Skip to main content
Open AccessOriginal Article

Socioeconomic Status and Self-Regard

Income Predicts Self-Respect Over Time

Published Online:https://doi.org/10.1027/1864-9335/a000536

Abstract

Abstract: Past research has shown that the socioeconomic status (e.g., income or education) is associated with people’s self-evaluation, such as global self-esteem. In the present research, we argue that socioeconomic status also affects people’s belief of possessing the same rights as others (i.e., self-respect). In a cross-sectional study (N = 298) and a longitudinal study (N = 379), we investigated the relationships between income and education with three forms of self-regard. The only consistent finding was that income was related to self-respect over time even when controlling for self-love and self-competence, the core and well-studied components of global self-esteem. We discuss the significance of our findings with regard to social justice and democracy.

Socioeconomic status has an undeniable influence on many aspects of people’s lives and self-understanding. As Manstead (2018) summarized, socioeconomic status (SES) affects humans’ thoughts, feelings, and behavior. Among other things, high socioeconomic status has been associated with better mental and physical health (Adeline & Delattre, 2017; Lorant et al., 2003). It has also been argued that the socioeconomic status of a person influences their self-esteem (Kraus & Park, 2014; see also Sandel, 2020). More specifically, several cross-sectional studies (Twenge & Campbell, 2002) and a few longitudinal studies (Orth, 2018; Orth et al., 2015) found correlations between socioeconomic status and global self-esteem.

Global self-esteem consists of a competence-based and a liking-based self-evaluation (Tafarodi & Swann, 1995, 2001). This means that the higher the socioeconomic status of people, the more competent they see themselves and the more they like or love themselves.

Recently, a third factor has been proposed, which has been shown to be independent of self-liking/self-love1 and self-competence (Renger, 2018). This third self-factor describes individuals’ belief of possessing the same rights as others and has been referred to as self-respect (see also Darwall, 1977; Feinberg 1970; Honneth, 1995). In the present research, we extend the earlier literature by investigating whether socioeconomic status is also associated with self-respect. In addition, we explore whether different indicators of SES are differentially associated with different forms of self-regard (i.e., self-love, self-competence, and self-respect).

Self-Respect as a Third Form of Self-Regard

The reasoning for considering individuals’ belief of possessing the same rights as others (i.e., self-respect) as a third form of self-regard in addition to self-love and self-competence stems from social recognition theory (Honneth, 1995). Based on a social philosophical analysis, Honneth (1995, 2012) first distinguished between three forms of recognition experiences, need-based care, achievement-based social esteem, and equality-based respect (see also Renger et al., 2017; Simon & Grabow, 2014). Whereas social esteem (e.g., positive feedback, praise) forms individuals’ self-competence (Battistelli et al., 2016; Honneth, 1995, 2012) and care contributes to self-love (see also Bowlby, 2005), experiences of equality-based respect shape self-respect (Möller & Danermark, 2007; Presbey, 2003; Renger et al., 2020).

According to Honneth (1995), the universal idea of human dignity and equal worth (Kant, 1977) has been codified in legal rights in constitutions. But individuals can only develop a sense of themselves as moral agents with equal rights and the same freedom to live a self-determined life to the extent that they are treated as someone of equal worth by others (Renger et al., 2017).

In other psychological research, self-respect has been regarded as an individual’s appraisal of their adherence to their own moral standards (appraisal self-respect, Clucas, 2020; Clucas et al., 2023). People high in appraisal self-respect perceive themselves as having an honorable character of high quality (Kumashiro et al., 2002). Recent research shows that appraisal self-respect is affected by adherence to morals and by competence appraisals (Clucas, 2020) and can also be bolstered through forgiving when a perpetrator signals that the victim will be safe (Luchies et al., 2010). Which character traits are considered as honorable depends on the honor code in a person’s society or environment (Clucas et al., 2023). In that respect, it is different from equality self-respect (Renger, 2018; Renger et al., 2020) that taps people’s evaluation of being equal to others and having the same dignity, worth, and rights as others.

Equality self-respect has been empirically distinguished from appraisal self-respect through factor analysis (cf. Clucas et al., 2023, Footnote 3). In the present contribution, we focus on equality self-respect. Picking up the theoretical idea of three forms of self-regard that are influenced through three forms of social recognition, Renger (2018) provided evidence that equality self-respect can be empirically distinguished from self-love and self-competence. Confirming prior research, self-love and self-competence loaded on the same factor as the Rosenberg global self-esteem items (cf. Richardson et al., 2009; Tafarodi & Swann, 2001), whereas self-respect constituted a third, independent factor (only the Rosenberg item “I feel that I am a person of worth, at least on an equal basis with others” loaded on this factor; see Table 3 in Renger, 2018). This means that perceiving oneself as being lovable or competent is different from perceiving oneself as being equal to others.

Socioeconomic Status and Self-Regard

Socioeconomic status can be defined both objectively and subjectively (Adler et al., 2000). Objective SES defines a person's status in terms of the absolute level of material resources that they possess, often indexed by income level (e.g., Howell & Howell, 2008), educational attainment (e.g., Witter et al., 1984), or a combination of both indices (e.g., Pinquart & Sörensen, 2000). In contrast to subjective measures of SES, where individuals report their own subjective perception of their socioeconomic position or rank within a society (Kraus et al., 2012), income and education involve factual reports of life circumstances that can be considered rather objective. In this research, we focus on income and education as objective measures.

Income and education have been shown to have independent and sometimes differential effects, for example, with regard to prejudice (Carvacho et al., 2013), psychopathic symptoms in children (Zhang, 2014), and political attitudes (Kitschelt & Rehm, 2022). In the present research, we thus look at differential associations with the different forms of self-regard.

Past research has demonstrated that socioeconomic status predicts self-esteem (e.g., Kraus & Park, 2014; Orth, 2018; Orth et al., 2015; Twenge & Campbell, 2002). More specifically, it has been found that income (Orth, 2018; Orth et al., 2015) and education (Orth et al., 2010) predict self-esteem over time. These longitudinal effects of income and education on global measures of self-esteem were, however, not robust and not replicated in later studies (e.g., Erol & Orth, 2011; Orth et al., 2012, 2015; von Soest et al., 2018). In the present research, we take a more differentiated perspective and separately examine the two subdimensions of self-esteem: self-love and self-competence. Prior research suggests that an association with socioeconomic status can be assumed for self-competence. More specifically, low SES individuals develop lower internalizations of competence (Durante et al., 2017) and lower motivation for being successful (Laurin & Engstrom, 2020; see also Dóci et al., 2023 for a discussion of self-efficacy beliefs as psychological capital). In addition to focusing on self-competence, we further extend prior research by investigating associations of income and education with self-love and with equality self-respect.

Socioeconomic Status and Self-Respect

Why is socioeconomic status predicted to influence self-respect? People are not born with high or low self-respect but form this self-view in response to how they are treated by others (Bratu, 2019). As a consequence, existing legal rights (e.g., one’s basic rights as a citizen) and subjective perceptions of one’s own rights (i.e., self-respect) do not always coincide (see also Young & Billings, 2020). This is supported by research showing that not all people have internalized their theoretical equal rights and the associated entitlements to the same degree (e.g., Jost, 1997; Major, 1994; Major et al., 1984; Renger, 2018). Although two individuals living in the same country are legally entitled to the same basic rights based on the country’s constitution, they may differ in the extent to which they have internalized these rights. The ease of internalization depends on whether they have been treated and respected as someone of equal worth and as equal counterparts by others in their life.

This is the reason why socioeconomic status in terms of income or education can affect people’s sense of being equal to others in terms of their basic rights. Individuals with lower socioeconomic status, like people from other disadvantaged groups, experience more social exclusion and discrimination (e.g., Augoustinos & Reynolds, 2001; Brooks, 2019; Sue et al., 2007) and often show a diminished sense of entitlement to equal treatment (Martiny et al., 2023; Jost, 1997; Major, 1994). Social discrimination involves both structural, institutionalized practices or policies and interpersonal biased treatment or discrimination.

Related to the first aspect, lower SES individuals suffer from reduced geographic mobility (Stephens et al., 2007) as well as higher rates of inadequate and unstable housing (Marmot, 2002), which makes it much more difficult to perceive an equal space for oneself (e.g., Symonds, 1968). People of lower SES additionally face unequal treatment with regard to educational resources and opportunities (e.g., Paulus et al., 2021) and voting availability (e.g., through poll taxes or voter ID laws that create obstacles for poorer people; cf. Peterman, 2018). Related to the second aspect (i.e., biased treatment or discrimination), individuals with lower SES experience discrimination in many areas, including teacher evaluations, housing offers by landlords, and selection by employers (for overviews, see Peterman, 2018; Sisselman-Borgia et al., 2021). Taking all these factors into account, we argue that it is less likely that people with lower socioeconomic status develop a sense of possessing the same rights as others (i.e., self-respect) than people with higher socioeconomic status.

The Present Research

So far, empirical research has only tested the association between socioeconomic status and global self-esteem. As global self-esteem includes two components, self-love and self-competence (Tafarodi & Swann, 1995, 2001), we know from this research that socioeconomic status might affect how much people like or love themselves and how much they perceive themselves as competent. In the present research, we focus on self-respect as a third component of self-regard and test whether it is affected by a person’s income and/or education level. To this end, we present a cross-sectional study (Study 1) and a longitudinal study (Study 2). Because higher socioeconomic status enables more opportunities for participation and fewer reasons for social discrimination, SES should be positively associated with self-respect. We focus on income and education, and for both, we predict positive relationships with self-respect because a low level of either dimension should hinder positive respect experiences and social participation.

Ethical approval for all studies was obtained by the Ethics Committee of the first author’s institution. Participants in all studies gave their informed consent prior to their participation and agreed to data protection declarations concerning collection, storage, and publication of their data. We report all exclusions in these studies. Data and code are available at https://osf.io/d5xey/ and further information and additional statistical analyses can be found in the Supplementary Material.

Study 1

In Study 1, we used a cross-sectional design to test whether income and education are related to self-love, self-competence, and/or self-respect.

Method

Participants

Following Schönbrodt and Perugini (2013), who found that the necessary sample size to achieve stable estimates for correlations should be at least 250, we aimed to recruit this minimum sample size. Participants were a convenience sample recruited via social media and had to be at least 18 years old to participate. Three hundred and three participants completed the measures as part of a larger online survey2 in Germany. Participants were able to participate in a lottery for three rewards worth about 50 Euro each. We excluded five participants who failed attention checks. Mean age of the remaining 298 persons was 33.7 years (SD = 16.1 years; range 18–80 years). There were 199 women and 99 men. 46.3% of the participants were employed, 39.3% were university students, and 10.1% were unemployed, retired, or high school students (4% missing values). 34.6% had a university degree, 49% had the (Fach-)Abitur (general qualification for university entrance), and 12.4% indicated a lower education level (4.4% missing values).

Measures

All responses were provided on 7-point Likert scales ranging from 1 (= not true at all) to 7 (= completely true). Self-respect, self-love, and self-competence were measured with Renger’s (2018) 4-item scales (self-respect, e.g., “In everyday life I always see myself as a person with equal rights;” self-love, e.g., “I look at myself with warmth and affection;” self-competence, e.g., “I think that I’m very good at the things I do”).

Participants filled out demographic information (age, gender, and occupation), including personal monthly net income (including after-tax income, support services, etc.) using 250-Euro ranges (€ 0–249; € 250–499; € 500–749; …; € 3,750–3,999, more than € 4,000). Education was measured as 1 = no school-completion qualification, 2 = Hauptschulabschluss (secondary school completion certificate), 3 = Realschulabschluss (realschule certificate), 4 = Fachabitur (advanced technical college entrance qualification), 5 = Abitur (university entrance qualification), and 6 = Hochschulabschluss (university diploma).

Results

For M, SDs, bivariate correlations, and Cronbach’s αs, see Table 1. As both age and gender differences have been documented for income (e.g., Eurostat, 2019; Lee & Mason, 2007), we included them as control variables in our analyses. In addition, we included the measure of income and the ordinal measure of education3 as markers of socioeconomic status. We tested the hypothesized model with path analyses with four manifest variables (income, education, gender, and age) and three latent variables (self-respect, self-love, and self-competence) using the lavaan package (Rosseel, 2012) in R.

Table 1 M, SD, bivariate correlations, and Cronbach’s αs of Study 1 (N = 298)

Following the recommendations by Yu (2002; see also van de Schoot et al., 2012), the initial CFA showed a good fit to the data confirming the assumed tripartite latent factor structure (χ2 = 87.808, df = 51, p = .001, TLI = .978, CFI = .983, RMSEA = .040, SRMR = .035). In the next step, the predictors and covariates, which had been standardized before the analyses, were included in the structural part of the model (χ2 = 142.825, df = 87, p = .001, TLI = .948, CFI = .962, RMSEA = .049, SRMR = .041). As expected, income was related to self-respect (β = .22) and additionally to self-competence (β = .35; see Table 2 and Figure 1). Education only predicted self-respect (β = .14). Without the covariates, income was related to all three forms of self-regard while education was only related to self-respect (see supplementary material, Table E1, which also includes robustness checks with manifest modeling).

Table 2 SEM for the relationships between income and education (manifest variables) and self-love, self-competence, and self-respect (all latent variables) controlling for gender and age in Study 1
Figure 1 Path model for the relationships between income and education (manifest variables) and self-love, self-competence, and self-respect (all latent variables) controlling for gender and age in Study 1. Note. Standardized coefficients are shown. For better readability, only significant correlations and regression coefficients are depicted (see Table 2 for all parameters).

Study 2

Study 1 provided initial evidence that income and education are indeed positively associated with self-respect. Income was also related to self-competence. In Study 2, we analyzed data from a longitudinal data set to explore the relationship between income and education with the three forms of self-regard over time.

Method

Study 2 was part of a larger research project that investigated the relationship between social recognition experiences and people’s self-views in Germany. One published article used data from Time 1 (Renger et al., 2017). Data from Time 2 (between 6 and 8 months later) have not been published.

Participants

At Time 1 (T1), 383 persons participated. Participants were able to take part in a lottery for one of five prizes of € 30. Four participants who reported being younger than 18 years were excluded, resulting in a final data set with 379 participants. There were 251 women and 128 men (Mage = 27.8, SD = 8.4, range 18–75 years). 38.8% had a university degree, 54.6% had the (Fach-)Abitur (general qualification for university entrance), and 5.5% indicated a lower education level (1.1% missing values). Regarding employment status, 42.0% of the participants were employed (including apprenticeship), 59.1% were university students, and 12.7% were unemployed, retired, or high school students (multiple answers were possible).

Two hundred twenty participated again at Time 2 (T2), and of these, 164 participants’ personal codes could be matched with the corresponding data from Time 1.

Data Analysis

The goal of Study 2 was to investigate the longitudinal reciprocal relationships between income and education4 and each of the three forms of self-regard. We employed a standard cross-lagged panel model (CLPM, Jöreskog, 1970, see also, Usami et al., 2019) to investigate these dynamic effects. The CLPM as implemented in our study analyzes the extent to which change in one variable (i.e., development from T1 to T2) can be predicted by another variable measured at T1. By always controlling for the states of the outcome variable(s) at T1 (the autoregressive paths), the cross-lagged parameters represent partial correlations. The CLPM was implemented using the structural equation modeling framework of the lavaan package in R. This latent modeling approach allowed us to control for measurement error.

Furthermore, to address the dropouts, we applied the full information maximum likelihood (FIML) approach, which is integrated in lavaan. FIML is superior to traditional missing data treatments such as listwise deletion, as they can substantially reduce missing data bias under at least the missing at random (MAR) assumption (Enders, 2010). We ran Little’s multivariate test of the missing completely at random (MCAR) assumption (e.g., Enders, 2010). The test statistic was not significant, meaning that the assumption that this more restrictive pattern of missing data (MCAR) was present did not need to be rejected. All key findings were also confirmed with the listwise-deleted data set (N = 164 see supplementary material, Table E8; see also additional robustness checks, Table E7 ff.).

A post hoc Monte Carlo power analysis was run with the cross-lagged model applied. We assumed the parameter estimates and covariances found were the true population parameters (Beaujean, 2014; Muthén & Muthén, 2002). Furthermore, we imposed the missing data rate of 57% on the outcome variable self-respect (T2), as in the real data. The simulation study (with 1,000 iterations) estimated a power of .83 (95% CI = .73–.94) to detect the effect (β = .22) of income (T1) on self-respect (T2) controlling for self-respect (T1), gender, age, education (T1), self-love (T1), and self-competence (T1; see also Figure E1, https://osf.io/d5xey/).

Measures

Measures5 were identical at T1 and T2. Net income, education, self-respect, self-love, and self-competence were measured with the same items as in Study 1. Responses were provided on 7-point Likert scales ranging from 1 (= not true at all) to 7 (= completely true).

Results

M, SDs, bivariate correlations, and Cronbach’s αs are presented in Table 3. To establish measurement invariance (MI) over time, we followed standard practices to compare models and added stepwise parameter restrictions (e.g., Little, 2013; van de Schoot et al., 2012). Fit measures of the initial configural model with no equality constraints over time indicated acceptable fit (χ2 = 452.79, df = 225, p < .001, TLI = .933, CFI = .945, RMSEA = .048, SRMR = .059). Next, we successively constrained factor loadings and item intercepts to be equal over time. Comparing the resulting models, likelihood-ratio tests, as well as differences in CFI, RMSEA, and SRMR (following the recommendation by Chen, 2007), implied that metric measurement invariance (MI; ∆χ2 = 10.245, ∆df = 12, p = .331, ∆CFI = −.001, ∆RMSEA = −.001, ∆SRMR = .005) and scalar MI (∆χ2 = −17.69, ∆df = 12, p = .126, ∆CFI = −.001, ∆RMSEA = −.001, ∆SRMR = .002) could be accepted. Based on the strong MI model, we proceeded and tested a model with income and education and the three forms of self-regard and again controlled for gender and age (Table 4). In addition to the autoregressive paths, the cross-path from income (T1) to self-respect (T2; β = .22) was significant. In addition, the cross-path from education (T1) to self-love (T2; β = .13) and the path from self-competence (T1) to self-love (T2; β = .22) were significant. All other cross-paths were nonsignificant (see Table 4; results for all covariances are included in the supplementary material, Table E5).6 The pattern for the relationship between income and self-respect is depicted in Figure 2. All the results were also confirmed without the covariates.

Table 3 M, SD, bivariate correlations, and Cronbach’s αs of Study 2 (N = 164)
Table 4 Cross-lagged panel model for the relationships between income and education (manifest variables) and self-love, self-competence, and self-respect (all latent variables) controlling for gender and age in Study 2
Figure 2 Cross-lagged panel model showing significant temporal relationships between income (manifest variable) and self-respect (latent variable) controlling for age, gender, education, self-love, and self-competence in Study 2. Note. Standardized coefficients are shown. For better readability, only significant coefficients of covariates are depicted (see Table 4 for all parameters).

Discussion

Numerous studies have shown that income affects people’s self-evaluation in terms of global self-esteem (Orth, 2018; Orth et al., 2015; Twenge & Campbell, 2002), but findings have not always been consistent and indicate bidirectionality (e.g., e.g., Erol & Orth, 2011; Orth et al., 2012, 2015; von Soest et al., 2018). In the present research, we integrated interdisciplinary theorizing suggesting a tripartite model of self-esteem/self-regard differentiating self-love, self-competence, and self-respect (Honneth, 1995, 2012). Whereas self-love and self-competence represent the core components of global self-esteem (Tafarodi & Swann, 1995, 2001), self-respect refers to individuals’ belief of possessing the same rights as others. We included income and education as indicators of socioeconomic status and examined associations with self-love, self-competence, and self-respect in a cross-sectional study (Study 1) and a longitudinal study (Study 2). In the estimated path model using cross-sectional data in Study 1, income was related to both self-respect and self-competence, whereas education was solely associated with self-respect. Bivariate correlations also showed correlations between income and self-love (see Table 1); this link, however, was not significant using the latent modeling strategy in the SEM when the other forms of self-regard, age, and gender were controlled for (see Figure 1).

Using longitudinal data, in Study 2, we found that income predicted self-respect over time (see Table 4). No such effect was found for the other two components of global self-esteem, namely self-love or self-competence. In prior research, some studies found longitudinal relationships between income and self-esteem (Orth, 2018), whereas others did not find this effect (Erol & Orth, 2011; von Soest et al., 2018) or found correlations in the opposite direction (Orth et al., 2012). By taking a differentiated look at the three forms of self-regard, our findings for self-love and self-competence did not confirm a longitudinal effect of income on the main components of global self-esteem (cf. Tafarodi & Swann, 1995, 2001). Going beyond prior research, the present findings shed new light on hitherto neglected aspects of individuals’ self-regard, namely self-respect. The present studies show that the more money people earned at Time 1, the more convinced they were that they possessed the same rights as others at Time 2.

For education, no longitudinal association with self-respect was observed, but education was associated with self-love over time. This is in line with Twenge and Campbell’s (2002) finding that self-esteem (which includes a strong self-love component, cf. Renger, 2018) is more strongly correlated with education than with income. Overall, our findings show that different forms of self-regard might be affected by different indicators of socioeconomic status. However, the only path that was consistent and stable across both studies (and also in all robustness checks, see supplementary material), was the one between income and self-respect. The association between education and self-love over time found in Study 2 was not observed in Study 1.

Limitations and Further Directions

Because we used a correlational design in Study 1, causal inferences cannot be made. Our longitudinal findings from Study 2, however, confirm that self-respect is affected by income over time. In the present research, we focused on income and education as indicators of socioeconomic status. We found that education was correlated with self-respect in both studies, but no longitudinal effect was observed. Future research should thus further investigate associations of education as well as other indicators of socioeconomic status, such as job status, wealth, or household conditions (Galobardes et al., 2006) with self-respect.

Furthermore, the research questions were examined with rather homogeneous samples (regarding income and education). Although this homogeneity can be assumed to result in a more conservative test of the relationships (see also supplementary material, Tables E3 and E11, for robustness checks without students) as they should tend to be larger in a more heterogeneous sample, the findings should be replicated with more representative samples.

Future research should also investigate the further consequences of the negative association between income and self-respect. Both low income (e.g., Adeline & Delattre, 2017; Lorant et al., 2003; Marmot, 2002) and low self-respect (Renger et al., 2023) have been associated with poorer mental health, and a next step could be to test self-respect as a mediator of the relationship between income and mental health. Similarly, it may be worthwhile to investigate self-respect as a mediator of the relationship between income and aggressive behavior (e.g., Greitemeyer & Sagioglou, 2018). Not perceiving an equal space or effective voice in society has been suggested as paving the way to diverse forms of aggression (Moghaddam, 2005).

The Need to Study Self-Respect as Internalized Equality in the Context of Low-Income Groups

Our finding that the more money people earn, the more equal they feel in terms of basic rights is alarming (see also Young & Billings, 2020). The resulting self-evaluation of people with low SES may hinder self-assertive behavior (Renger, 2018) and protest against injustice (Renger et al., 2020),7 which is an important cornerstone of democracies (Passini, 2011). It also hinders a sense of entitlement to legitimately participate in political discourses and to have and express political opinions more generally (Bourdieu, 1979, 1984).

Our findings indicate that lower SES individuals show lower levels of self-respect, i.e., internalization of equal rights. We would like to stress that this deficiency should not be regarded as a dispositional deficit of lower SES individuals (cf. Frankenhuis & Nettle, 2020), but rather as a systemic deficit in societies’ ability to ensure equal treatment and discrimination-free institutions for all their citizens (see e.g., Peterman, 2018, for an overview of laws that create unequal opportunities and treatment regarding socioeconomic status).

Society treats persons of certain groups as less equal and thus conveys the idea that poorer people are less worthy as humans (e.g., Sainz et al., 2021), which is then internalized by them (cf. Renger et al., 2016). In line with this, recent research finds that disadvantaged group members’ perceived lack of public respect experiences (i.e., being treated as someone of equal worth in state/public institutions, society, and/or workplace/university) is more strongly associated with lowered self-respect than a perceived lack of private respect experiences (i.e., within family, partnership, and/or friendships; Martiny et al., 2023).

Implications and Intervention Strategies

Our findings suggest that income is especially related to self-respect. In light of Bourdieu's assumption that economic capital (e.g., money) is more fluid and thus more easily gained than cultural capital (e.g., education; Bourdieu, 2021; Bourdieu & Passeron, 1990), this gives us hope for potential changes in the self-respect of low SES individuals, for example, through redistribution policies. Recent research testifies to the possibility for change and class transitions over time (Phillips et al., 2020) when adequate initial assets are provided to poorer people (Balboni et al., 2022; Iqbal et al., 2021).

Redistribution policies as well as social programs should aim directly at enabling respect experiences and equal treatment for all regardless of demographics or social categories. Programs (e.g., in schools) such as citizenship education programs, diversity training, antiracism programs, or implicit bias training are approaches that could be implemented (e.g., Beelmann & Heinemann, 2014; Beelmann & Lutterbach, 2020; Fitzgerald et al., 2021; Moghaddam, 2005). Also, it could be beneficial to organize schools with balanced ratios of low and high socioeconomic status students, as research shows that such schools are more inclusive and equal than schools with predominantly high socioeconomic background (Grütter et al., 2022).

Overall, ensuring equal treatment and respect for all members of society has been repeatedly stressed as a crucial aspect of justice from scholars of different scientific disciplines (Honneth, 2012; Rossiter, 2014). In his theory of justice, John Rawls described the social bases of self-respect as “perhaps the most important” primary good (Rawls, 1971, p. 396; see also Crego, 2022) that must be ensured by governments, especially for people of lower social classes (see also Stark, 2012).

We thank Theresa Sohst, Ronja Zimmermann, Lilli Stief, and Viviane Füser for the help with collecting data and preparing the article and Elizabeth J. Parks-Stamm for her valuable feedback on the manuscript.

References

1Note that in Renger (2018), self-love is referred to as self-confidence. In light of the diverse usage of the term self-confidence, we think that self-love better captures the idea of seeing the self as a likable, lovable person.

2Further scales included in the questionnaires: social recognition experiences, assertiveness (Studies 1 and 2), mental health (Study 2), satisfaction with life (Study 2), and human rights attitudes (Study 1).

3For robustness checks with a dichotomous version of this measure, see Tables E2 and E4, https://osf.io/d5xey/.

4All results are also replicated with a dichotomous version of this measure, see Tables E7 and E12, https://osf.io/d5xey/.

5A measure of global self-esteem (Rosenberg, 1965) was also available in the data set. We present associations with income and education in Table E6 in https://osf.io/d5xey/. Analyses showed that self-esteem (T1) predicted income (T2; replicating Orth et al., 2012) and income (T1) predicted self-esteem (T2; replicating Orth, 2018; Orth et al., 2015), the latter effect was only significant without age and gender as covariates.

6This effect is also replicated when including education (T2) in a fully cross-lagged model, see Tables E9 and E10 (https://osf.io/d5xey/).

7Self-respect was also positively and significantly correlated with a measure of assertiveness available in the data sets used in this article (see https://osf.io/d5xey/, Tables E4 and E12).