46, p < .001 $(η_{p}^{2} = . 23)$ , CI_95% = [1.031, 1.945]. Hypothesis 4 that academic competence will influence SE was, therefore, supported. This effect remained after adjusting for moral ratings, F(1, 141) = 18.75, p < .001 $(η_{p}^{2} = . 12)$ , CI_95% = [0.483, 1.294], or anticipated SR, F(1, 140) = 14.18, p < .001 $(η_{p}^{2} = . 09)$ , CI_95% = [0.361, 1.160], alongside anticipated SR, F(1, 145) = 91.39, p < .001 $(η_{p}^{2} = . 40)$ . This pattern of results suggests that both SR and academic competence uniquely influence SE.

A significant effect on anticipated SR was also found, F(1, 140) = 24.45, p < .001 $(η_{p}^{2} = . 15)$ , CI_95% = [0.567, 1.323], with higher anticipated SR in the HC versus LC condition. However, the effect became smaller after controlling for moral ratings, F(1, 140) = 7.56, p = .007 $(η_{p}^{2} = . 05)$ , CI_95% = [–0.391, 0.620], and disappeared after controlling for anticipated SE, F(1, 139) = 1.51, p = .221 $(η_{p}^{2} = . 01)$ , CI_95% = [–0.132, 0.564]. The main effect of competence was, therefore, explained by differences in feelings of being morally principled and anticipated SE, as opposed to directly influenced by feelings of competence.

Subsidiary analysis

A test of moderation did not find a differential relationship of morally principled behavior with SR according to virtue-contingent self-esteem levels after adjusting for SE, but showed a stronger relationship of morally principled behavior with SE for participants who were higher versus lower on virtue-contingent self-esteem, which was in line with past literature. The full results are presented in Supplemental Appendix C.

In sum, Studies 1a and 1b provided support for the hypotheses and for SR to be a component of SE tied to moral integrity, distinct from IA, and academic competence. Adherence to morals was found to influence anticipated SR (over and above anticipated SE), and anticipated SR mediated the effect of adherence to morals on anticipated SE. In contrast, IA and academic competence no longer influenced anticipated SR after controlling for ratings of being morally principled and anticipated SE but influenced SE independently of SR. The studies portrayed widely endorsed moral principles; however, there could be variations in how people perceive immoral behaviors as going against their principles (Liu & Ditto, 2013). Study 2 aimed to generalize the findings relating to adherence to morals and academic competence by asking participants to respond to personally selected situations they had experienced.

Study 2

Study 2 aimed to provide additional support for the conclusions of Studies 1a and 1b and enhance ecological validity by asking participants to relive situations in their life when they had (or had not) compromised on their morals and did particularly well (or badly) academically and indicate their SR and SE at that moment in time. The study played an important role in informing Study 3 and provided some useful results in relation to academic competence specifically (as opposed to broader task-related competence) but presented a number of difficulties, and is, therefore, reported here in less detail (additional details can be found in Supplemental Appendix E).

Although the study originally planned to use a 2 (adherence to morals: high/low) × 2 (academic competence: high/low) design, a pilot study to test the task’s feasibility found that asking participants to think back to a situation when they had performed particularly well on an academic task but had compromised on their morals disrupted the academic competence manipulation. As a result, the study tested the influence of adherence to morals and competence on SR and SE by comparing the HM and LM conditions when competence was low (HM-LC/LM-LC), and the HC and LC conditions when adherence to morals was high (HM-HC/HM-LC). Participants were pseudo-randomly allocated to one of these three conditions and the study was analyzed in two parts.

Methods

Of 259 English-speaking Western university students recruited through Prolific Academic for £1.50, and through RPS, 178 recalled a relevant situation (69%): 68 (out of 83) an HM-LC situation, 52 (out of 105) an LM-LC situation,⁶ and 58 (out of 71) an HM-HC situation (see Table 1 for sample characteristics).

Participants were asked in an online questionnaire to relive a scenario/experience in their life in an academic setting (e.g., school, college, university) in which they had (a) performed particularly badly on an academic/scholarly task and had compromised on their morals (LM-LC), (b) performed particularly badly on an academic/scholarly task and had not compromised on their morals (HM-LC), or (c) performed particularly well on an academic/scholarly task and had not compromised on their morals (HM-HC), and were provided with examples to help them recall a specific experience, such as cheating, being unfair, or making false excuses in the LM condition and opposite behaviors in the HM condition (see Supplemental Appendix A). Participants were instructed to take 3 min to relive the scenario/experience, including associated emotions, and to describe it.

Manipulation checks, outcome and baseline measures were identical to those used in Study 1b with similar internal reliabilities, except the SE measure included an additional item: “I feel good about myself” (see Rubin & Hewstone, 1998; rated from 1 = not very true of me, 7 = very true of me, r_SB = .91),⁷ and the baseline state SR measure included an additional item: I feel I have a high degree of self-respect (1 = strongly disagree to 7 = strongly agree) with standardized item scores averaged into one index (α = .83).

Parallel analysis and exploratory factor analysis (EFA) fit indices using principal axis factoring⁸ supported the five SR and SE outcome items representing two factors as opposed to one (see Table 6); the SR and SE items loaded on their respective factors (loadings between .51 and .90), with low factor loadings on the other factor (between .02 and .10), and a factor correlation of .77, indicating that the SR and SE measures represent two related but distinct constructs (see Supplemental Appendix F for full details).

Table 6.

Fit Indices for Factor Analysis Models.

Model	χ²(df), p	CFI	TLI	RMSEA (90% CI)	SRMR	BIC
Study 2: Outcome state measures
Two-factor EFA	χ²(1) = 4.99, .026	NA	.95	NA	NA	−0.38
One-factor EFA	χ²(5) = 116.69	NA	.72	NA	NA	89.86
Study 3: Outcome state measures
Two-factor EFA	χ²(19) = 54.18	NA	.97	.089 [0.061, 0.116]	NA	−50.03
One-factor EFA	χ²(27) = 234.32	NA	.88	.181 [0.158, 0.2]	NA	86.23
Two-factor E/CFA	χ²(19) = 39.29, .004	.99	.98	.071 [0.039, 0.102]	.02	3,950.33
One-factor E/CFA	χ²(27) = 146.31	.93	.91	.157 [0.133, 0.183]	.05	4,056.42
Study 3: Baseline state measures
Five-factor EFA	χ²(166) = 322.75	NA	.93	.066 [0.052, 0.073]	NA	−587.73
Six-factor EFA with global SE items	χ²(225) = 371.55	NA	.95	.057 [0.043, 0.061]	NA	−862.53
Exploratory bifactor	χ²(166) = 322.75	NA	.93	.066 [0.052, 0.073]	NA	−587.73
Bifactor CFA	χ²(118) = 248.13	.93	.91	.078 [0.065, 0.092]	.06	9,753.03
Bifactor CFA (SR items on g factor only)	χ²(121) = 340.62	.89	.86	.101 [0.088, 0.114]	.07	9,860.30
Studies 2 and 3: Baseline trait measures
Four-factor EFA	χ²(87) = 352.82	NA	.91	.078 [0.069, 0.085]	NA	−190.93

Note. TLI and CFI values ≥ .90 and RMSEA values < .08 indicate acceptable model fit, whereas TLI and CFI values ≥ .95 and RMSEA values ≤ .06 indicate good model fit; SRMR values close to 0 indicate perfect fit (Browne & Cudeck, 1992; Hu & Bentler, 1999). A lower BIC and χ²/df ratio indicate better model fit. p < .001 unless indicated otherwise. NA because not provided by the EFA or not considered robust because of the very low degrees of freedom (see Kenny et al., 2015). CFI = comparative fit index; TLI = Tucker–Lewis index; RMSEA = root mean square error of approximation; CI = confidence interval; SRMR = standardized root mean square residual; BIC = Bayesian information criterion; EFA = exploratory factor analysis; E/CFA = EFA within a CFA framework; CFA = confirmatory factor analysis; NA = not applicable; SE = self-esteem; SR = self-respect.

Results and Discussion

Although the LC adherence to morals and HM academic competence conditions did not differ according to participant demographics, recruitment source, trait, or overall state SE, LM-LC participants scored significantly lower on baseline performance state SE and trait and state SR compared with HM-LC participants, and HM-LC participants scored significantly higher on trait SR compared with HM-HC participants. These differences likely reflect theoretically consistent self-selection biases, whereby high SR individuals who gain a sense of worth from being principled are less likely to experience situations in which they compromise on their morals and more likely to resist the urge of compromising on their morals despite a potential need. Individuals with lower performance SE may feel more of a need for moral compromise to not do badly academically and protect their SE.

As a result, analyses testing the impact of adherence to morals controlled for baseline differences in performance SE and state SR⁹ and analyses testing the impact of academic competence controlled for baseline differences in trait SR in MANOVAs/ANOVAs by adding these as covariates. MANOVAs were used for main analyses because experienced SR and SE were strongly correlated (ρ = .72 and ρ = .61, respectively). Baseline overall state SE was also added as a covariate in the MANOVAs and experienced SE ANCOVAs testing the impact of academic competence¹⁰ but was too strongly related to performance state SE (ρ = .80) to be added as a covariate in analyses testing the impact of adherence to morals.

Tables 7 and 8 present descriptive statistics for manipulation check and outcome measures. The manipulations were effective. Participants rated themselves significantly more morally principled in the HM-LC versus LM-LC condition, F(1, 108) = 31.38, p < .001 $(η_{p}^{2} = . 23)$ , with no effect on academic competence ratings. Participants also rated themselves significantly more competent in the HM-HC versus HM-LC condition, F(1, 122) = 22.03, p < .001 $(η_{p}^{2} = . 15)$ , with no effect on moral ratings.

Table 7.

Effects of Adherence to Morals on Self-Respect and Self-Esteem Ratings (and Moral and Competence Ratings).

Morals	Moral ratings			Competence ratings			Self-respect			Self-esteem
Morals	M	SD	N	M	SD	N	M	SD	N	M	SD	N
High	7.46	1.68	67	5.47	2.58	68	4.77	1.36	68	4.29	1.59	68
Low	5.08	2.13	51	4.68	2.29	51	3.51	1.30	51	3.10	1.57	50
Total	6.43	2.22	118	5.13	2.48	119	4.23	1.47	119	3.79	1.68	118

Table 8.

Effects of Academic Competence on Self-Respect and Self-Esteem Ratings (and Moral and Competence Ratings).

Comp	Moral ratings			Competence ratings			Self-respect			Self-esteem
Comp	M	SD	N	M	SD	N	M	SD	N	M	SD	N
High	7.53	1.64	57	7.06	2.10	58	4.78	1.30	58	4.76	1.53	58
Low	7.46	1.68	67	5.47	2.58	68	4.77	1.36	68	4.29	1.59	68
Total	7.49	1.66	124	6.20	2.49	126	4.78	1.33	126	4.51	1.57	126

Using real situations, Study 2 was able to replicate the main findings of Studies 1a and 1b. Following significant MANOVA effects (see Supplemental Appendix E), the study showed a main effect of adherence to morals on SR (when academic competence was held constant and low), F(1, 108) = 22.76, p < .001 $(η_{p}^{2} = . 17)$ , CI_95% = [0.600, 1.453], which remained significant after adjusting for experienced SE, F(1, 107) = 12.15, p = .001 $(η_{p}^{2} = . 10)$ , CI_95% = [0.273, 0.995]. This effect mediated the effect of adherence to morals on SE, F(1, 108) = 9.53, p = .001 $(η_{p}^{2} = . 08)$ , CI_95% = [0.304, 1.393] (indirect effect = .77, CI_95% = [0.47, 1.14]; see Figure 1 for a schematic representation and Table 9 for all regression coefficients).

Figure 1.

Mediation model for Study 2.

Table 9.

Hierarchical Regression Analysis: Test of Mediation.

	Experienced self-respect			Experienced self-esteem
	Step 1			Step 2
	B	t	B	t	B	t
Adherence to morals	1.03	4.77***	0.85	3.09**	0.07	0.30
Experienced self-respect					0.75	7.58***
Baseline state self-respect	1.04	7.12***	0.91	4.89***	0.13	0.70
Performance self-esteem	−0.02	−0.77	0.02	0.70	0.03	1.42

Note. Adding self-respect increased the amount of variance explained for self-esteem from R² = .35 to R² = .58.

p < .05. **p < .01. ***p < .001.

Academic competence did not influence SR (when adherence to morals was held constant and high), F(1, 120) = 2.13, p = .148 $(η_{p}^{2} = . 02)$ , CI_95% = [–0.103, 0.678], with the effect close to zero after controlling for experienced SE, F(1, 121) = 0.06, p = .803 $(η_{p}^{2} = . 001)$ , CI_95% = [–0.397, 0.308], but influenced SE, F(1, 120) = 11.73, p = .001 $(η_{p}^{2} = . 09)$ , CI_95% = [0.346, 1.295], independently of SR, F(1, 115) = 9.40, p = .003, $(η_{p}^{2} = . 08)$ , CI_95% = [0.230, 1.07]. Participants often drawing on experiences of cheating to describe moral compromise in an academic setting made it difficult to study the impact of competence in the low morals condition, and baseline differences in SR and performance SE emerged. Study 3 was carried out to address these limitations.

Study 3

Study 3 aimed to replicate Study 2 using real situations but focusing on more broadly defined task-related competence to test the full 2 (adherence to morals: high [HM]/low [LM]) × 2 (competence: high [HC]/low [LC]) design. Additional measures were also put in place to reduce selection biases leading to differences in baseline SR and performance SE between conditions.

Methods

Participants, materials, and procedure

Participants were asked in an online questionnaire to recall and relive a scenario/experience in their life in which (1) they had performed particularly badly (e.g., at work, school, or other activity) despite trying their best and (1a) had compromised on their morals (LM-LC condition) or (1b) had not compromised on their morals (HM-LC condition), or (2) they had performed particularly well and (2a) had compromised on their morals but still felt competent (LM-HC condition) or (2b) had not compromised on their morals (HM-HC). Of 241 English-speaking Western adults recruited through Prolific Academic for £1.65, 204 recalled a relevant situation (85%): 51 (out of 62) an HM-LC, 51 (out of 59) an LM-LC, 51 (out of 57) an HM-HC, 51 (out of 63) an LM-HC situation. An older nonstudent sample was recruited to ensure participants had a wider range of experiences to draw from (see Table 1 for sample characteristics, Supplemental Appendix B for additional details).

Participants were provided with a broad range of examples of situations to facilitate retrieval of a relevant experience and reduce selection biases, such as lying or concealing information, behaving unfairly or being selfish, adopting unethical organizational practices in the LM conditions (see Supplemental Appendix A). Participants were also told most people had experienced at least one such situation in their lives and to think back to different periods in their life when this may have happened to encourage deeper self-reflection, and in the LM conditions, that this may not be indicative of the kind of person they are in general to reduce defensive reactions, particularly from high SR individuals who do not view themselves as immoral. Participants were instructed to take 3 min to think of and relive the scenario/experience, including associated emotions, and to describe it.

Manipulation checks

The adherence to morals manipulation check was the same as in Study 2 (with the addition of “at that time” at the end of the statement; r_SB = .94). For the competence manipulation check, participants indicated how high they felt their performance was at the time from 1 (very low) to 10 (very high) and how competent they felt at the time from 1 (very incompetent) to 10 (very competent; r_SB = .90).

Outcome measures

The measures were the same as in Study 2, except the SR measure included an additional item: “I have little respect for myself,” and the SE measure included an additional three items adapted from the General Self-Concept subscale of the Self-Description Questionnaire III (Marsh & O’Neill, 1984): “I have pretty positive feelings about myself,” “I have pretty negative feelings about myself,” and “I view myself positively,” rated on the same 7-point Likert-type scale as the other items.

Parallel analysis and EFA fit indices supported the nine items representing two factors as opposed to one; the SR and SE items loaded well on their respective factors (loadings of .50-.97) with a factor correlation of .79, indicating that the SR and SE measures represent two related but distinct constructs (see Table 6 and Supplemental Appendix F for full details). Moreover, an EFA within a confirmatory factor analysis (CFA) framework (E/CFA),¹¹ which permits to statistically compare model fit, using maximum likelihood robust (MLM) estimator, showed an acceptable to good fit for the two-factor model distinguishing between SR and SE items (see Supplemental Appendix F for the factor loadings), which was significantly better than the one-factor model, χ² difference = 82.21, p < .001. These findings support the SR and SE items used in Studies 1a, 1b, and 2 tapping two distinct constructs.

Nevertheless, an even better fit for a three-factor model suggested the measures could be improved further, by removing the SR item “I wish I could have more respect for myself” and the SE item “I feel good about myself,” which appeared to be less consistent or pure markers.¹² The first item loaded most strongly on a third factor and the second item cross loaded on the SR factor (factor loading of .69 on the SE factor and .43 on the SR factor; see Table 2 in Supplemental Appendix F). The three-item SR factor (α = .90) correlated at .78 with the four-item SE factor (α = .95).

Baseline measures

The same measures were used as in Study 1b with similar internal reliabilities, except state SR was measured with the same three items used for the revised SR outcome measure (α = .92; data were also collected for the item “I wish I could have more respect for myself” at baseline).

Analytical approach

A 2 × 2 MANOVA was used because experienced SR and SE outcome measures were strongly correlated (ρ = .77). Despite the high correlation, 41% (1-.77²) of the variance in state SR and SE remains available to be uniquely explained for each.

Results and Discussion

Preliminary analyses

The conditions did not differ in terms of background or baseline characteristics. Baseline state SR and SE were added as covariates in MANOVAs and experienced SR and SE ANCOVAs, respectively, because they were significantly correlated with these outcomes (see Supplemental Appendix D Table 4a).

Manipulation checks

Table 10 presents descriptive statistics for manipulation checks and outcome measures.

Table 10.

Effects of Adherence to Morals and Academic Competence on Self-Respect and Self-Esteem Ratings (and Moral and Competence Ratings).

Morals	Competence	Moral ratings			Competence ratings			Self-respect			Self-esteem
Morals	Competence	M	SD	N	M	SD	N	M	SD	N	M	SD	N
High	High	8.57	1.21	50	7.78	1.33	51	5.39	1.22	51	5.35	1.20	51
	Low	7.03	2.09	51	4.28	2.21	50	4.75	1.51	51	3.82	1.60	50
	Total	7.79	1.87	101	6.05	2.53	101	5.07	1.40	102	4.59	1.60	101
Low	High	5.15	2.08	51	6.94	1.62	50	4.19	1.43	50	3.94	1.26	51
	Low	3.18	1.99	51	3.25	2.06	51	3.36	1.57	51	2.80	1.49	51
	Total	4.17	2.26	102	5.07	2.62	101	3.77	1.55	101	3.37	1.49	102
Total	High	6.84	2.42	101	7.37	1.53	101	4.79	1.45	101	4.65	1.42	102
	Low	5.11	2.80	102	3.76	2.19	101	4.06	1.69	102	3.30	1.62	101
	Total	5.97	2.75	203	5.56	2.61	202	4.42	1.61	203	3.98	1.66	203

Moral ratings

A 2 × 2 ANOVA confirmed participants rated themselves more morally principled in the HM versus LM condition, F(1, 199) = 188.71, p < .001 $(η_{p}^{2} = . 49)$ .

Higher moral ratings were also found in the HC versus LC condition, F(1, 199) = 43.82, p < .001 $(η_{p}^{2} = . 18)$ , as in Study 1b, with no significant interaction. We, therefore, also present results for the effect of competence after controlling for moral ratings in anticipated SR and SE ANCOVAs.

Competence ratings

A 2 × 2 ANCOVA confirmed participants rated themselves significantly more competent in the HC versus LC condition, F(1, 198) = 193.44, p < .001 $(η_{p}^{2} = . 49)$ .

Higher competence ratings were also found in the HM versus LM condition, F(1, 195) = 13.18, p < .001 $(η_{p}^{2} = . 06)$ . Results are, therefore, also presented for the effect of adherence to morals after controlling for competence ratings in anticipated SR and SE ANCOVAs.

Main results

A 2 × 2 MANCOVA showed a significant multivariate main effect of adherence to morals, F(1, 187) = 26.34, p < .001, Wilks’s λ = .78, $η_{p}^{2} = . 22$ , and competence, F(1, 187) = 24.94, p < .001, Wilks’s λ = .79, $η_{p}^{2} = . 21$ , and no interaction.

Adherence to morals

Univariate tests showed a significant effect of adherence to morals on experienced SR, F(1, 188) = 51.45, p < .001 $(η_{p}^{2} = . 22)$ , CI_95% = [0.963, 1.693], and SE, F(1, 188) = 33.74, p < .001 $(η_{p}^{2} = . 15)$ , CI_95% = [0.725, 1.472], in the expected direction. The effect on SR remained after adjusting for experienced SE, F(1, 194) = 16.39, p < .001 $(η_{p}^{2} = . 08)$ , CI_95% = [0.302, 0.877], and competence ratings also, F(1, 191) = 16.29, p < .001 $(η_{p}^{2} = . 08)$ , CI_95% = [0.304, 0.886]. Hypothesis 1 that adherence to moral principles will influence SR was, therefore, supported. In contrast, the effect on SE was no longer significant after adjusting for experienced SR, F(1, 189) = 3.72, p = .055 $(η_{p}^{2} = . 02)$ , CI_95% = [–0.007, 0.598], and became even weaker after also adjusting for competence ratings, F(1, 186) = 2.24, p = .136 $(η_{p}^{2} = . 01)$ , CI_95% = [–0.07, 0.508].

A mediation analysis, adjusting for baseline state SR and SE and the competence manipulation supported experienced SR mediating the effect of adherence to morals on experienced SE (indirect effect = .93, CI_95% = [0.66, 1.24] > 0), and Hypothesis 2 (see Figure 2 for a schematic representation and Table 11 for all regression coefficients).

Figure 2.

Mediation model for Study 3.

Table 11.

Hierarchical Regression Analysis: Test of Mediation.

	Experienced self-respect			Experienced self-esteem
	Step 1			Step 2
	B	t	B	t	B	t
Adherence to morals	1.33	7.20***	1.09	5.80***	0.17	1.06
Anticipated self-respect					0.70	12.81***
Competence	0.69	3.78***	1.30	6.95***	0.82	5.76***
Baseline state self-esteem	0.01	1.45	0.04	3.64***	0.03	3.60***
Baseline state self-respect	0.11	3.86***	0.01	0.46	−0.07	−2.86**

Note. Adding anticipated self-respect increased the amount of variance explained for self-esteem from R² = .40 to R² = .68. The pattern of results was similar after controlling for competence ratings with an indirect effect of .69 (.45, .98). Social approval ratings had little effect on the pattern of results (see Supplemental Appendix A).

p < .05. **p < .01. ***p < .001.

Competence

Univariate tests showed a significant effect of competence on experienced SE, F(1, 188) = 48.29, p < .001 $(η_{p}^{2} = . 20)$ , CI_95% = [0.933, 1.673], in the expected direction, supporting Hypothesis 4 that competence will influence self-esteem. This effect remained after adjusting for moral ratings, F(1, 189) = 19.27, p < .001 $(η_{p}^{2} = . 09)$ , CI_95% = [0.485, 1.275], or experienced SR, F(1, 189) = 36.42, p < .001 $(η_{p}^{2} = . 16)$ , CI_95% = [0.582, 1.147], alongside experienced SR, F(1, 186) = 150.42, p < .001 $(η_{p}^{2} = . 44)$ . This pattern of results indicates that both SR and competence uniquely influence SE.

A significant effect on experienced SR was also found, F(1, 188) = 14.18, p < .001 $(η_{p}^{2} = . 07)$ , CI_95% = [0.329, 1.053], with higher SR in the HC versus LC condition. However, the effect was no longer significant after controlling for moral ratings, F(1, 194) = 2.27, p = .134 $(η_{p}^{2} = . 01)$ , CI_95% = [–0.09, 0.648], or experienced SE, F(1, 194) = 1.15, p = .286 $(η_{p}^{2} = . 01)$ , CI_95% = [–0.133, 0.450].

In sum, Study 3 replicated the findings of Study 2 using the full 2 × 2 design and in the absence of baseline differences between conditions.

Post hoc analyses: Further exploring the relationship between self-respect and self-esteem through factor analysis

Consistent with the main study results, factor analyses of baseline state and trait SR and SE data supported SR to be distinct from global SE and other specific self-evaluations or facets.

EFA of SSES and state SR items supported a five-factor solution, which showed SR to be a separate factor from social approval, appearance, and performance SE, and a fifth factor corresponding to self-deprecation. Correlations were .35, .39, .59, and .50 between SR and each factor, respectively, showing SR to be related to, yet distinct from, the other SE facets investigated. After adding four baseline state global SE items to the EFA (the same as those making up the outcome measure that were also taken at baseline), an additional positive global SE factor emerged on which the global SE items loaded, the SR factor became more clearly defined, and the SR–performance SE correlation decreased to .54, with the correlation with global SE being .54 (see Table 6 for EFA fit indices and Supplemental Appendix F for fuller details and results of all analyses in this section).

In contrast, an EFA of the 10 RSES items; the seven descriptive trait SRS items relating to being moral, principled, and honorable; and two global appraisal trait SR items (I have a lot of respect for myself” and “I feel I have a high degree of self-respect”) in combined samples from Studies 2 and 3 (N = 518) supported a four-factor solution that showed the descriptive SRS and global appraisal SR items to load on the same factor, which was distinct from global SE, itself divided into overall self-satisfaction and two other factors interpreted by Tafarodi and Swann (1995) as reflecting the “self-competence” (i.e., feeling confident, capable, and effective) and “self-liking” (i.e., social value ascribed) dimensions of global SE. The SR factor correlated at .67, .55, and .55 with each SE factor, respectively. This analysis further supports SR’s close link to attributes of being moral and principled.

An exploratory bifactor analysis, followed by a confirmatory bifactor analysis (see Reise, 2012), also supported SR being a subdomain of global SE distinct from other facets. In line with past research showing a bifactor model to provide the best factor representation for the SSES (see McCain, Jonason, Foster, & Campbell, 2015), an exploratory bifactor model showed the state SR items to behave as a fourth grouping factor, alongside social approval, appearance SE, and performance SE grouping factors, that explained additional common variance to that explained by the general (g) factor (here global SE or evaluation toward the self, underpinning all items) like the other three SE facets. A reduced bifactor CFA model¹³ presented an acceptable fit, with all SR items loading significantly on the SR grouping factor, suggesting notable amounts of unique variance for the SR items (Gomez et al., 2015), and was a significantly better fit than a bifactor model where the SR items loaded on the general factor only, χ² difference = 69.87, p < .001, or bifactor models in which the SR items were specified to load on the general and performance SE, appearance SE, or social approval grouping factors instead. SR thus shares communalities with the general factor global SE while also exhibiting specific elements that are separate and unique from it. This evidence, combined with Studies 1 to 3 showing SR to have distinctive association patterns from SE, further supports SR as a distinct construct in its own right (Gomez et al., 2015).

General Discussion

Although the term “self-respect” is commonly used in everyday language, the concept has been the object of very little psychological research. Yet, it is recognized to be an important aspect of self-esteem (e.g., Kernis, 2003; Rosenberg, 1965) and believed to be critical to living a satisfying and flourishing life (Dillon, 2010). This research sought to provide an empirically supported understanding of the bases of self-respect, and how it differs from self-esteem, to develop our understanding of this important concept. Findings support the conceptualization of self-respect as a component of self-esteem linked to morally principled conduct (Kristjansson, 2007; Kumashiro et al., 2002), rather than a feature of global self-esteem influenced by a range of diverse self-domains (Kernis, 2003).

Four experimental studies provide evidence that adherence to morals is a key feature of self-respect, with a significant effect of the adherence to morals manipulation on participants’ anticipated state self-respect when imagining themselves in scenarios and felt state self-respect in a past situation they relived. This is the first set of studies to provide direct evidence for the conceptualization of self-respect as linked to morally principled conduct, as put forth by philosophers and other authors (e.g., Kristjansson, 2007; Kumashiro et al., 2002; Roland & Foxx, 2003), thus establishing self-respect as likely to play an important role in the understanding of moral behavior. Study 3 unexpectedly provided additional support for this argument by showing that participants who recalled a situation in which they compromised on their morals had lower baseline self-respect than those who recalled a situation in which they did not, when performing badly academically. Future research should test this hypothesis more formally in view of research demonstrating the predictive value of trait self-esteem, when domain-specific self-esteem relevant to the outcome is assessed (Baumeister et al., 2003), and showing momentary feelings of moral self-worth to influence engagement in moral behaviors (Sachdeva, Iliev, & Medin, 2009). It should be noted that self-respect is conceptualized as involving a subjective judgment of honorable and moral character and behavior such that it may relate to generally agreed notions of moral behavior only when individuals embrace such notions.

An important finding in all studies was that self-respect mediated the effect of adherence to morals on self-esteem, furthering our understanding of the relationship between morally principled conduct and self-esteem, which has not benefited from much research (Jennings, Mitchell, & Hannah, 2015). Subsidiary analyses in Study 1b and post hoc analyses in Study 3 further supported moral integrity as being more defining of self-respect than self-esteem. In addition to moral behavior being associated with self-esteem because it allows individuals who value being moral to meet their “ought” or “ideal” selves (Jennings et al., 2015), the subsidiary analysis in Study 1b indicated self-evaluation as worthy of respect can also explain the link between moral behavior and self-esteem, even when one is not strongly invested in being moral per se. Self-respect involves an appraisal of oneself as worthy of high regard connected to feelings of dignity, honor, and being principled and is linked to, yet broader than moral self-approval (see Kumashiro et al., 2002; Luchies et al., 2010). For instance, Luchies et al. (2010) showed that self-respect diminished when forgiving a transgressor who had not made amends. Future research should investigate further the relationship between the moral self and self-respect, including factors facilitating an attitude of respect for oneself when engaging in morally principled behavior.

The influence of competence and IA on self-respect and self-esteem was also investigated. Studies 1b and 3 found academic or broader task-related competence to influence self-respect, however, only indirectly through feelings of being morally principled or anticipated/felt self-esteem. Study 2 showed no effect of academic performance per se on self-respect when adherence to morals was held constant. These findings suggest that self-respect does not appear to be based on intellectual qualities and talents/skills, such that its basis is narrower than that for interpersonal respect (Frei & Shaver, 2002; Hamilton & Fallot, 1974; Prestwich & Lalljee, 2009). Despite this, the findings suggest that the relationship between competence, self-respect, and self-esteem is complicated because competence could possibly modify perceptions of moral integrity, and vice versa.

Study 1 found a main effect of IA on self-respect, but the effect became small after adjusting for feelings of being morally principled and was no longer significant after adjusting for global self-esteem, whereas the effect of IA on self-esteem remained significant after adjusting for self-respect (as was the case for competence). Therefore, self-respect does not appear to be directly influenced by feelings of acceptance or rejection, although social (dis)approval by peers might function as a looking glass self (Cooley, 1902/1956), intensifying self-perceptions of being (im)moral and (un)principled, and indirectly influencing self-respect. The closer influence of IA on self-esteem than self-respect is consistent with social popularity being more strongly linked to liking than respect (Hamilton & Fallot, 1974).

Moreover, the studies indicated that self-respect influenced self-esteem independently of IA (Study 1a) or competence (Studies 1b, 2, and 3). Baseline post hoc analyses in Study 3 also demonstrated state self-respect to be a factor sufficiently distinct from performance, social and appearance self-esteem, as well as global self-esteem, suggesting that self-respect should be measured alongside these self-esteem domains to improve the prediction of self-esteem. The high bivariate correlations between state self-respect and global self-esteem may be related to principled behavior’s link to competence, enhanced self-efficacy, and greater confidence in being accepted by others associated with high self-respect. Also, the global appraisal nature of the self-respect items makes them liable to be strongly influenced by positive affectivity linked to self-esteem in the same manner as performance self-esteem also correlates highly with trait self-esteem as it contains global evaluative items such as “I feel like I’m not doing well.” On the whole, the findings support self-respect as a sense of worth derived from being morally principled, as opposed to competent or popular. This suggests that self-respect may be a more internal and controllable source of self-esteem and, therefore, more amenable to intervention as a trait (Crocker et al., 2003).

The research suffered from some limitations. The study used a single item to measure anticipated self-esteem in Studies 1a and 1b, yet, it has been well validated, performing similar to the RSES (Robins et al., 2001). Moreover, single items have been shown to demonstrate reasonable levels of reliability and validity, when not ambiguous, and assessing reasonably homogeneous unidimensional constructs as was the case here because perceptions of global self-esteem and self-respect were assessed (Postmes, Haslam, & Jans, 2013; Robins et al., 2001). Due to the lack of prior studies on which to base sample size power calculations, studies planned to recruit a minimum of 128 participants to detect a medium effect size (Cohen, 1992), but Study 2 fell a little short. However, power did not appear to be an issue because effect sizes were medium to large and detected, or were very small, indicating no or little meaningful effect to be detected.

An experimental design with set vignettes in Studies 1a and 1b afforded more control over the factors of interest, strengthening causal conclusions regarding their impact on self-respect and self-esteem, and was needed to separate the influence of interrelated sources of self-esteem. Moreover, it permitted investigation of immediate responses to undesirable behavior, in a potentially less threatening way, given the role-play involved, possibly facilitating their disclosure. However, responses in vignettes may differ to those in real life, although similarities have been found between hypothetical vignette and actual behavior (Evans et al., 2015), and some may believe that the ends justify immoral behavior. Therefore, Studies 2 and 3 were carried out to increase ecological validity by asking participants to select personally relevant situations they had experienced. This introduced some unpredicted but theoretically consistent differences between conditions in Study 2, notably in baseline self-respect, likely due to high self-respect individuals being better at resisting moral compromise when the need arises, such as when doing badly academically. Nonetheless, the main findings relating to adherence to morals and competence held after adjusting for these baseline differences in Study 2 and were replicated in Study 3, which was not subject to this potential bias, strengthening the conclusions.

Conclusion

The studies have helped advance our understanding of self-respect, a concept that has been neglected in the psychological literature and often used interchangeably with self-esteem. Consistent with theories that self-respect is an attitude of respect for oneself based on principled and honorable behavior, the experimental studies provided direct evidence for self-respect being influenced by moral integrity, and showed self-respect to mediate the effect of morally principled behavior on self-esteem. However, they did not support self-respect to be directly based on competence or social popularity. This work strengthens the view that self-respect is a specific self-evaluation that is more internal and as a trait likely to be more amenable to intervention and highlights its potentially important implications for the study of moral behavior. This work also suggests self-respect should be measured alongside other well-established self-esteem components to enhance prediction of self-esteem, although future research is needed to study the relationship between self-respect and other self-esteem domains not investigated here. Strengthening self-respect has the potential to enhance individual well-being while benefiting society at large.

Supplemental Material

Clucas_OnlineAppendix – Supplemental material for Understanding Self-Respect and Its Relationship to Self-Esteem

Supplemental material, Clucas_OnlineAppendix for Understanding Self-Respect and Its Relationship to Self-Esteem by Claudine Clucas in Personality and Social Psychology Bulletin

Footnotes

Acknowledgements

We thank Emma Bubb for assistance with data reporting and Dr. Astrid Schepman, Dr. Annie Scudds, Dr. Heather Wilkinson, Dr. Julian Lloyd, and Dr. Lindsay St Claire for their comments on earlier or later drafts of the article and/or assistance with the wording of materials. Some of the findings were presented at the 2015 BPS joint Developmental Section and Social Psychology Section Annual Conference.

Authors’ Note

For all experiments, all measures, conditions, and data exclusions are reported. Any exclusion made not detailed in the main text was minor and is detailed in the .

Declaration of Conflicting Interests

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

Funding

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

ORCID iD

Claudine Clucas

Supplemental Material

Supplemental material is available online with this article.

Notes

References

Asparouhov

Muthén

Morin

A. J. S.

(2015). Bayesian Structural Equation Modeling with cross-loadings and residual covariances: Comments on Stromeyer et al. Journal of Management, 41, 1561-1577. doi:10.1177/0149206315591075

Baumeister

R. F.

Campbell

J. D.

Krueger

J. I.

Vohs

K. D.

(2003). Does high self-esteem cause better performance, interpersonal success, happiness, or healthier lifestyles? Psychological Science in the Public Interest, 4, 1-44. doi:10.1111/1529-1006.01431

Brown

T. A.

(2014). Confirmatory factor analysis for applied research. New York, NY: Guilford Press.

Browne

M. W.

Cudeck

(1992). Alternative ways of assessing model fit. Sociological Methods and Research, 21, 230-258. doi:10.1177/0049124192021002005

Clucas

St Claire

(2017). How can respectfulness in medical professionals be increased? A complex but important question. Journal of Bioethical Inquiry, 14, 123-133. doi:10.1007/s11673-016-9758-5

Clucas

Wilkinson

(2017, May 5). The value of self-respect for moral and social behaviour: Development of a trait self-respect measure. Paper presented at the British Psychological Society Annual Conference, Brighton, UK. Available from https://www.bps.org.uk

Cohen

(1992). A power primer. Psychological Bulletin, 112, 155-159. doi:10.1037/0033-2909.112.1.155

Cooley

C. H.

(1956). The two major works of Charles H. Cooley: Social organization and human nature and the social order. Glencoe, IL: Free Press. (Original work published 1902)

Coopersmith

(1967). The antecedents of self-esteem. San Francisco, CA: W.H. Freeman.

10.

Crocker

Luhtanen

R. K.

Cooper

M. L.

Bouvrette

(2003). Contingencies of self-worth in college students: Theory and measurement. Journal of Personality and Social Psychology, 85, 894-908. doi:10.1037/0022-3514.85.5.894

11.

Dahlsgaard

Peterson

Seligman

M. E. P.

(2005). Shared virtue: The convergence of valued human strengths across culture and history. Review of General Psychology, 9, 209-213. doi:10.1037/1089-2680.9.3.203

12.

Dillon

R. S.

(2010). Respect. In Zalta

E. N.

(Ed.), Stanford encyclopedia of philosophy. Retrieved from http://plato.stanford.edu/entries/respect

13.

Eisinga

Te Grotenhuis

Pelzer

(2013). The reliability of a two-item scale: Pearson, Cronbach, or Spearman-Brown? International Journal of Public Health, 58, 637-642. doi:10.1007/s00038-012-0416-3

14.

Evans

S. C.

Roberts

M. C.

Keeley

J. W.

Blossom

J. B.

Amaro

C. M.

Garcia

A. M.

. . . Reed

G. M.

(2015). Vignette methodologies for studying clinicians’ decision-making: Validity, utility, and application in ICD-11 field studies. International Journal of Clinical and Health Psychology, 15, 160-170. doi:10.1016/j.ijchp.2014.12.001

15.

Fiske

S. T.

Cuddy

A. J. C.

Glick

(2002). A model of (often mixed) stereotype content: Competence and warmth respectively follow from status and competition. Journal of Personality and Social Psychology, 82, 878-902. doi:10.1037/0022-3514.82.6.878

16.

Fitts

W. H.

(1964). Tennessee Self-Concept Scale. Nashville, TN: Counselor Recordings and Tests.

17.

Frei

J. R.

Shaver

P. R.

(2002). Respect in close relationships: Prototype definition, self-report assessment, and initial correlates. Personal Relationships, 9, 121-139. doi:10.1111/1475-6811.00008

18.

Glass

G. V.

Peckham

P. D.

Sanders

J. R.

(1972). Consequences of failure to meet assumptions underlying the fixed effects analyses of variance and covariance. Review of Educational Research, 42, 237-288. doi:10.3102/00346543042003237

19.

Gomez

McLaren

Sharp

Smith

Hearn

Turner

(2015). Evaluation of the Bifactor Structure of the Dispositional Hope Scale. Journal of Personality Assessment, 97, 191-199. doi:10.1080/00223891.2014.938158

20.

Hamilton

D. L.

Fallot

R. D.

(1974). Information salience as a weighting factor in impression formation. Journal of Personality and Social Psychology, 30, 444-448. doi:10.1037/h0037033

21.

Harman

Jones

(1966). Factor analysis by minimizing residuals (minres). Psychometrika, 31, 351-368. doi:10.1007/BF02289468

22.

Harter

(1999). The construction of the self. A developmental perspective. New York, NY: Guilford Press.

23.

Heatherton

T. F.

Polivy

(1991). Development and validation of a scale for measuring state self-esteem. Journal of Personality and Social Psychology, 60, 895-910. doi:10.1037/0022-3514.60.6.895

24.

L.-T.

Bentler

P. M.

(1999). Cut-off criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6, 1-55. doi:10.1080/10705519909540118

25.

Jennings

P. L.

Mitchell

M. S.

Hannah

S. T.

(2015). The moral self: A review and integration of the literature. Journal of Organizational Behavior, 36, S104-S168. doi:10.1002/job.1919

26.

Kenny

D. A.

Kaniskan

McCoach

D. B.

(2015). The performance of RMSEA in models with small degrees of freedom. Sociological Methods & Research, 44, 486-507. doi:10.1177/0049124114543236

27.

Kernis

M. H.

(2003). Toward a conceptualization of optimal self-esteem. Psychological Inquiry, 14, 1-26. doi:10.1207/S15327965PLI1401_01

28.

Kristjansson

(2007). Measuring self-respect. Journal for the Theory of Social Behaviour, 37, 225-242. doi:10.1111/j.1468-5914.2007.00339.x

29.

Kumashiro

Finkel

E. J.

Rusbult

C. E.

(2002). Self-respect and pro-Relationship behavior in marital relationships. Journal of Personality, 70, 1009-1050. doi:10.1111/1467-6494.05030

30.

Leary

M. R.

Haupt

A. L.

Strausser

K. S.

Chokel

J. T.

(1998). Calibrating the sociometer: The relationship between interpersonal appraisals and the state self-esteem. Journal of Personality and Social Psychology, 74, 1290-1299. doi:10.1037/0022-3514.74.5.1290

31.

Liu

B. S.

Ditto

P. H.

(2013). What dilemma? Moral evaluation shapes factual belief. Social Psychological and Personality Science, 4, 316-323. doi:10.1177/1948550612456045

32.

Luchies

L. B.

Finkel

E. J.

McNulty

J. K.

Kumashiro

(2010). The doormat effect: When forgiving erodes self-respect and self-concept clarity. Journal of Personality and Social Psychology, 98, 734-749. doi:10.1037/a0017838

33.

MacKinnon

D. P.

Fairchild

A. J.

Fritz

M. S.

(2007). Mediation analysis. Annual Review of Psychology, 58, 593-614. doi:10.1146/annurev.psych.58.110405.085542

34.

Marsh

H. W.

Craven

R. G.

Martin

A. J.

(2013). What is the nature of self-esteem? Unidimensional and multidimensional perspectives. In Kernis

M. H.

(Ed.), Self-esteem issues and answers: A sourcebook of current perspectives (pp. 16-24). New York, NY: Psychology Press.

35.

Marsh

H. W.

O’Neill

(1984). Self-Description Questionnaire III: The construct validity of multidimensional self-concept ratings by late adolescents. Journal of Educational Measurement, 24, 153-174. doi:10.1111/j.1745-3984.1984.tb00227.x

36.

McCain

J. L.

Jonason

P. K.

Foster

J. D.

Campbell

W. K.

(2015). The bifactor structure and the “dark nomological network” of the State Self-Esteem Scale. Personality and Individual Differences, 72, 1-6. doi:10.1016/j.paid.2014.08.006

37.

O’Brien

E. J.

Epstein

(1998). MSEI: The multidimensional self-esteem inventory professional manual. Lutz, FL: Psychological Assessment Resources.

38.

Postmes

Haslam

S. A.

Jans

(2013). A single-item measure of social identification: Reliability, validity, and utility. British Journal of Social Psychology, 52, 597-617. doi:10.1111/bjso.12006

39.

Prestwich

Lalljee

(2009). The determinants and consequences of intragroup respect: An examination within a sporting context. Journal of Applied Social Psychology, 39, 1229-1253. doi:10.1111/j.1559-1816.2009.00480.x

40.

Reise

S. P.

(2012). The rediscovery of bifactor measurement models. Multivariate Behavioral Research, 47, 667-696. doi:10.1080/00273171.2012.715555

41.

Renger

(2017). Believing in one’s equal rights: Self-respect as a predictor of assertiveness. Self and Identity, 17, 1-21. doi:10.1080/15298868.2017.1313307

42.

Revelle

(2018). psych: Procedures for psychological, psychometric, and personality research (R package version 1.8.4). Retrieved from https://personality-project.org/r/psych-manual.pdf

43.

Robins

R. W.

Hendin

H. M.

Trzesniewski

K. H.

(2001). Measuring global self-esteem: Construct validation of a single-item measure and the Rosenberg Self-Esteem Scale. Personality and Social Psychology Bulletin, 27, 151-161. doi:10.1177/0146167201272002

44.

Roland

C. E.

Foxx

R. M.

(2003). Self-respect: A neglected concept. Philosophical Psychology, 16, 247-288. doi:10.1080/09515080307764

45.

Rosenberg

, (1965). Society and the adolescent self-image. Princeton, NJ: Princeton. University Press.

46.

Rosenberg

Schooler

Schoenbach

Rosenberg

(1995). Global self-esteem and specific self-esteem: Different concepts, different outcomes. American Sociological Review, 60, 141-156. Retrieved from http://www.jstor.org/stable/2096350

47.

Rubin

Hewstone

(1998). Social identity theory’s self-esteem hypothesis: A review and some suggestions for clarification. Personality and Social Psychology Review, 2, 40-62. doi:10.1207/s15327957pspr0201_3

48.

Sachdeva

Iliev

Medin

D. L.

(2009). Sinning saints and saintly sinners: The paradox of moral self-regulation. Psychological Science, 20, 523-528. doi:10.1111/j.1467-9280.2009.02326.x

49.

Sedikides

Wildschut

Arndt

Routledge

(2006). Affect and the self. In Forgas

J. P.

(Ed.), Affect in social thinking and behavior (pp. 197-215). New York, NY: Psychology Press.

50.

Tafarodi

R. W.

Swann

W. B.

Jr. (1995). Self-linking and self-competence as dimensions of global self-esteem: Initial validation of a measure. Journal of Personality Assessment, 65, 322-342. doi:10.1207/s15327752jpa6502_8

51.

Van Quaquebeke

Eckloff

. (2010). Defining respectful leadership: What it is, how it can be measured, and another glimpse at what it is related to. Journal of Business Ethics, 91, 343-358. doi:10.1007/s10551-009-0087-z

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

0.53 MB