This is an open access article distributed under the terms of the Creative Commons
Attribution License (The heterogeneity of (post-)modern societies entails that various social groups claim recognition for their place in society (Honneth, 1995; Parekh, 2008; Simon, 2020). This claim for recognition in society is frequently expressed in the form of a demand for respect. The granting of respect in turn possesses pivotal potential for pacifying intergroup disagreements in plural societies. It is therefore of significance to examine how the granting of respect towards various social groups can be fostered. The studies underlying this article focus on the degree to which respect for outgroups can be fostered through the feeling of being respected as group members.
Respect as Equality Recognition
Our conceptualization of respect draws on Kant’s (1974) moral philosophy as well as Honneth’s (1995) recognition framework. The moral principles derived from the former demand that one does not reduce others to mere means to one’s own ends, but that one recognizes them as persons with dignity and thus also as ends in themselves. Since all rational beings possess dignity, all deserve consideration as equals (Kant, 1974). This egalitarian conception of respect informs the recognition framework proposed by Honneth (1995). According to Honneth, respect is rooted in the sphere of legal and political relations governed by the equality principle. It is in this sphere of political relations in which societal groups are demanding recognition, which they often frame as a claim for respect. The foci of the present article, i.e., respect dynamics between societal groups, are located in this political sphere. In accordance with Honneth, we therefore conceive equality recognition as being at the core of respect in this context.
Previous findings on the meaning of respect are in line with this conception of respect. In previous research, lesbians and gay men in Germany rated to which degree they felt respected and recognized by the societal majority (Simon & Grabow, 2014), thereby providing information about participants’ understanding of the concept of respect. When relating the feeling of being respected to various forms of being recognized, the feeling of respect was more strongly associated with equality recognition than with other recognition principles identified by Honneth (1995; namely achievement and need recognition). Experimental work with ad hoc groups further confirmed this link. In a study by Renger and Simon (2011), it was demonstrated that participants’ feeling of being respected depended on their experimentally prompted feeling of being recognized as an equal.
The Positive Effects of Respect
The results of Renger and Simon’s (2011) experiments further demonstrated that the feeling of being respected was associated with the identification with an ingroup and with group serving behaviour, thereby pointing to the psychological significance of feeling respected. The link between equality recognition and group identification was also examined with natural groups on a societal level. In a longitudinal study on the experiences of Muslims living in Germany, it was demonstrated that Muslims’ perception of being recognized by society as equals yielded stronger identification with Germany (Schaefer & Simon, 2020), and this identification, in turn, fostered recognition of other societal subgroups, thus pointing to the wider significance of equality recognition (which we conceive as the core of respect) in shaping societal group dynamics. This result further corroborated the link between (superordinate) group membership and equal treatment within this (superordinate) group.
The positive effects of granting respect are particularly pertinent for intergroup dynamics in (post-)modern societies which are encompassing a multiplicity of social, ethnic, cultural, or religious groups. As a corollary of this plethora, it is inevitable that people disapprove of some outgroups’ beliefs, preferences, or practices. Given this plethora, tolerance is vital for enabling open, plural societies (Popper, 2013; Rawls, 2005; Scanlon, 2003), while respect is a constituent of tolerance. We build on conceptions that recognize disapproval or disagreement as a “definitional condition” for tolerance (Gibson, Duch, & Tedin, 1992, p. 338; Simon & Schaefer, 2016; Verkuyten & Yogeeswaran, 2017), as it would make little sense to say we tolerate something that we already approve of or even appreciate (see Scanlon, 2003). As argued by Forst (2013) and demonstrated in social psychological research (Simon, Eschert, et al., 2019; Simon & Schaefer, 2016, 2018), respect for others as equal members of society restrains people’s disapproval of others and thus makes tolerance possible. Tolerance thus is the combination of respect and disapproval.
The Present Research
As a form of social recognition, respect should be characterized by reciprocity (Gouldner, 1960; Honneth, 1995; Sennett, 2003). Thus, group members should be more likely to respect an outgroup when they feel respected by this particular outgroup. In line with this assumption, Simon and Grabow (2014) found in a cross-sectional study with lesbians and gay men in Germany that perceived respect from the Muslim community negatively predicted anti-Muslim attitudes, which is conceptually related to outgroup (dis)respect. Additionally, Simon and Schaefer (2018) observed that among the Muslim minority in Germany, perceived equality recognition from disapproved outgroups (i.e., homosexuals, atheists, and feminists) longitudinally predicted equality recognition for those outgroups. We tried to replicate these findings by the studies underlying the present article (H1: Intergroup Respect Reciprocity Hypothesis).
Moreover, intergroup relations are usually embedded in a wider societal context, within which the general population or society as a whole is expected to function as an important source of respect for one’s ingroup. Due to the link between group membership and equal treatment, respect from society in the sense of recognition as an equal should signal to group members that their group is accepted as part of larger society. In fact, it was already observed that members of minority groups (lesbians and gay men as well as Muslims living in Germany) who feel recognized as equals by society are more likely to identify with society (Schaefer & Simon, 2020; Simon & Grabow, 2014). As the belongingness of minority groups to society can be more easily challenged than the belongingness of majority groups, minority groups are likely to be more in need of, and thus to appreciate, respect from society as a whole, which may then encourage them to extend their respect towards outgroups. Majority groups, in contrast, can usually take societal respect for granted, so that they are likely to be less in need of (and less sensitive to) confirmation of their position in the form of respect by society. We therefore expected that especially minority groups are sensitive to respect from society, colouring their relations to wider society, so that they also respect other outgroups. In line with this expectation, Simon and Grabow (2014) found in a cross-sectional study with lesbians and gay men in Germany that perceived respect from the majority population negatively predicted anti-Muslim attitudes, while Schaefer and Simon (2020) demonstrated with a sample of Muslims living in Germany that perceived recognition from society fosters the recognition of outgroups (Schaefer & Simon, 2020). In line with these findings, we anticipated that respect from society would predict respect for outgroups, especially so among members of minority groups (H2: Societal Respect Reciprocity Hypothesis).
Finally, respect from one’s fellow ingroup members should confirm self-categorization as a group member, thereby increasing self-stereotyping, rendering both ingroup favouritism as well as outgroup devaluation more likely (Turner, Hogg, Oakes, Reicher, & Wetherell, 1987). While we derive from self-categorization theory that there is a link between group membership and equality expectations, we acknowledge that the (de facto) degree of given and experienced equality recognition can vary. When one’s own membership in a group is confirmed by other ingroup members through equality recognition, this should increase group salience as well as the perceived prototypically of the self. As a consequence of increased group salience and self-stereotyping, adherence to group norms should be reinforced (Turner et al., 1987). These group norms in turn can define relationships to other groups, such as to outgroups with opposing ideologies. We therefore formulate the novel (and, to our knowledge, so far untested) expectation that the experience of being respected by other ingroup members would turn into disrespect for disapproved outgroups (H3: Ingroup Respect Outgroup Disrespect Hypothesis).
The two studies that had so far provided support for the Intergroup Respect Reciprocity Hypothesis and the Societal Respect Reciprocity Hypothesis (Simon & Grabow, 2014; Simon & Schaefer, 2018) examined German minority group samples, analysing their attitudes and respect towards other minority outgroups.
We extend the aforementioned work in two ways: We focus – similar to Simon and Schaefer (2018) – on antecedents of respect for disapproved outgroups. We expand the possible sources of respect, however, by adding respect from ingroup members to the two previously examined sources of respect. As a further extension of previous studies, we examine the dynamics of respect in novel contexts. That is, in both of our studies, we examine antecedents of mutual respect between groups associated with opposite poles of the political and cultural spectrum, thus forming two nearly symmetrical constellations. In Study 1, we recruited research participants from a group that claims to represent traditional values and the majority’s way of life, namely, supporters of the U.S. Tea Party movement, with homosexuals in the United States serving as the target outgroup. In Study 2, we recruited research participants from a minority group representing liberal, progressive, or (post)modern values and ways of life, namely, members of the German LGBTI (lesbian, gay, bisexual, trans, intersex) community, with supporters of the German populist, right-wing political party „Alternative für Deutschland“ (AfD) serving as the target outgroup. Taken together, we investigate the dynamics of mutual respect in two complementary sets of intergroup relations. Finally, by using a longitudinal study approach, we are able to examine effects that develop in intergroup dynamics over time. By using field studies in two different national and political contexts, we are tracking these effects under different context conditions, thereby supplementing experimental work.
Study 1
With a sample of Tea Party supporters in the United States as research participants, we tested the hypothesis that respect from the disapproved outgroup of homosexuals increases respect for this outgroup (H1: Intergroup Respect Reciprocity Hypothesis). We also explored whether receiving societal respect plays a role in the development of respect for homosexuals among these (self-proclaimed) representatives of the societal majority (H2: Societal Respect Reciprocity Hypothesis). Finally, we investigated whether respect from the conservative ingroup may turn into disrespect towards homosexuals (H3: Ingroup Respect Outgroup Disrespect Hypothesis).
We have chosen the Tea Party movement as participants to examine a group close to the conservative end of the traditional vs. liberal spectrum. The Tea Party is a populist movement of conservative, mostly middle-class and middle-aged or older, white Americans claiming for fiscal, social, and cultural restoration (“I want my country back!”, Skocpol & Williamson, 2012). In the area of fiscal policies, they lobby for a lean state with a focus of providing security to its citizens. In the social and cultural domain, they tend towards a conservative orientation regarding gender roles and family values. Around the time we started our research, approximately 17% of Americans identified themselves as Tea Party supporters (Norman, 2015). Homosexuals served as a disapproved target group because they represent one of Tea Party supporters’ liberal, progressive or postmodern societal antagonists.
Method
Procedure
We recruited our research participantsi online via Qualtrics (www.qualtrics.com) using convenience sampling. Between August 11 and September 12, 2016, the first wave of data collection was conducted, following a two-step procedure. Those participants who self-identified as Tea Party supporters, were U.S. citizens, and completed the electronic Web-based questionnaire were re-contacted for a second wave of data collection that took place between February 22 and March 30, 2017. Four hundred and twenty-three of the Time 1 respondents (45%, not including those who no longer self-identified as Tea Party supporters) constituted our final panel sample.ii
Participants
The mean age in the panel sample (measured at Time 1) was 55 years (SD = 13); sixty percent of the participants were male (no missing data) and 47.5% had a college or university degree (one missing education specification).
Our research participants indeed disapproved of homosexuals: They indicated their disapproval for homosexuals by specifying whether they regarded homosexuality as ‘something bad or something good’ (7-point Likert-scales from -3 [clearly bad] to +3 [very good] with 0 [neutral] as midpoint). The mean value was significantly below the scale midpoint (i.e., 0), M = -1.35, SD = 1.59, t(419) = -17.74, p < .001, 95% CI [-1.51, -1.20].
When comparing the panel sample and the nonrespondent group (i.e., Tea Party supporters who had participated only at Time 1, N = 516 after exclusion of 20 self-identified homosexuals or bisexuals), we observed that the participants in the panel sample were somewhat older (nonrespondent group: Mage = 49 years, SD = 15, p < .001), and that the proportions of men and persons with a college or university degree were higher in the panel sample (nonrespondent group: 50% were male and 37% had a college or university degree, ps ≤ .003).
Measures
As predictor variables, we assessediii three sources of respect at Time 1 and respect for a disapproved outgroup at Time 1 as well as at Time 2. Specifically, respect from society as a whole, from (the disapproved outgroup of) homosexuals in the U.S., as well as from the (wider ingroup of) conservativesiv in the U.S. were measured each with two items adapted from previous research (Simon, Eschert, et al., 2019; Simon & Grabow, 2014; Simon & Schaefer, 2016, 2018). Respondents provided straightforward ratings whether “the U.S. population in general”, “homosexuals in the U.S.”, or “conservatives in the U.S.” respect them (i.e., direct respect measurement: “Are Tea Party supporters respected by the following groups?”). In addition, they specified the extent to which they felt recognized as equal fellow citizens by the members of these groups (i.e., equality recognition: “Are Tea Party supporters recognized as equal fellow citizens by the following groups?”)v. Responses were provided on 5-point Likert-scales ranging from 0 (not at all) to 4 (very much/absolutely) and the two scores were averaged to create a single index for each respect measure (rSBs ≥ .66, ps < .001).
At time 2, we measured respect for “homosexuals in the U.S.”, again, with the two items (direct respect measurement: “How much do you respect the following people?”; equality recognition: “Do you view the following people as equal fellow citizens?”), both on 5-point Likert-scales from 0 [not at all] to 4 [absolutely]). Time 1 respect for homosexuals served as a control for Time 2 respect for homosexuals (lagged criterion, rSBs ≥ .67, ps < .001).
At Time 1, the items for the direct respect measurement preceded the items for equality recognition. The groups were presented in the following order: population in general, conservatives, homosexuals. At Time 2, the presentation order of the items and the groups was randomized. All scales were accompanied by a no-answer option (I cannot or do not want to answer), the selection of which was treated as a missing value.
Results
As main tests of our hypotheses, we performed longitudinal path analyses with observed variables and maximum-likelihood estimation (Mplus; Muthén & Muthén, 1998–2017). We conducted path analyses with manifest variables. To construct measures for our constructs, we computed scale scores by aggregating over item responses. Means, standard deviations, and intercorrelations are presented in Table 1. In each analysis, age (z-standardized), sex (0 = female, 1 = male), income (z-standardized), and education (0 = no university or college degree, 1 = university or college degree) were included as Time 1 control variables.
Table 1
Means, Standard Deviations, and Intercorrelations for Tea Party Supporters in the United States – Study 1
| Measure | M | SD | n | 5 | 6 | 7 | 8 | 9 |
|---|---|---|---|---|---|---|---|---|
| 1. Age (years) | 54.92 | 12.80 | 421 | .05 | -.20 | .13 | -.07 | -.04 |
| 2. Sex (59.6% male) | 423 | .08 | -.04 | .04 | -.07 | -.02 | ||
| 3. Education (47.5% college or university degree) | 422 | -.05 | .02 | .00 | .07 | .09 | ||
| 4. Income ($ U.S.) | 67,975 | 50,065 | 400 | -.04 | -.08 | -.02 | .08 | .09 |
| 5. Societal respect, Time 1 | 1.98 | 0.84 | 418 | .32 | .48 | .07 | .02 | |
| 6. Respect from homosexuals, Time 1 | 0.84 | 0.97 | 377 | .00 | .33 | .30 | ||
| 7. Respect from conservatives, Time 1 | 2.54 | 0.98 | 417 | .02 | -.07 | |||
| 8. Respect for homosexuals, Time 1 | 2.36 | 1.27 | 421 | .71 | ||||
| 9. Respect for homosexuals, Time 2 | 2.42 | 1.20 | 422 |
Note. For all variables, a higher score indicates a higher level of the construct in question, except for sex (0 = female, 1 = male). For all respect variables, scores could range from 0 to 4. Correlation coefficients in bold are significant (ps ≤ .05, two-tailed).
We specified a model in which Time 1 respect from society as a whole, from homosexuals, and from conservatives predicted Time 2 respect for homosexuals. Time 1 respect for homosexuals, as the lagged criterion, served as an additional control variable. We thus followed the frequently used procedure of Heise (1970), using auto-regression to control the effects of interest for the Time 1 measure of the criterion. By this inclusion of the lagged criterion, we went beyond a mere predictive relationship and moved closer to a causal analysis of the role of receiving respect.
As shown in Figure 1, the hypothesized (intergroup respect reciprocity) effect of respect from homosexuals on respect for homosexuals was significant, b = 0.109, SE = 0.052, β = 0.088, p = .036. Respect from the conservative ingroup, in contrast, decreased respect for homosexuals, b = -0.088, SE = 0.048, β = -0.072, p = .070.vi Respect from society as a whole did not affect respect for homosexuals, b = -0.033, SE = 0.061, β = -0.023, p = .586. All influences of sociodemographic variables were nonsignificant (ps ≥ .260).
Figure 1
Antecedents of respect for homosexuals in a sample of U.S. Tea Party supporters (N = 423).
Note. For better legibility, correlations and influences of sociodemographic variables are not shown (see text).
†p ≤ .10. *p ≤ .05. **p ≤ .01. ***p ≤ .001 (all two-tailed).
To examine whether the results were influenced by attrition, we repeated the longitudinal analysis using full-information maximum likelihood estimation (FIML) in combination with auxiliary variables on the basis of all the Time 1 respondents, including those that exclusively participated at Time 1. By this procedure, we assured that also the respondents who only participated at Time 1 contributed to the estimation of the covariance structure or relationships between the variables. Ten auxiliary variables (e.g., willingness to cooperate with homosexuals; evaluation of conservative values) were selected due to their correlations with both the predictors and the criterion. The pattern of results remained largely unchanged. As hypothesized, respect from homosexuals increased respect for homosexuals, b = 0.090, SE = 0.051, β = 0.076, p = .077, whereas respect from the conservative ingroup decreased it, b = -0.091, SE = 0.048, β = -0.078, p = .056. Respect from society as a whole did not affect respect for homosexuals, b = -0.020, SE = 0.060, β = -0.015, p = .736. (To complete the analytical picture of both Study 1 and Study 2, we also used an alternative way of analysing our longitudinal data specifying a multilevel model, see Appendix in the Supplementary Materials).
In order to examine whether there were reverse effects from the criterion variable to the main predictor variables, we supplemented our analyses by a full cross-lagged model. This model comprised the variables of our main panel model. That is, respect for homosexuals at Time 2 was predicted by the lagged criterion, by the four control variables, and by respect from homosexuals, from conservatives, and from society as a whole at Time 1. In addition to this, the full model included further criteria, namely Time 2 respect from homosexuals, from conservatives, and from society as a whole. Each criterion was predicted by the same set of predictors that were employed in our main model. In this full cross-lagged model, the general pattern of results for our main model remained unchanged. Additionally, there was an effect of respect for homosexuals on respect from homosexuals, b = 0.083, SE = 0.033, β = 0.106, p = .014. Respect for homosexuals also negatively predicted respect from conservatives, b = -0.088, SE = 0.032, β = -0.122, p = .006. There was no effect of respect for homosexuals on respect from society as a whole, b = 0.006, SE = 0.034, β = 0.009, p = .850.
Discussion
All of our analyses indicated that Tea Party supporters in the United States who experienced increased respect from the disapproved outgroup of homosexuals also respected this particular outgroup to a higher degree (H1: Intergroup Respect Reciprocity Hypothesis). The demonstration that respect from an outgroup can predict respect for this outgroup, while controlling for the lagged criterion, points in the direction of a causal effect. For societal respect, in contrast, there was no effect on Tea Party supporters’ respect for homosexuals (H2: Societal Respect Reciprocity Hypothesis). Respect from their conservative ingroup decreased respect for the disapproved outgroup of homosexuals (H3: Ingroup Respect Outgroup Disrespect Hypothesis).
We also observed reverse effects. Respect for homosexuals increased (perceived) respect from homosexuals and decreased (perceived) respect from conservatives. The former effect can be interpreted as another promising sign of mutual reinforcement in a respect dynamic. The latter effect suggests that respectful interactions with a disapproved outgroup can undermine respect from the ingroup, possibly via reducing one’s prototypicality in the ingroup (Turner et al., 1987) when violating conservative ingroup norms.
Given their conservative political agenda, Tea Party supporters are likely to see themselves as representatives of the majority population. In Study 2, we therefore tested whether the effects hold for minority group members as well.
Study 2
We suspect that the ineffectiveness of societal respect in Study 1 may have been due to Tea Party supporters’ (self-proclaimed) status as majority representatives. For minority members, in contrast, societal respect should pave the way for respect towards outgroups as suggested by Simon & Grabow (2014). Study 2 represents a reversal of Study 1, in that members of the German LGBTI community now serve as research participants, while supporters of a conservative political group, namely the German populist right-wing political party “Alternative für Deutschland” (AfD, Alternative for Germany), serve as the disapproved outgroup. The AfD is mainly known for its anti-immigration agenda, but it also advocates traditional gender roles and conservative family values. Amongst possible outgroups, we have chosen the AfD as the most extreme amongst the major political parties that support conservative values. During Time 1 of the study, the AfD reached between 10% and 15% in opinion polls on the federal level (infratest dimap, 2019).
Method
Procedure
Research participants were recruited via mailing lists of various organizations within, or associated with, the LGBTI community in Germany as well as via mailing lists assembled in prior research concerned with LGBTI issues (e.g., Simon & Grabow, 2014)vii. We conducted the first wave of data collection between January 15 and May 12, 2016. Nine hundred and twenty German citizens who self-identified as members of the LGBTI community in Germany completed the electronic Web-based questionnaire. The second wave of data was collected between December 9, 2016, and February 6, 2017. Five hundred and seventy-six German citizens who self-identified as members of the LGBTI community in Germany completed the Time 2 questionnaire. We were able to match 262 respondents that had participated at both waves using anonymous personalized codes.
Participants
The panel sample of Study 2 consisted of 262 German citizens who self-identified as members of the LGBTI community. The mean age in the panel sample was 42 years (SD = 13). Fifty-eight percent of the participants were male (sex assigned at birth; no missing data) and 63% had a university degree (ten missing education specifications).
Our research participants indeed disapproved of supporters of the populist right-wing political party AfD: They indicated their disapproval by specifying whether they regarded the goals of the AfD as ‘something bad or something good’ (7-point Likert-scales from -3, very bad, to +3, very good). The mean rating significantly differed from the scale midpoint (i.e., 0), M = -2.48, SD = 0.82, t(254) = -48.43, p < .001, 95% CI [-2.58, -2.38].
Comparisons between the panel sample and the nonrespondent group (i.e., members of the LGBTI community who had participated only at Time 1, N = 658) revealed that the participants in the panel sample were somewhat older (nonrespondent group: Mage = 40 years, SD = 14, p = .052). Furthermore, the proportions of men and individuals with a university degree were higher in the panel sample (nonrespondent group: 50% were male and 57% had a university degree, ps ≤ .033).
Measures
We used the same measures as in Study 1, that is, as predictor variables, we assessed Time 1 respect from three sources: respect from society as a whole, from supporters of the AfD (outgroup), and from the LGBTI community (ingroup). Each respect measure comprised two items. Specifically, respondents provided ratings whether “the German majority population”, “supporters of the AfD”, or “other LGBTI individuals” respect and recognize them as equals (direct respect measurement: “According to your opinion, are LGBTI individuals in Germany respected by the following groups?”; equality recognition: “According to your opinion, are LGBTI individuals in Germany recognized as equal fellow citizens by the following groups?”). Responses were provided on 5-point Likert-scales ranging from 0 (not at all) to 4 (completely), and the two scores were averaged to create a single index for each respect measure (rSBs ≥ .75, ps < .001).
The items for the direct respect measurement preceded the items for equality recognition. The groups were presented in the following order: society as a whole, AfD supporters, members of the LGBTI community. All response scales also ranged from 0 (not at all) to 4 (completely).
Results
As main tests of our hypotheses, we performed longitudinal path analyses with observed variables and maximum-likelihood estimation (Mplus; Muthén & Muthén, 1998–2017). We conducted path analyses with manifest variables. Means, standard deviations, and intercorrelations are presented in Table 2. In each analysis, age (z-standardized), sex (0 = female, 1 = male), income (z-standardized), and education (0 = no university or college degree, 1 = university or college degree) were included as Time 1 control variables.
Table 2
Means, Standard Deviations, and Intercorrelations for Members of the German LGBTI Community – (Study 2)
| Measure | M | SD | n | 5 | 6 | 7 | 8 | 9 |
|---|---|---|---|---|---|---|---|---|
| 1. Age (years) | 41.95 | 12.60 | 260 | -.03 | -.08 | -.03 | -.09 | -.11 |
| 2. Sex (58.4% male) | 262 | .09 | .07 | .03 | -.14 | -.09 | ||
| 3. Education (62.6% college or university degree) | 252 | -.00 | .17 | .10 | .25 | .13 | ||
| 4. Monthly Income (€) | 2,819 | 1,654 | 247 | .03 | -.02 | .10 | -.07 | -.20 |
| 5. Societal respect, Time 1 | 2.55 | 0.68 | 262 | .39 | .30 | .17 | .22 | |
| 6. Respect from AfD supporters, Time 1 | 0.48 | 0.59 | 233 | .13 | .33 | .32 | ||
| 7. Respect from other members of the LGBTI Community (Time 1) | 3.43 | 0.68 | 259 | .03 | -.03 | |||
| 8. Respect for AfD supporters, Time 1 | 1.91 | 1.24 | 257 | .68 | ||||
| 9. Respect for AfD supporters, Time 2 | 2.01 | 1.18 | 260 |
Note. For all variables, a higher score indicates a higher level of the construct in question, except for sex (0 = female, 1 = male). For all respect variables, scores could range from 0 to 4. Correlation coefficients in bold are significant (ps ≤ .05, two-tailed).
As shown in Table 2, perceived respect from AfD supporters had a mean of only 0.48 and a standard deviation of 0.59, which on the one hand illustrates the cleavage between the ingroup and outgroup, and on the other hand hints at the possibility of range restriction.
We specified a model in which Time 1 societal respect, Time 1 respect from AfD supporters, and Time 1 respect from other members of the LGBTI community predicted Time 2 respect for AfD supporters. As in Study 1, the lagged criterion served as an additional control variable, to move closer to a causal analysis of the effects of respect. As hypothesized, respect from society as a whole increased respect for AfD supporters, b = 0.157, SE = 0.090, β = 0.090, p = .083 (see Figure 2), whereas respect from the ingroup decreased it, b = -0.144, SE = 0.090, β = -0.084, p = .083 (Figure 2). Respect from AfD supporters, however, did not affect respect for AfD supporters, b = 0.151, SE = 0.109, β = 0.075, p = .167. All influences of sociodemographic variables were nonsignificant (ps ≥ .136).
Figure 2
Antecedents of respect for supporters of the German populist right-wing political party “Alternative für Deutschland” (AfD) in a sample of German LGBTI community members (N = 262).
Note. For better legibility, correlations and influences of sociodemographic variables are not shown (see text).
†p ≤ .10. *p ≤ .05. **p ≤ .01. ***p ≤ .001 (all two-tailed).
To examine whether the results were influenced by attrition, we repeated the longitudinal analysis using full-information maximum likelihood estimation (FIML) in combination with auxiliary variables on the basis of all the Time 1 respondents, including those that exclusively participated at Time 1. Eight auxiliary variables (e.g., evaluation of the outgroup; identification with Germany) were selected due to their correlations with both the predictors and the criterion. The effect of respect from society as a whole on respect for AfD supporters was no longer significant, b = -0.044, SE = 0.064, β = -0.032, p = .491, while the negative effect of respect from the ingroup remained intact, b = -0.093, SE = 0.051, β = -0.080, p = .065. Respect from AfD supporters now increased respect for AfD supporters, b = 0.119, SE = 0.101, β = 0.075, p = .029.
As in Study 1, we supplemented our analyses by a full cross-lagged model in order to identify reverse effects. This model included Time 2 respect for AfD supporters as a criterion, and, as additional criteria, Time 2 respect from AfD supporters, Time 2 respect from the ingroup, and Time 2 respect from society as a whole. Each criterion was predicted by the same set of predictors that were employed in our main panel model. In this full cross-lagged model, the pattern of results for our main model remained unchanged. Additionally, there was a positive effect of respect for AfD supporters on respect from AfD supporters, b = 0.053, SE = 0.027, β = 0.104, p = .049. Respect for AfD supporters did not affect respect from the LGBTI community, b = 0.023, SE = 0.030, β = 0.040, p = .446, nor respect from society as a whole, b = -0.005, SE = 0.027, β = -0.010, p = .852.
Discussion
In our main panel analysis, we did not find an effect of respect from the outgroup of AfD supporters on respect for this outgroup (H1: Intergroup Respect Reciprocity Hypothesis). This effect, however, emerged in the analysis that included all Time 1 participants and used auxiliary variables in combination with FIML. (It also appeared on the between-person level in a multilevel analysis, see Appendix in the Supplementary Materials, which is similar to the average of the cross-sectional effects.) Responsible for this pattern could be that the power of our main panel analysis was too low due to the smaller number of participants compared to the other two analyses, or that relevant characteristics of participants varied as a consequence of selective attrition. Possibly, these or unmeasured characteristics of the samples are moderators for the effect of respect from outgroups on respect for outgroups. In any case, the effect was less robust for this minority group of LGBTI community members than for Tea Party supporters and therefore needs to be interpreted with caution.
In our main panel analysis, respect from society as a whole did increase respect for AfD supporters, as predicted (H2: Societal Respect Hypothesis). This effect, however, did not reach significance in the sample that included all Time 1 participants. The fact that the effect was not observed in the analysis that included all Time 1 participants points again to the possibility of selective attrition, meaning that relevant characteristics of participants might have been different between the panel sample and the sample of respondents who only participated at Time 1.
Respect from other LGBTI members (H3: Ingroup Respect Outgroup Disrespect Hypothesis) reduced respect for the outgroup of AfD supporters in our main panel analysis. This decrease in respect was also observed in the analysis including all Time 1 participants, using auxiliary variables and FIML.
We also observed a reversed effect. Respect for a disapproved outgroup positively influenced respect from this outgroup. This effect can be interpreted as a sign of mutual reinforcement within a respect-fostering dynamic.
General Discussion
The research in this article examined the role of respect received from others in the development of respect for disapproved outgroups. We focused on respect from three sources, namely from a disapproved outgroup, from society as a whole, and from an ingroup. Our first hypothesis states that group members who feel respected by an outgroup respond in kind by increasing their respect for that outgroup (H1: Intergroup Respect Reciprocity Hypothesis). The second hypothesis is that group members who feel respected by society as a whole increase their respect for outgroups (H2: Societal Respect Hypothesis). According to the third hypothesis, group members who feel respected by fellow ingroup members decrease their respect for disapproved outgroups (H3: Ingroup Respect Outgroup Disrespect Hypothesis). We will now discuss the overall pattern of results with its converging as well as differentiating aspects, before viewing the results in light of the two differing social and political contexts. We will conclude with the wider implications of the findings for theories on respect and identity, for future research agendas, and for practical consequences.
The Observed Effects of Respect From Three Different Sources on Respect for a Disapproved Outgroup
When looking at the effect of respect received from a disapproved outgroup, we observed that, among supporters of the Tea Party movement in the United States, respect from an outgroup increased respect for that outgroup. For members of the LGBTI community in Germany, this effect was not observed for our main panel analysis, but for the analysis including all Time 1 participants. Thus, there was fair evidence for the Intergroup Respect Reciprocity Hypothesis (H1) from two different national and political contexts. This finding supplements previous studies that pointed to a negative effect of respect from outgroups on anti-outgroup attitudes (Simon and Grabow, 2014) and a positive effect on respect for outgroups (Simon & Schaefer, 2018). However, these previous findings were only conducted with minorities in Germany as participants, whereas the present work extends these findings by providing corroborating evidence with new samples that are situated in different contexts.
The fact that the effect was more robust for the former sample than for the latter sample is possibly attributable to differences in the two contexts. In the political arena of Germany, the outgroup of AfD supporters is a group which is perceived to be outside of the normative limits by the mainstream in Germany. Consequently, it is normatively widely acceptable or even demanded in the German mainstream to disapprove of AfD supporters. As indicated, perceived respect from AfD supporters had a low standard deviation and a low mean, which on the one hand hints at range restriction, and on the other hand illustrates the cleavage between the ingroup and outgroup, whose width is likely to pose a challenge for a bridging effect of respect.
We also investigated the effect of respect from society as a whole on respect for disapproved outgroups (H2: Societal Respect Hypothesis). For the members of the Tea Party movement, receiving societal respect had no effect on respect for a disapproved outgroup. For members of the LGBTI community as a minority group, however, respect from society fostered respect for an outgroup in our main panel analysis. This suggests that society has a role to play in respectful intergroup relations; at least in some contexts. The observed effect corroborates earlier studies which suggested that respect from society decreases anti-outgroup attitudes (Simon & Grabow, 2014), and that societal recognition increases recognition for outgroups (Schaefer & Simon, 2020). In our analysis with all Time 1 respondents, however, this effect did not emerge. This could mean that the effect of societal respect needs some time and/or cumulative experiences to develop, and/or that the group of respondents who only participated at Time 1 have different characteristics from that of the panel sample.
With regard to the question why this societal effect could not be observed for the Tea Party supporters in Study 1, it is likely that group members who perceive themselves as being part of the societal majority can take respect from society for granted. As a result, being respected by society should be less surprising and less valuable for (self-proclaimed) majority members than for minority members.
Our examination of the effect of respect from other ingroup members indicated that – for both the Tea Party supporters and the LBTI community members – ingroup respect decreased respect for a disapproved outgroup, thereby supporting the Ingroup Respect Outgroup Disrespect Hypothesis (H3). This effect is in line with self-categorization theory (Turner et al., 1987), insofar as individuals who feel respected by fellow ingroup members should experience an endorsement and heightened salience of their group membership, which in turn should increase ingroup favouritism and outgroup devaluation. Additionally, a heightened salience of a group membership should reinforce self-stereotyping and adherence to group norms. When group norms define the relationships to certain outgroups with opposing ideologies as negative, heightened endorsement or salience of a group membership should increase negative attitudes towards these outgroups.
Thus, our two samples delivered converging evidence that ingroup respect leads to a decrease of outgroup respect, but needs some time to unfold. From a normative perspective, it appears to be regrettable that ingroup respect can decrease respect for a disapproved outgroup. This element of disapproval, however, should be dependent on the specific norms of the ingroup. If increased ingroup respect leads to increased ingroup identification and increased self-stereotyping, ingroup norms are likely to mediate the effect of ingroup respect on decreased outgroup respect. Consequently, to encourage respectful intergroup relations, a change of ingroup norms and a change of the ingroup’s definition of its relationships to other groups would be a desirable path. Group norms and self-definitions in relation to other groups, in turn, are likely to be affected by respect from outgroups and, at least for minority groups, respect from wider society.
Future Research
We had included in our study design two time points and controlled for lagged criteria, to move closer to causal analysis. While we see advantages of examining the development of respect dynamics in two different national and political contexts, we do not regard our studies as substitutes for experiments, but as supplementary investigations that track these effects with natural groups in their given environments. As field studies can never conclusively rule out causation by third variables, future research should examine these effects further in more controlled experimental settings.
An agenda for future research can also be derived on the basis of potential interrelationships between respect, identification, and tolerance. Previous research found that respect from (disapproved) outgroups predicts respect for these outgroups as well as tolerance towards these outgroups (Simon, Eschert, et al., 2019; Simon & Schaefer, 2018). Future research could investigate to what degree respect from other sources (ingroup and society) has similar effects.
Given the role of social identifications in social behaviour (Turner et al., 1987), it could further be examined how respect affects social identifications. While Schaefer and Simon’s (2020) findings suggest that recognition from society increases societal identification, which in turn increases recognition for outgroups, future research could investigate the effects of respect from other sources and how identifications affect respect for outgroups and tolerance towards outgroups. A more fine-grained research programme could also examine to what degree changes in group categorizations, group salience, self-stereotyping, and adherence to group norms is associated with and fosters outgroup respect and tolerance.
Differences Between the National and Political Contexts in Which the Two Samples are Embedded
Part of the differences that we observed between members of the LGBTI community and supporters of the Tea Party may be attributable to the different national contexts. Central dimensions on which these contexts can be differentiated concern the domain of values. Schwartz (2008) compared various nations along value dimensions, using a classification system consisting of seven dimensions to allocate societies. In this multi-dimensional system, Eastern and Western German societies are high in egalitarianism, whereas the United States is characterized as being particularly high in mastery. The degree of attaching relative importance to egalitarianism is also likely to influence the dynamics of mutual recognition, insofar as feeling recognized as equals should have different implications in different contexts. Given the more individualistic and less egalitarian nature of the U.S. culture, these differing value priorities could imply that societal respect in terms of equality recognition as opposed to, for example, achievement recognition has less significance in U.S. society than in German society. Such a difference in significance might further contribute to our observation that societal respect had an effect solely in the context of Germany. In order to disentangle the effects of societal contexts on the one hand and specific position of the sampled group in the respective society on the other hand, it is desirable that future research would compare several minority and majority groups in different national and regional contexts.
By comparing multiple countries, a potential weakness of our first study could also be addressed, namely that the position of the Tea Party movement in the United States is not unambiguous. While the Tea Party movement is a numerical minority, their self-understanding as representing the interests of “proper” citizens appears to imply a self-definition as the societal majority. A comparison between different minority and majority groups in different contexts could more clearly distinguish between, on the one hand, effects attributable to numerical vs. self-proclaimed minority or majority positions of groups, and, on the other hand, effects attributable to different value orientations of different national and political contexts.
Conclusion
The observation that receiving respect from a disapproved outgroup facilitates respect for that particular outgroup suggests that giving respect is a central step for groups toward receiving respect for themselves. Importantly, the concept of respect here does not require to yield to an outgroup (e.g., to a majority outgroup), but to regard and treat outgroup members as equal fellow citizens. In both of our samples, we also found the reverse longitudinal effect that respecting an outgroup increases the feeling of being respected. Unless this pattern is produced by a third variable (e.g., an increased sensitivity to respect in general), this result is insofar encouraging as it suggests that giving and receiving respect can set in motion a mutually reinforcing spiral of respect.
The importance of the present research on the development of respect for disapproved others becomes even clearer when looking at the further consequences of respect. Respect longitudinally predicts and may in fact increase tolerance for disapproved others, which is a prerequisite for the functioning of a modern, plural society (Simon, Eschert, et al., 2019). Our results do not only point to the responsibility of both majority and minority members for respectful societal relations, but also tie in with studies showing that common superordinate identities are facilitated by the experience of being respected (Renger, Renger, Miché, & Simon, 2017; Simon, Mommert, & Renger, 2015). A common ingroup identity, in turn, has been demonstrated to have the potential of reducing intergroup bias (Gaertner et al., 2000). This is not to say that a common superordinate identity should replace subgroup identifications. In fact, previous research indicated that natural groups frequently feel close ties to their subgroups and might therefore experience identity threat when being prompted to give up their subgroup identity (Dovidio, Gaertner, & Saguy, 2007). Additionally, Glasford and Dovidio (2011) suggested that a common ingroup identity can decrease the motivations of disadvantaged groups to struggle for social change. In contrast to a common ingroup identity, a dual identity, or identity as a “different equal” (Simon, 2017, 202024), which combines a subgroup identification with a superordinate group identification, has the advantage of reducing identity threat (Dovidio et al., 2007). Additionally, dual identity preserves both motivation for social change and willingness for intergroup contact (Glasford & Dovidio, 2011).
In conclusion, the present research affords the hopeful prospect that a positive respect dynamic is set in motion when group members feel respected by outgroup members so that they feel encouraged, if not obligated, to respond in kind. They may then even form a common superordinate identity or a dual identity (Dovidio et al., 2007; Simon et al., 2015), which can operate as additional sources of mutual respect and thus enable constructive social and political life in plural societies characterized by a multiplicity of beliefs, preferences, or practices.