Sources of Differences in Right-Wing Authoritarianism and Social Dominance Orientation from Adolescence to Adulthood: A Multi-Cohort Twin Family Study

doi:10.21203/rs.3.rs-6148484/v1

Download PDF

Research Article

Sources of Differences in Right-Wing Authoritarianism and Social Dominance Orientation from Adolescence to Adulthood: A Multi-Cohort Twin Family Study

https://doi.org/10.21203/rs.3.rs-6148484/v1

This work is licensed under a CC BY 4.0 License

Journal Publication

published 03 Nov, 2025

Read the published version in Behavior Genetics →

You are reading this latest preprint version

Past research indicates that both genetic and environmental sources contribute to two core dimensions of individual differences in socio-political attitudes: Right-Wing Authoritarianism (RWA) and Social Dominance Orientation (SDO). However, less is known about differences between the sources of variance in RWA and SDO, and how these sources vary by age. Based on data from 1440 twin families including 1198 complete twin pairs, 435 non-twin full siblings, and 2016 parents of twins from the German TwinLife project, nuclear twin family modeling (NTFM) was used to estimate different genetic and environmental variance components of RWA and SDO, and to examine the extent to which the variance components vary across three age cohorts of twins (average age 15, 21, and 27 years). We hypothesized increasing levels of genetic differences across age cohorts, reflecting more active genotype-environment transactions with development, and declining levels of passive genotype-environment correlation. The results showed some differences regarding the sources of variance in RWA and SDO. RWA showed stronger environmental effects shared in families due to intergenerational non-genetic transmission from parents to offspring, and both passive genotype-environment covariance and additive genetic variance declined across cohorts, while nonadditive genetic differences increased. For SDO, passive genotype-environment covariance components were generally negligible, while additive genetic variance components tended to increase across cohorts. These findings were only partially consistent with our expectations. Counterintuitively, the influence of within-familial environmental factors on RWA were larger in adulthood than adolescence. The results are discussed against a background of developmental theories and potential cohort differences.

Right-Wing Authoritarianism

Social Dominance Orientation

multi-cohort nuclear twin family model

genotype-environment correlation

Right-Wing Authoritarianism (RWA) and Social Dominance Orientation (SDO) have been proposed as two personality-related constructs that describe important individual differences in socio-political attitudes. RWA is thought to encompass an individual’s preference for collective security while maintaining established social conventions and rejecting social change (Altemeyer, 1981), whereas SDO captures a preference for intergroup stratification while favoring one’s own group over other groups (Pratto et al. 1994). The two constructs show some theoretical and empirical overlap, but are also said to provide unique information on socio-political attitudes (Duckitt and Sibley, 2010; Nacke and Riemann, 2023).

As individual differences in both RWA and SDO appear to substantially contribute to variance in prejudice and discrimination (e.g., Duckitt and Sibley 2010; Kandler, Lewis et al. 2015), examining the sources of such differences is clearly essential. Although previous research generally indicates that RWA and SDO are influenced by both genetic and environmental factors (de Vries et al. 2022; Hatemi et al. 2014; Kleppestø et al. 2024), questions remain as to the specific nature of those factors and how they may combine and relate to each other, especially during the formative years of socio-political development. A related question that remains unanswered pertains to whether the form those influences take varies through the life course. In the present study, we critically addressed those questions. In doing so, we expanded on previous research (1) by using a nuclear twin family design that included data from twins, their non-twin siblings, and their parents, and (2) by analyzing data from three different age cohorts of twins that spanned the period from adolescence to young adulthood—a time that is deemed to be a crucial stage in life for the formation of socio-political orientations (Jennings et al. 2009).

Different Sources for RWA and SDO?

Given their common and unique variance, RWA and SDO might show some similar as well as some different properties regarding their genetic and environmental sources. Several studies have consistently reported moderate to substantial genetic influences on RWA, accounting for 30–50% of its variance, while environmental influences shared by twins have explained 10–30% (e.g., Funk et al. 2013; Kleppestø et al. 2024; Lewis and Bates 2014; Ludeke and Krueger 2013; McCourt et al. 1999; Nacke and Riemann 2023). Somewhat lower genetic and shared environmental contributions (on average about 30% and 10% of the variance, respectively) have been found for SDO (de Vries et al. 2022; Kleppestø et al. 2024; Kandler, Bell et al. 2015; Kandler 2015; Nacke and Riemann 2023). Hence, individual-specific environmental factors (including error of measurement) appear to account for most of the differences in SDO.

After controlling for error of measurement, familial resemblance based on reliable latent variable scores across self- and informant reports was found to be considerable for both RWA and SDO (Kandler et al. 2016). In particular, the within-generational similarity of monozygotic (MZ) and dizygotic (DZ) twins, corrected for random error of measurement and systematic rater specificity, indicated substantial environmental influences shared by twins on variance in both RWA (r_MZ = .88 and r_DZ = .60) and SDO (r_MZ = .66 and r_DZ = .62). However, the correlations between twins’ parents were also substantial, being r = .45 for RWA and r = .42 for SDO. This is indicative of assortative mating, which occurs if parents have not mated in an entirely random fashion, but instead on the basis of their phenotypic resemblance to one another (e.g., Briley et al. 2019). Assortative mating increases the genetic resemblance of twins’ parents and thus the genetic resemblance of their DZ twin offspring, but not the genetic resemblance of their already genetically identical MZ twin offspring. If assortative mating is of relevance but is not accounted for in variance decomposition analyses, shared environmental influences will be overestimated and additive genetic influences underestimated. A recent meta-analysis yielded considerable partner correlations for political values (r_meta = .58), supporting the idea that assortative mating could play a crucial role for RWA and SDO (Horwitz et al. 2023). Nuclear twin family models (NTFMs) based on data from twins and their parents allow one to examine the resemblance of twins’ parents on the trait of interest, and thus to account for assortative mating effects. The latter effects were evident in the NTFM analyses of Kandler and colleagues (2016), which indicated that estimates of twins’ shared environmental influences on variance in RWA and SDO were primarily sibling-specific or twin-specific.

RWA and SDO appear to have substantial genetic but almost no to little environmental overlap, indicating that environmental effects are rather construct-specific (Lewis and Bates 2014; Nacke and Riemann 2023). Kandler et al. (2016) found that controlling for their shared variance with RWA, individual differences specific to SDO were not significantly influenced by genetic factors, but primarily attributable to unique environmental factors not shared between parents and their offspring. For RWA, by contrast, genetic factors were found to form a considerable source of individual differences. However, other studies have concluded that the level of RWA might be more malleable by societal circumstances than is the case with SDO (Zubielevitch et al. 2023). Given this seeming inconsistency in the literature, we aimed to determine whether or not certain genetic and environmental factors are differently important for RWA and SDO and also whether one or both constructs are subject to unique influences.

Different Sources for Different Ages?

Not only different genetic and environmental sources but also different kinds of nonrandom links between genetic and environmental factors could be of relevance to the formation of individual differences in RWA and SDO. Such nonrandom links include, for instance, passive and (re)active genotype-environment correlations (rGE). Passive rGE refers to situations in which children are exposed to environmental contexts that their parents provide that are affected by their parents’ genetically influenced characteristics and behaviors. Since children share substantial portions of their genotype with their parents, there may be a correlation between the children’s genotypes and the nature of the environments in which the children are raised. In the case of RWA and SDO, parents may provide genetically influenced environmental cues that reinforce their children’s natural inclinations toward those two constructs. Developmental theories suggest a declining impact of within-family influences including passive rGE on individual differences during psycho-social development, likely due to increasing autonomy and self-determination (Scarr 1992, 1993; Scarr and McCartney 1983). Therefore, in the late teen years diminishing effects of environments shared by twins and lower passive rGE can be expected, especially after children have grown up and left the familial household.

Active or reactive rGE (the latter is also referred to as evocative rGE) occur in situations in which individuals’ genetically-based predispositions lead them to choose environments that provide a good fit for those predispositions, or lead them to provoke or construe their environmental circumstances in a particular way (Plomin et al. 1977; Kandler et al. 2021). For example, a politically conservative young adult may choose news sites or friendship circles in which their conservative leanings are welcomed and affirmed. These sorts of rGE are thought to become increasingly relevant during adolescence and early adulthood, as individuals in this period of life begin to physically and psychologically distance themselves from the parental home environment and transact with their broader environments more autonomously (Kandler et al. 2019; Scarr and McCartney 1983).

One study that used cross-sectional data from twins and their parents in a NTFM to examine the genetic and environmental contributions to the variance in these constructs found evidence of passive rGE for RWA but not for SDO, even though the sample included a broad age-range of adult participants (Kandler et al. 2016). Another study that examined correlates of both RWA and SDO, such as general left-right political orientations across two age-groups, found a decrease in the extent of passive rGE from adolescence to young adulthood (Hufer et al. 2020).

Also, Hufer et al. (2020) reported a greater additive genetic variance component for political orientation for young adults than for adolescents, which seems consistent with the supposed increasing genetically driven extrafamilial socialization through greater active and evocative rGE during young adulthood (Scarr and McCartney 1983; Kandler et al. 2021). This increase in additive genetic effects from adolescence to young adulthood is also consistent with findings on other socio-political attitudes. For overall liberalism-conservatism, Hatemi et al. (2009) reported minimal additive genetic variance in adolescents but substantial levels of it after age 21. A similar pattern was observed by Eaves et al. (1997) for political conservatism.

As RWA and SDO show parallels and links to general left-right political orientations in terms of etiology, comparable age trends can be expected for RWA and SDO. A recent cohort-sequential study on the developmental trends in RWA and SDO found that, on average, the levels of both of these socio-political orientations tended to increase with age, although cohort differences and, thus, contextual influences seemed to play a greater role for RWA in this respect (Zubielevitch et al. 2023). However, research from a developmental perspective on the extent to which genetic and environmental variance and covariance account for individual differences in RWA and SDO has, to our knowledge, never been conducted. The present extended twin family study with twins at three different developmental stages (adolescence, emerging adulthood, and young adulthood) endeavored to fill that gap.

The Current Study

The key aims of this research were to disentangle different genetic and environmental sources of individual differences in RWA and SDO, and to determine whether the importance of those sources differed between the two constructs and across three age cohorts. The youngest of these three age cohorts consisted of twins born in the years 2003–2004 (M_age ≈ 15), which we designated as “adolescents.” The intermediate cohort, with twins born in 1997–1998 (M_age ≈ 21), was classified as “emerging adults.” The oldest cohort, with twins born from 1990–1993 (M_age ≈ 27), was described as “young adults.” Non-twin siblings that were roughly the same age were also investigated. The additional consideration of data from twins’ parents, who are at a different developmental stage than their offspring, is crucial when investigating different sources of siblings’ similarities and differences in the offspring generation. These sources include (1) within-generational sibling- and twin-specific similarities due to nonadditive genetic and age- or generation-specific environmental influences, (2) the role of intergenerational genetic and environmental transmission, (3) nonrandom links between genetic factors and family environment (i.e., passive rGE), while (4) taking nonrandom spouse similarity (i.e., assortative mating) of twins’ parents into account (Hufer et al. 2020; Keller et al. 2009).

Unlike classical twin models, NTFMs can consider the simultaneous relevance of nonadditive genetic factors and environmental influences shared by twins, which otherwise would result in biased estimates of genetic and environmental components (Instinske and Kandler 2024; Keller et al. 2010). With the use of a more fine-grained NTFM analysis, we looked for potential differences in the sources of variance in RWA and SDO, taking construct-specific variance in RWA and SDO as well as variance due to measurement error into account.

Fitting multi-cohort nuclear twin family models (MC-NTFMs) based on the three different age cohorts further allowed us to compare the genetic and environmental contributions to the variance from adolescence to young adulthood. Based on the findings from previous research discussed above, we hypothesized that the role of passive genotype-environment correlation decreases with higher age across the three age groups for RWA (Hypothesis 1a) and SDO (Hypothesis 1b). In addition, we hypothesized that at the same time the variance attributable to genetic sources increases with age across the three age cohorts for RWA (Hypothesis 2a) and SDO (Hypothesis 2b), which would be indicative of greater effects originating from active and evocative genotype-environment transactions. Both hypotheses were preregistered as part of an overarching project (https://osf.io/6asfj/).

Study Design and Data Collection Procedures

We analyzed data from the German twin family panel TwinLife, a genetically informative, longitudinal study that tracks individual development and social inequality through the life course (Diewald et al. 2024; Hahn et al. 2016). MZ and same-sex DZ twin pairs from four different birth cohorts as well as their parents and, if available, non-twin siblings (when old enough) were interviewed. Participants were randomly selected based on national population registration offices, which produced a representative sample of German twin families. The TwinLife core project encompasses five waves of data collection that began in 2014 and concluded in 2024 (see also www.twin-life.de/documentation/).

Participants have been surveyed on various constructs, alternating between face-to-face interviews in the household and computer-assisted telephone interviews. At the time of the third face-to-face interview, conducted in the period from 2018 to 2020, information on RWA and SDO was assessed for participants 13 years of age and older. The constructs were not measured again in any other data wave. See Rohm et al. (2023) for details on the study design of TwinLife.

To determine if a twin pair was monozygotic or dizygotic, a standardized zygosity questionnaire was used at the time of the twins’ first participation (Lenau et al. 2017). The questionnaire comprised items concerning the respondent’s perception of the twin’s physical similarity to the respective co-twin. For the two older age-cohorts, the twins’ themselves answered the relevant questions, whereas for the younger age-cohort, the questionnaire was completed by the twins’ parents. Compared to DNA-based classifications, the standardized questionnaire provides high rates of correct zygosity classification, ranging from 92–97% (Lenau et al. 2017). The present study assessed zygosity using the questionnaire data, and those classifications were checked (and corrected where necessary) using DNA information in cases for which such information was available.

Sample

Overall, data from 1440 twin families were available, which involved 5089 different participants who provided self-reports on RWA and SDO. We aimed to use the highest sample size possible based on the data available. Therefore, we included data from twin pairs for which information on RWA or SDO was available for at least one of the co-twins. The resulting sample includes 1270 MZ twins (58% female) from 687 pairs, with 583 pairs for which data from both twins were available, and 1368 DZ twins (53% female) from 753 pairs, with 615 complete pairs. From these twins, 1129 were designated as adolescents (Age: M = 15.04, range: 14–16), 820 as emerging adults (Age: M = 21.09, range: 20–22), and 689 as young adults (Age: M = 27.11, range: 25–29).

In addition to the twins, RWA and SDO self-report data from 435 non-twin full siblings were available, including 197 siblings (48% female) of adolescent twins, 150 siblings (47% female) of emerging adult twins, and 88 siblings (56% female) of young adult twins. From the 1440 twin families, 1173 mothers (81%) and 843 fathers (59%) provided RWA and SDO self-reports. See Table 1 for numbers of twins and their family members by twins’ birth cohort and zygosity as well as sex distributions and age statistics.

Table 1

Descriptive Statistics of the Twin Family Sample
	Birth cohort and zygosity
	2003–2004		1997–1998		1990–1993
	MZ	DZ	MZ	DZ	MZ	DZ	Total
N of twin families	248	328	219	237	220	188	1440
N of twin a	242	323	204	219	199	160	1347
N of male twin a	108	166	83	91	77	67	592
N of female twin a	134	157	121	128	122	93	755
N of twin b	244	320	189	208	192	138	1291
N of male twin b	110	166	76	87	76	66	581
N of female twin b	134	154	113	121	116	72	710
N of complete twin pairs	238	315	174	190	171	110	1198
N of non-twin sibs	88	109	76	74	47	41	435
N of male non-twin sibs	40	63	42	38	15	24	222
N of female non-twin sibs	48	46	34	36	32	17	213
N of mothers	235	315	165	190	150	118	1173
N of fathers	168	248	121	145	90	71	843
Age mean of twins	15.04	15.04	21.12	21.05	27.17	26.99	20.08
Age range of twins	14–16	14–16	20–22	20–22	25–29	25–29	14–29
Age mean of non-twin sibs	18.53	18.39	22.39	22.80	27.30	27.32	21.67
Age range of non-twin sibs	13–28	13–27	13–32	13–38	14–38	18–38	13–38
Age mean of mothers	46.78	48.00	52.15	53.15	56.49	57.52	51.22
Age range of mothers	33–58	35–62	43–62	44–67	46–67	46–67	33–67
Age mean of fathers	50.47	51.14	54.91	55.71	59.43	59.46	53.92
Age range of fathers	34–77	38–77	45–68	43–78	47–70	46–77	34–78

MZ: monozygotic; DZ: dizygotic.

RWA and SDO Measures

RWA and SDO were assessed by means of self-reports. RWA was measured using a four-item version of the German Right-Wing Authoritarianism Scale RWA³D (Funke 2005; Hebler et al. 2001; example item: “The real keys to the 'good life' are obedience, discipline, and virtue.”). A German variant (Kämpfe 2002) of the four-item scale for measuring Social Dominance Orientation (Pratto et al. 1994) was used to assess SDO (example item: “If some groups kept to themselves, we would have fewer problems.”). Participants were asked to respond to the items contained in both instruments on a 5-point-Likert scale according to their level of agreement (1: “disagree very strongly” – 5: “agree very strongly”). Two RWA and one SDO items were negatively formulated regarding the construct and thus had to be reversely coded. For more details on the measurement instruments see the TwinLife data documentation website (https://www.twin-life.de/documentation/).

Internal consistency estimates based on Cronbach’s α (Cronbach, 1951) and McDonald’s ω (McDonald, 1999) were α = .59 and ω = .60 for RWA and α = .50 and ω = .54 for SDO. Given the brevity of the RWA and SDO measures and that both measures should encompass the theoretical bandwidth of both constructs, the rather low internal consistency can be treated as acceptable. In addition, confirmatory factor analyses (CFA), allowing for two dimensions representing RWA and SDO as well as systematic residual correlations between the three recoded items to consider systematic method variance, were run. To evaluate overall model fit, the comparative fit index (CFI; Bentler, 1990) and the root mean square error of approximation (RMSEA; Steiger and Lind, 1980) were used, with CFI > .95 and RMSEA < .05 indicating good fit, and CFI > .90 and RMSEA < .08 suggesting adequate fit (Hu and Bentler 1999; Steiger 2007). The CFA supported the two-dimensional structure with acceptable model fit to the data (CFI = .925, RMSEA = .067) and thus provided factorial validity of the RWA and SDO scores.

The RWA and SDO measures were tested for different levels of measurement invariance (Putnick and Bornstein 2016) across the two zygosity groups, the three age cohorts, and five samples representing different groups of family members (i.e., firstborn twins, second-born co-twins, non-twin siblings, mothers, and fathers). We started the measurement invariance (MI) tests with a baseline model that assumed the same factor structure across the different subsamples (configural MI). If the model fit was acceptable, we tested for metric MI that requires invariant item factor loadings across all subsamples. Metric MI is the basic prerequisite to assume that the same constructs have been measured in each subsample and thus the basic prerequisite for variance-covariance decomposition in the NTFM analyses. Nonetheless, we also tested for scalar MI that additionally requires equal item intercepts across all subsamples and strict MI that requires equal item residual variances across subsamples. To compare the model fit of different MI levels, we used the common descriptive criterion to identify lack of MI: A more constrained model should not show a decrease in the CFI value > .01 (Cheung and Rensvold 2002). Table 2 summarizes the MI model tests. In sum, strict MI could be assumed across zygosity groups and metric MI could be assumed across twins’ birth cohorts and the five family members as a basic fundament for the NTFMs.

Intercorrelated RWA and SDO mean scores (r = .42), uncorrelated RWA and SDO factor scores based on the Anderson-Rubin factor-score estimation method (DiStefano et al. 2009), and latent RWA and SDO variable scores corrected for item specificity including random error of measurement were used in the ongoing twin family analyses. Means and variances of these scores for different zygosity groups, birth cohorts of twins, and family members are shown in the supplementary Table S1 (see https://osf.io/mq83d/). Generally, the adolescent siblings tended to show higher RWA and SDO scores compared to older birth cohorts, while the variance in RWA and SDO tended to increase across cohorts.

Table 2

Measurement Invariance (MI) Testing of RWA and SDO Measures
Models	χ²	df	Δχ²	Δdf	p	RMSEA	CFI	ΔCFI
Measurement invariance testing across zygosity groups
Configural MI	427.09	32				.070	.920
Metric MI	442.75	38	15.66	6	.016	.065	.918	.002
Scalar MI	450.95	44	8.20	6	.224	.060	.918	.000
Strict MI	458.40	52	7.45	8	.489	.055	.918	.000
Measurement invariance testing across twins' birth cohorts
Configural MI	416.40	48				.067	.925
Metric MI	445.88	60	29.47	12	.003	.062	.921	.004
Scalar MI	550.65	72	104.77	12	< .001	.063	.903	.018
Strict MI	569.47	88	18.82	16	.278	.057	.902	.001
Measurement invariance testing across family members
Configural MI	457.88	80				.068	.925
Metric MI	513.55	104	55.67	24	< .001	.062	.919	.006
Scalar MI	855.46	128	341.91	24	< .001	.075	.855	.064
Strict MI	937.15	160	81.69	32	< .001	.069	.845	.010

Configural MI: same factor structure across groups; Metric MI: configural MI + equal factor loadings across groups; Scalar MI: metric MI + equal item intercepts across groups; Strict MI: scalar MI + equal residual variances across groups; best fitting MI models based on RMSEA < .08, CFI > .90, and ΔCFI < 0.01 compared to lower-level MI are shown in bold.

Analyses

The preparative psychometric analyses subsumed in the previous section (see the corresponding HTML-file ‘Psychometric Analyses’ at https://osf.io/mq83d/ for the entire output) and the main analyses were run with the statistical software package JASP version 0.18.3 (JASP Team 2024). For the most part, we approached the question of whether the sources of variance in RWA and SDO differed across and between adolescents, emerging adults, and young adults by making use of the three different age cohorts of twins embedded in a nuclear twin family design. Hence, we applied all of our analyses to the entire sample as well as to the three different twin-cohort families separately. In this manner, we investigated whether differences between these groups were discernable, and which corresponding trends emerged.

Family Correlation Analyses

First, we approached the family similarity by calculating correlations of mean scores, factor scores, and latent variable scores of RWA and SDO, respectively, for each set of family member dyads based on the full sample and based on the families from the three different twin birth cohorts separately. The inspection of family similarity provides an initial insight into the roles of genetic and environmental sources of differences in RWA and SDO. Then, structural equation modeling was applied and NTFMs as well as MC-NTFMs were fitted to the data on the basis of full information maximum likelihood estimation to handle missing data (Bentler 2006; Little and Rubin 2019).

Nuclear Twin Family Model Analyses

To investigate the contributions of genetic factors, environmental factors and passive rGE to individual differences in RWA and SDO, NTFM analyses were applied (Heath et al. 1985; see also Keller et al. 2009). Here, we utilized information from monozygotic and dizygotic twins reared together, non-twin full-siblings, and their biological parents to decompose the observed variance in the respective constructs into components due to genetic and environmental sources (see Fig. 1).

Genetic sources can be decomposed into additive genetic factors (A) and nonadditive genetic factors (NA), with the latter either arising from allelic dominance within gene loci or gene-gene interactions, also called emergenesis or epistasis (Lykken 2006). All genetic factors are perfectly correlated between MZ twins, whereas A factors’ correlation is .5 for other first-degree relatives and NA factors are not correlated among other relatives, except the correlation of nonadditive genetic factors due to allelic dominance, which is .25 for non-MZ first-degree siblings.

Environmental sources can be broken down into environmental influences shared by twins only (T), environmental influences shared by all siblings (S), family environmental influences (FE) transmitted from parents to their offspring, and individual-specific environmental influences (E) not shared by family members, with the latter including random error of measurement, if not controlled for. In addition, the covariance between A and FE (w) can be regarded as an estimate that represents passive rGE, as w captures the extent to which parental phenotypes are genetically linked to the offspring’s phenotype and associated with the environment provided by the parents (Keller et al. 2009). All these different sources of variance were estimated while considering the effects of assortative mating (µ). This was done because, as noted, assortative mating can increase genetic similarity between parents and their children as well as among their children, except for already genetically identical MZ twins, and can thus affect genetic variance (Heath and Eaves, 1985).

The NTFM does not have sufficient information to estimate, in the presence of each other, nonadditive genetic factors, environmental factors shared by all offspring due to parental influences, sibling-specific common environmental influences not shared with parents, and twin-specific shared environmental influences (Keller and Coventry 2005). For that reason, we tested NTFMs assuming nonadditive genetic effects or sibling-specific environmental influences to be zero, further specifying nonadditive genetic effects as being either due to allelic dominance or due to epistatic gene-gene interaction (epistasis). We compared the three models descriptively to one another using the Akaike Information Criterion (AIC; Akaike, 1987) and the Bayesian Information Criterion (BIC; Schwarz, 1978) to decide which of them fitted the data best. The respective best fitting model was used to derive the single parameter estimates, whose statistical significance was evaluated using the 95% confidence intervals. See Hufer et al. (2020) for more details on the NTFMs used in this study and Instinske and Kandler (2024) for NTFM user tutorials based on the open-source and user-friendly statistical program JASP.

Multi-Cohort Nuclear Twin Family Model Analyses

After investigating the contributions to the variance in RWA and SDO scores for the entire twin family sample, we scrutinized the role of rGE for the three age cohorts of twins. We specified a MC-NTFM using the three age cohorts as different groups. To test whether the contribution of passive rGE to individual differences decreased between adolescence and adulthood (Hypothesis 1) and to determine whether the relevance of active or evocative rGE increased across the three age cohorts (Hypothesis 2), we examined whether a MC-NTFM in which all model parameters were constrained to be equal across the three cohorts fitted the data significantly worse than a corresponding MC-NTFM in which parameters were estimated freely for different cohorts using the likelihood ratio test (Δχ²). If so, we inspected whether the model parameter w (passive rGE) declined with age across the three cohorts and whether genetic variance components A and NA increased. In addition, we explored which of the other model parameters might be responsible for potential differences between cohorts. See also https://osf.io/mq83d/ for lavaan codes of the NTFMs and MC-NTFMs in the corresponding folder.

Family Correlations

We estimated the correlations of different family member dyads for correlated mean scores, orthogonal factor scores, and latent variable scores of RWA and SDO. This was done for the full sample and the three separate family groups with different twin birth cohorts. All detailed family correlations including 95% confidence intervals can be found at https://osf.io/mq83d/ in the HTML-file ‘Family Correlations’.

Figure 2 summarizes the family correlations for the family dyads relevant in the nuclear twin family design. The parent-offspring and the twin-nontwin sibling correlations are weighted averages across all potential combinations (i.e., across parent-twin a, parent-twin b, and parent-nontwin sib, as well as across twin a-nontwin sib and twin b-nontwin sib). The orthogonal factor scores yielded lower correlations, which is not surprising, given that the common variance in RWA and SDO has been shown to be primarily genetic in nature (Nacke and Riemann 2023) and adjusting RWA and SDO factors for this common component would result in reduced family correlations compared to those based on correlated mean scores (−[.07-.08] for MZ twins and −[.03-.06] for other core family relatives). The latent variable scores provided the highest correlations, because they are not attenuated due to item-specific method variance and random error variance.

The general patterns of family correlations were comparable across all three kinds of measures. Family resemblance and spouse similarity was higher for RWA than SDO, indicating larger influences shared by family members on variance in RWA and larger individual-specific environmental effects on variance in SDO. In all cases, the MZ twin correlations were higher than the DZ twin correlations, indicating contributions of genetic factors to individual differences in both RWA and SDO. The DZ twin correlations tended to be higher compared to the twin-nontwin sibling and the parent-offspring correlations, suggesting environmental influences that are shared by twins only. Assortative mating played a larger role for RWA compared to SDO given the higher mother-father correlation.

An inspection of the cohort-specific family correlations (see Fig. 3) yielded the highest twin correlations in RWA for the oldest (27-year old) cohort of young adult twins compared to emerging young adult (21-year old) and adolescent (15-year old) twins. For SDO, the highest twin correlations were consistently found for the emerging adult twins compared to the other cohorts. These differences indicate lower twin resemblance in RWA and SDO for the adolescent twin cohort compared to adult twins.

The other family correlations did not vary that systematically, except the spouse correlations in SDO, which were substantially higher in families with the oldest twin cohorts. The latter is probably a sample-specific deviation. The NTFMs can deal with those deviations which otherwise would lead to more or less biased estimates of genetic and environmental contributions to the variance in RWA and SDO.

Nuclear Twin Family Modeling

Descriptive model comparisons of different NTFM models either allowing for sibling-specific shared environmental influences or nonadditive genetic effects due to epistasis or allelic dominance yielded fairly consistent results across the different measures of RWA and SDO. The NTFM allowing for nonadditive genetic effects instead of sibling-specific shared environmental influences on RWA variance showed descriptively the best model fit, whereas all three models provided almost identical fit statistics and indices for SDO (see Table 3; additional fit measures and all model parameter estimates including exact p-values and confidence intervals can be found in the NTFMs_RWA_SDO.html file at https://osf.io/mq83d/).

Based on the model parameter estimates of the best fitting model, we calculated the variance components of RWA and SDO measures as follows: The phenotypic variance can be decomposed into \(\:a²q+n²+x+t²+s²+e²+2aw\), where \(\:a²q\) reflects the genetic component due to additive genetic effects, \(\:n²\) represents the genetic component due to nonadditive genetic effects, \(\:x=m²+f²+2mf\mu\:\) (if phenotypic variance is standardized, i.e. \(\:VAR=1\)) is the environmental variance attributable to environmental transmission from parents to their offspring, \(\:t²\) reflects the environmental component due to twin-specific influences, \(\:s²\) represents the environmental component attributable to sibling-specific influences, \(\:e²\) is the individual-specific environmental component, and \(\:2aw\) is the component due to passive rGE (Hufer et al. 2020; Keller et al. 2009). These variance components are summarized in Fig. 4.

Table 3

Nuclear Twin Family Model (NTFM) Fit Statistics and Indices Based on Different RWA and SDO Measures
NTFMs	Fit statistics/indices
and measures	AIC	BIC	χ²	df	p	CFI	RMSEA
RWA z-standardized mean score
sibling effects	13638.12	13680.30	36.94	32	.251	.994	.015
epistasis	13637.37	13679.55	36.19	32	.279	.995	.013
allelic dominance	13636.51	13678.69	35.33	32	.314	.996	.012
RWA orthogonal factor score
sibling effects	13791.90	13834.08	34.07	32	.368	.997	.009
epistasis	13791.44	13833.62	33.62	32	.389	.998	.008
allelic dominance	13790.95	13833.13	33.12	32	.412	.998	.007
RWA latent variable score
sibling effects	59126.00	59605.78	778.48	369	< .001	.876	.039
epistasis	59123.73	59603.52	776.21	369	< .001	.877	.039
allelic dominance	59126.00	59605.78	778.48	369	< .001	.876	.039
SDO z-standardized mean score
sibling effects	14156.60	14198.77	32.50	32	.442	.988	.005
epistasis	14156.61	14198.79	32.51	32	.442	.988	.005
allelic dominance	14156.61	14198.79	32.51	32	.442	.988	.005
SDO orthogonal factor score
sibling effects	14273.13	14315.31	28.77	32	.631	1.000	.000
epistasis	14273.13	14315.31	28.77	32	.631	1.000	.000
allelic dominance	14273.13	14315.31	28.77	32	.631	1.000	.000
SDO latent variable score
sibling effects	59856.08	60335.87	668.74	369	< .001	.835	.034
epistasis	59856.08	60335.87	668.74	369	< .001	.835	.034
allelic dominance	59856.08	60335.87	668.74	369	< .001	.835	.034
NTFM with sibling effects: s ≠ 0 and n = 0; NTFM with epistasis: n ≠ 0 (monozygotic twin correlation: 1, dizygotic twin/nontwin sibling correlation: 0) and s = 0; NTFM with allelic dominance: n ≠ 0 (monozygotic twin correlation: 1, dizygotic twin/nontwin sibling correlation: 0.25) and s = 0; AIC: Akaike Information Criterion; BIC: Bayesian Information Criterion; df: degrees of freedom; CFI: Comparative Fit Index; RMSEA: Root Mean Square Error of Approximation

The analyses yielded consistent differences across the three measures regarding the sources of variance in RWA and SDO. While passive rGE, environmental parental transmission, and nonadditive genetic effects played a significant role for variance in RWA, none of these factors significantly contributed to the variance in SDO. Even after correction for item-specific and random error variance in the latent variable models, individual-specific environmental influences contributed substantially to the variance in SDO (36%), as compared to a small but significant contribution to individual differences in RWA (13%). There were also some similarities across the two constructs: The entire (i.e., additive and nonadditive) genetic variance and the variance attributable to twin-specific shared environmental influences were comparable in size for RWA and SDO.

Multi-Cohort Nuclear Twin Family Modeling

In a final step, we tested for differences between twins’ birth cohorts regarding genetic and environmental variance components of RWA and SDO. We used a MC-NTFM allowing for nonadditive genetic effects in case of RWA and a reduced MC-NTFM for SDO, in which sibling-specific shared environmental influences and nonadditive genetic factors were assumed to be negligible, as these models turned out to be the best fitting models based on the full sample (see Fig. 4). The cohort differences in standardized variance components are summarized in Fig. 5.

In line with the expectation, the plots in Fig. 5 suggest declining contributions of passive rGE from > 10% in adolescence to 0% in young adulthood and increasing contributions of nonadditive genetic factors from 0% in adolescence to 20% in young adulthood to the variance in RWA. Also, increasing contributions of additive genetic factors from 9–17% in adolescence to 21–49% in young adulthood to the variance in SDO were implied. However, model comparisons indicated no significant differences based on the likelihood ratio test across cohorts for SDO (see Table 4). For RWA, by contrast, the cohort differences in variance components turned out to be statistically significant for the mean scores and latent variable scores, albeit not for the factor scores.

Table 4

Multi-Cohort Nuclear Twin Family Model (MC-NTFM) Fit Statistics and Model Comparisons Based on Different RWA and SDO Measures
MC-NTFMs	Fit statistics/indices
and measures	AIC	RMSEA	χ²	df	p	Δχ²	Δdf	p
RWA z-standardized mean score
different	13616.78	.034	122.08	96	.037
equal	13623.36	.042	156.66	110	.002	34.58	14	.002
RWA orthogonal factor score
different	13786.88	.026	111.23	96	.137
equal	13781.22	.030	133.56	110	.063	22.33	14	.072
RWA latent variable score
different	59116.70	.056	1963.78	1113	< .001
equal	59127.23	.057	2002.31	1127	< .001	38.53	14	< .001
SDO z-standardized mean score
different	14149.43	.033	124.87	99	.040
equal	14144.68	.035	144.12	111	.019	19.25	12	.083
SDO orthogonal factor score
different	14273.69	.028	117.75	99	.096
equal	14265.48	.029	133.53	111	.072	15.79	12	.201
SDO latent variable score
different	59814.48	.046	1671.91	1116	< .001
equal	59809.36	.046	1690.78	1128	< .001	18.87	12	.092
MC-NTFMs with different and equal model parameters (a, n, m, f, t, e, and µ for RWA and a, m, f, t, e, and µ for SDO) across cohorts of twins; AIC: Akaike Information Criterion; RMSEA: Root Mean Square Error of Approximation; df: degrees of freedom; Δχ²: likelihood ratio

Not in line with the hypotheses, however, the plots in Fig. 5 indicate that environmental factors shared by twins only (\(\:t²\)) and by all siblings within a family attributable to parental transmission (\(\:x\)) represent the primary sources that became more important for RWA during the transition from adolescence (0% for mean scores up to 2% for latent variable scores) to young adulthood (22% for mean scores up to 36% for latent variable scores), accounting for the increasing variance in RWA across cohorts. Comparable trends, primarily due to environmental factors shared by twins, appeared for SDO from adolescence (5% for mean scores up to 8% for latent variable scores) to emerging adulthood (19% for mean scores up to 26% for latent variable scores).

The current study examined whether the importance of different genetic and environmental sources of individual differences in RWA and SDO varies between these two constructs and across three age cohorts. For this purpose, we used extended twin family data encompassing families with three different twin birth cohorts. Our twin family model analyses yielded new insights into both similarities and differences regarding the sources of variance in RWA compared to SDO, and how these variance components vary across different age cohorts.

Different Sources for RWA and SDO?

Our study indicates there were no substantive differences between RWA and SDO concerning the entirety of the genetic variance, which includes both additive and nonadditive genetic factors. This finding is consistent with some previous studies (e.g., Kleppestø et al. 2024), but contradicts other findings based on a different German twin family sample that included older twins which suggested a smaller genetic component for SDO compared to RWA (Kandler 2015) or even a negligible genetic component after exclusion of the common variance with RWA (Kandler et al. 2016). However, differences regarding genetic influences could be observed between RWA and SDO when considering the additive and nonadditive genetic variance components separately. Both additive and nonadditive genetic sources of variance were statistically significant for RWA, while only additive genetic sources were relevant for individual differences in SDO.

Compared to models based on data from twins only, employing an NTFM provides more precise estimates of the genetic components when there are also relevant environmental influences shared by twins (Instinske and Kandler 2024). In our study, we found that twin-specific shared environmental factors played a crucial role for both RWA and SDO. This suggests a potential impact of age-related effects among twin siblings, or possibly the effects of same-aged twin siblings on each other created by a twin-specific coevolution that produces socio-political preferences and attitudes that are more strongly shared by twins than by other siblings of different ages. Consistent with the latter interpretation is the observation that components due to twin-specific shared environmental effects were lower when based on the orthogonal factor scores compared to the mean or latent variable scores of both RWA and SDO. This indicates that twin-specific shared environmental effects might account for a considerable amount of the common variance in RWA and SDO, and thus yield a common impact on both constructs.

Our findings also indicate that environmental transmission from generation to generation was statistically significant for RWA, but not for SDO. This meant that passive rGE was only relevant for RWA as well. This finding is consistent with a pervious German twin family study using data from an older sample of twins and their family members (Kandler et al. 2016). About 16% of the variance in RWA in the previous study and 18% in the current study could be attributed to passive rGE, after correction for measurement error variance. Generally, RWA exhibited a higher familial resemblance, which can be attributed to larger environmental influences shared among family members. In contrast, SDO appears to be substantially more influenced by individual-specific environmental effects, even after adjustments for measurement error variance. Accordingly, individual differences in RWA may be more malleable by within-familial sources, while extra-familial circumstances may represent the primary environmental determinants of variance in SDO.

Different Sources for Different Ages?

Generally, our study implied that individual differences in RWA and SDO tend to increase as individuals age. In the light of developmental theories (Scarr 1992; Scarr and McCartney 1983), these trends could reflect the increasing importance of individual experiences and environments as well as the individualizing interdependence between genetic and environmental factors: As people grow older and mature, such factors may play a more decisive role in shaping their socio-political attitudes. Initially, we hypothesized that passive rGE declines (Hypothesis 1), while genetic differences increase with age potentially reflecting active rGE (Hypothesis 2) for both RWA and SDO. Our study indicated that these preregistered hypotheses could not be consistently confirmed. The results revealed a decline in passive rGE for RWA only, which is consistent with Hypotheses 1a but not 1b, because passive genotype-environment covariance components were generally negligible for SDO. Furthermore, overall levels of genetic variance remained fairly constant for RWA across cohorts, while these levels generally tended to increase for SDO, which provided only partial confirmation of Hypothesis 2. Although the latter hypothesis predicted increasing levels of genetic variance with age without distinguishing between additive and nonadditive genetic variance, an increase in the nonadditive variety and a decrease in additive genetic variance was found in RWA, while an increasing trend was observed in the additive genetic variance component of SDO.

Interestingly, we noted an increase in the twin-specific shared environmental component for RWA across the three cohorts, and for SDO, an increasing trend was observed from adolescence to emerging adulthood. Although not expected, these findings seem to be in accord with the idea that individuals increasingly distance themselves from the parental home and interact more autonomously with their self-chosen environments. In other words, as twins transition from adolescence to adulthood, they might talk more on political, particularly RWA-related, issues with each other and with their peers. Since interactions with peers are thought to play an important role in the shaping of political attitudes (e.g., Dey 1997), the increase in the twin-specific shared environmental component could be reflective of an increased political engagement with peers with increasing age. Peer or co-twin influences, which act to increase twin similarity compared to other family members, may reflect the presence of important social role models, which could have a significant impact on socio-political attitudes during the formative years of political development. These interpretations align with the conclusions drawn by Kandler et al. (2012) that environmental influences within but not across generations provide relevant sources of political orientations. Given the apparently important role of twin-specific shared environmental influences, understanding the specific underlying processes could, prospectively, provide insights into the development and evolution of socio-political attitudes.

As an alternative interpretation, twin-specific shared environmental effects could potentially mimic maturity-related genetic differences affecting both RWA and SDO. However, the fact that non-twin siblings of a twin pair were on average of a comparable age and that sibling-specific influences were negligible contradicts this interpretation. Generally, the influences of within-familial environmental factors on RWA appear to increase with age. Conversely, these influences are small for SDO and tend to decline in young adults. These findings again emphasize the differing dynamics of environmental influences on the development of these two socio-political traits over time.

Limitations and Future Directions

Despite the advantages of our study, including the application of a twin family design to disentangle various genetic and environmental variance components, the use of data from different age cohorts from adolescence to adulthood, a substantial sample size, and the differentiation between mean, orthogonal factor, and latent variable scores of RWA and SDO, some limitations also exist. One of the limitations of our study is the inability to disentangle between cohort and age effects. In other words, we could not clearly differentiate whether differences between the three different age groups were due to genuinely chronological age-related changes (age effects), or if they were related to the specific time period in which the different cohorts were born and raised (cohort effects). This limits the implications of our study for an understanding of the development of RWA and SDO. To address this limitation, subsequent studies could utilize data from multiple timepoints in addition to data from multiple birth cohorts to apply longitudinal twin family model analyses.

Another limitation in our study was the lack of consideration for sex effects. Sex, as a biological variable, can significantly influence the manifestation of both RWA and SDO due to variations in hormone levels, societal influences, and other sex-specific factors. Future studies should therefore seek to incorporate opposite-sex DZ twins for sex limitation modeling (Neale et al. 2006) to provide a more exhaustive and nuanced understanding of the genetic and environmental influences on RWA and SDO.

The different ages of family members in an NTFM also represents a limitation in our study. These age differences can lead to distortions, as it is assumed that the genetic and environmental variance is the same across the different generations. However, such an assumption may not be tenable, especially since socio-political preferences and attitudes may be influenced by different social contexts experienced by different generations (Zubielevitch et al. 2023). Although data collection of family members when they are in the same age or developmental phase may be difficult to realize, this would enable a more comprehensive understanding of the transmission of RWA and SDO from one generation to the next.

Finally, our measures of RWA and SDO were rather short ones, as is typically the problem in large panel studies. Nevertheless, future studies should use multi-item and multi-faceted measures of RWA and SDO to further improve internal reliability and to examine these constructs at the facet level.

The manifestations of both RWA and SDO appear to be to some extent similar within families, although spouse similarity and family resemblance across all core family members appear to be consistently higher for RWA than for SDO. Our nuclear twin family model analyses suggest that the higher family resemblance in RWA is attributable to significant environmental transmission from parents to offspring and passive genotype-environment correlation. For SDO, however, these two influences turned out to be negligible, such that individual-specific environmental circumstances seemed to primarily account for differences in SDO. Although additive and nonadditive genetic sources appear to drive individual differences in RWA and only additive genetic sources seem to have an impact on individual differences in SDO, the overall level of genetic variance did not differ between RWA and SDO.

Individual differences and twin resemblance in RWA and SDO tended to be smaller for adolescents compared to emerging adult and young adult twins. However, in contrast to our expectation, especially for RWA, the larger variance in adulthood was not attributable to larger genetic differences as a consequence of an increasing importance of active and evocative genotype-environment correlation. Rather, the larger variance with age might be the result of an increasing importance of twin-specific shared environments. These findings point to the relevance of the social environment common to twins but not to other siblings during the formative years of socio-political development that takes place as adolescents mature into adulthood. These environmental influences shared by twins could include having common friends or mentors, or even the influence same-aged twins have on each other.

This paper is a reworked and finalized version of a paper previously entitled “Genetic and Environmental Differences in Right-Wing Authoritarianism and Social Dominance Orientation from Adolescence to Adulthood: A Multi-Cohort Twin Family Study”, which was presented as an oral presentation at the 54th Behavior Genetics Association Annual Meeting in London, UK, June 26-29, 2024 (Abstract in: Behavior Genetics, 54, pp. 534-535).

Author Contribution

C.K. wrote the main manuscript with the help of E.B., who provided conceptual ideas.C.K. and J.I. conducted the analyses.All authors reviewed the manuscript and provided helpful feedback.

Acknowledgments.

The present research is based on data from the TwinLife Study that was funded by the German Research Foundation (DFG, Grant Number 220286500, https://gepris.dfg.de/gepris/projekt/220286500).

Data Availability

The entire data set is available as Scientific Use File at https://doi.org/10.4232/1.14331. In addition, reduced anonymized input data files are provided at OSF. All data analytic scripts and codes for the performed model analyses used within this research paper are openly available at OSF: https://osf.io/mq83d/.

Akaike H (1987) Factor analysis and AIC. Psychometrika 52(3):317–332. https://doi.org/10.1007/BF02294359
Altemeyer B (1981) Right-wing authoritarianism. Winnipeg University of Manitoba Press
Bentler PM (1990) Comparative fit indexes in structural models. Psychol Bull 107(2):238–246. https://doi.org/10.1037/0033-2909.107.2.238
Bentler PM (2006) EQS 6 structural equations program manual. Multivariate Software, Inc.
Briley DA, Livengood J, Derringer J, Tucker-Drob EM, Fraley RC, Roberts BW (2019). Interpreting behavior genetic models: Seven developmental processes to understand. Behav Genet 49(2):196–210. https://doi.org/10.1007/s10519-018-9939-6
Cheung GW, Rensvold RB (2002) Evaluating goodness-of-fit indexes for testing measurement invariance. Structural Equation Modeling: a Multidisciplinary Journal 9:233–255. https://doi.org/10.1207/S15328007SEM0902_5
Cronbach LJ (1951) Coefficient alpha and the internal structure of tests. Psychometrika 16(3):297–334. https://doi.org/10.1007/bf02310555
De Vries RE, Wesseldijk LW, Karinen AK, Jern P, Tybur JM (2021) Relations between HEXACO personality and ideology variables are mostly genetic in nature. Eur J Pers 36(2):200–217. https://doi.org/10.1177/08902070211014035
Dey EL (1997) Undergraduate Political Attitudes: Peer Influence in Changing Social Contexts. The Journal of Higher Education 68:398. https://doi.org/10.2307/2960009
Diewald M, Kandler C, Riemann R et al. (2024). TwinLife. GESIS, Cologne. ZA6701 Data file Version 8.0.0, https://doi.org/10.4232/1.14331
DiStefano C, Zhu M, Mindrila D (2009) Understanding and using factor scores: Considerations for the applied Researcher. Practical Assessment, Research & Evaluation 14(20):20. https://doi.org/10.7275/da8t-4g52
Duckitt J, Sibley CG (2010) Personality, Ideology, Prejudice, and Politics: A Dual‐Process Motivational Model. J Pers 78(6):1861–1894. https://doi.org/10.1111/j.1467-6494.2010.00672.x
Eaves L, Martin N, Heath A, Schieken R, Meyer J, Silberg J, Neale M, Corey L (1997) Age changes in the causes of individual differences in conservatism. Behav Genet 27(2):121–124. https://doi.org/10.1023/a:1025633307992
Funk CL, Smith KB, Alford JR, Hibbing MV, Eaton NR, Krueger RF, Eaves LJ, Hibbing JR (2013) Genetic and environmental transmission of political orientations. Political Psychology 34(6):805–819. https://doi.org/10.1111/j.1467-9221.2012.00915.x
Funke F (2005) The Dimensionality of Right‐Wing Authoritarianism: Lessons from the Dilemma between Theory and Measurement. Political Psychology 26(2):195–218. https://doi.org/10.1111/j.1467-9221.2005.00415.x
Hahn E, Gottschling J, Bleidorn W, Kandler C, Spengler M, Kornadt AE, Schulz W, Schunck R, Baier T, Krell K, Lang V, Lenau F, Peters A-L, Diewald M, Riemann R, Spinath FM (2016) What drives the development of social inequality over the life course? the German TwinLife study. Twin Res Hum Genet 19(6):659–672. https://doi.org/10.1017/thg.2016.76
Hatemi PK, Funk CL, Medland SE, Maes HM, Silberg JL, Martin NG, Eaves LJ (2009) Genetic and environmental transmission of political attitudes over a life time. The Journal of Politics 71(3):1141–1156. https://doi.org/10.1017/s0022381609090938
Hatemi PK, Medland SE, Klemmensen R, Oskarsson S, Littvay L, Dawes CT, Verhulst B, McDermott R, Nørgaard AS, Klofstad CA, Christensen K, Johannesson M, Magnusson PKE, Eaves LJ, Martin NG (2014) Genetic Influences on Political Ideologies: Twin Analyses of 19 Measures of Political Ideologies from Five Democracies and Genome-Wide Findings from Three Populations. Behav Genet 44(3):282–294. https://doi.org/10.1007/s10519-014-9648-8
Heath AC, Eaves LJ (1985) Resolving the effects of phenotype and social background on mate selection. Behav Genet 15(1):15–30. https://doi.org/10.1007/bf01071929
Heath AC, Kendler KS, Eaves LJ, Markell D (1985) The resolution of cultural and biological inheritance: Informativeness of different relationships. Behav Genet 15(5):439–465. https://doi.org/10.1007/bf01066238
Hebler M, Booh AT, Wieczorek S, Schneider JF (2014) Right-Wing Autoritarismus. Zusammenstellung sozialwissenschaftlicher Items und Skalen (ZIS). https://doi.org/10.6102/zis81
Horwitz TB, Balbona JV, Paulich KN, Keller MC (2023) Evidence of correlations between human partners based on systematic reviews and meta-analyses of 22 traits and UK Biobank analysis of 133 traits. Nat Hum Behav 7:1568–1583. https://doi.org/10.1038/s41562-023-01672-z
Hu L, Bentler PM (1999) Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling a Multidisciplinary Journal 6(1):1–55. https://doi.org/10.1080/10705519909540118
Hufer A, Kornadt AE, Kandler C, Riemann R (2020) Genetic and environmental variation in political orientation in adolescence and early adulthood: A Nuclear Twin Family analysis. J Pers Soc Psychol 118(4):762–776. https://doi.org/10.1037/pspp0000258
Instinske J, Kandler C (2024) Benefits of nuclear twin family models for estimating heritability: An investigation of 93 personality-relevant constructs using JASP. Personality Science 5. https://doi.org/10.1177/27000710241299222
JASP Team (2024) JASP (Version 0.19.2)[Computer software]. https://jasp-stats.org/
Jennings MK, Stoker L, Bowers J (2009) Politics across Generations: Family Transmission Reexamined. The Journal of Politics 71(3):782–799. https://doi.org/10.1017/s0022381609090719
Kämpfe N (2002) Persönlichkeit, soziale Einstellungen und Fremdenfeindlichkeit: Eine empirische Untersuchung mit Selbst- und Bekannteneinschätzungen [Personality, social attitudes, and xenophobia: A study using self- and peer report] (Unpublished diploma thesis). Friedrich-Schiller-University Jena, Germany.
Kandler C (2015) Quellen politischer Orientierung: Genetische, soziale, kulturelle und Persönlichkeitsfaktoren [Sources of political orientation: genetic, social, cultural, and personality factors]. In: Nomos eBooks. pp 44–69. https://doi.org/10.5771/9783845254418-44
Kandler C, Bell E, Riemann R (2016) The structure and sources of right–wing authoritarianism and social dominance orientation. Eur J Pers 30(4):406–420. https://doi.org/10.1002/per.2061
Kandler C, Bell E, Shikishima C, Yamagata S, Riemann R (2015) Genetic foundations of attitude formation. In: Scott RA, Kosslyn SM (eds) Emerging Trends in the Social and Behavioral Sciences. Wiley, pp 1–21. https://doi.org/10.1002/9781118900772.etrds0144
Kandler C, Bleidorn W, Riemann R (2012) Left or right? Sources of political orientation: the roles of genetic factors, cultural transmission, assortative mating, and personality. J Pers Soc Psychol 102:633–645. https://doi.org/10.1037/a0025560
Kandler C, Lewis GJ, Feldhaus LH, Riemann R (2014) The Genetic and Environmental Roots of Variance in Negativity toward Foreign Nationals. Behav Genet 45(2):181–199. https://doi.org/10.1007/s10519-014-9700-8
Kandler C, Waaktaar T, Mõttus R, Riemann R, Torgersen S (2019) Unravelling the Interplay between Genetic and Environmental Contributions in the Unfolding of Personality Differences from Early Adolescence to Young Adulthood. Eur J Pers 33(3):221–244. https://doi.org/10.1002/per.2189
Kandler C, Zapko-Willmes A, Richter J, Riemann R (2021) Synergistic and dynamic genotype-environment interplays in the development of personality differences. In: Rauthmann J (ed) The Handbook of Personality Dynamics and Processes. Elsevier, pp 155–181. https://doi.org/10.1016/B978-0-12-813995-0.00007-8
Keller MC, Coventry WL (2005) Quantifying and addressing parameter indeterminacy in the classical twin design. Twin Res Hum Genet 8(3):201–213. https://doi.org/10.1375/twin.8.3.201
Keller MC, Medland SE, Duncan LE, Hatemi PK, Neale MC, Maes HHM, Eaves LJ (2009) Modeling Extended Twin Family Data I: Description of the cascade model. Twin Res Hum Genet 12(1):8–18. https://doi.org/10.1375/twin.12.1.8
Keller MC, Medland SE, Duncan LE (2009) Are extended twin family designs worth the trouble? A comparison of the bias, precision, and accuracy of parameters estimated in four twin family models. Behav Genet 40(3):377–393. https://doi.org/10.1007/s10519-009-9320-x
Kleppestø TH, Czajkowski NO, Sheehy‐Skeffington J, Vassend O, Roysamb E, Eftedal NH, Kunst JR, Ystrom E, Thomsen L (2024) The genetic underpinnings of right‐wing authoritarianism and social dominance orientation explain political attitudes beyond Big Five personality. J Pers 92(6):1744–1758. https://doi.org/10.1111/jopy.12921
Lenau, F., Hahn, E., Peters, A.-L., Gottschling, J., Thiel, W., & Spinath, F. M. (2017). Zygosity determination in twin studies. (TwinLife Working Paper Series, 01). Project TwinLife “Genetic and social causes of life chances”. https://www.twin-life.de/images/TwinLife/Series/twinlife_wp_01.pdf
Lewis GJ, Bates TC (2013) Common heritable effects underpin concerns over norm maintenance and In‐Group favoritism: evidence from genetic analyses of Right‐Wing authoritarianism and traditionalism. J Pers 82(4):297–309. https://doi.org/10.1111/jopy.12055
Little RJA, Rubin DB (2019) Statistical analysis with missing data (3rd rev. edition). Wiley.
Ludeke SG, Krueger RF (2013) Authoritarianism as a personality trait: Evidence from a longitudinal behavior genetic study. Pers Individ Dif 55(5):480–484. https://doi.org/10.1016/j.paid.2013.04.015
Lykken DT (2006) The mechanism of emergenesis. Genes Brain Behav 5(4):306–310. https://doi.org/10.1111/j.1601-183x.2006.00233.x
McCourt K, Bouchard TJ, Lykken DT, Tellegen A, Keyes M (1999) Authoritarianism revisited: genetic and environmental influences examined in twins reared apart and together. Pers Individ Dif 27(5):985–1014. https://doi.org/10.1016/s0191-8869(99)00048-3
McDonald RP (1999) Test theory: a unified treatment. Lawrence Erlbaum Associates Publishers.
Nacke L, Riemann R (2023) Two sides of the same coin? On the common etiology of Right-Wing Authoritarianism and Social Dominance Orientation. Pers Individ Dif 207:112160. https://doi.org/10.1016/j.paid.2023.112160
Neale MC, Røysamb E, Jacobson K (2006) Multivariate genetic analysis of sex limitation and G x E interaction. Twin Res Hum Genet 9:481–489. https://doi.org/10.1375/183242706778024937
Plomin R, DeFries JC, Loehlin JC (1977) Genotype-environment interaction and correlation in the analysis of human behavior. Psychol Bull 84(2):309–322. https://doi.org/10.1037/0033-2909.84.2.309
Pratto F, Sidanius J, Stallworth LM, Malle BF (1994) Social dominance orientation: A personality variable predicting social and political attitudes. J Pers Soc Psychol 67(4):741–763. https://doi.org/10.1037/0022-3514.67.4.741
Putnick DL, Bornstein MH (2016) Measurement invariance conventions and reporting: The state of the art and future directions for psychological research. Dev Rev 41:71–90. https://doi.org/10.1016/j.dr.2016.06.004
Rohm T, Andreas A, Deppe M, Eichhorn H, Instinske J, Klatzka CH, Kottwitz A, Krell K, Mönkediek B, Paulus L, Piesch S, Ruks M, Starr A, Weigel L, Diewald M, Kandler C, Riemann R, Spinath FM (2023) Data from the German TwinLife Study: Genetic and Social Origins of Educational Predictors, Processes, and Outcomes. JOPD 11(1):4. https://doi.org/10.5334/jopd.78
Scarr S (1992) Developmental Theories for the 1990s: Development and Individual Differences. Child Development 63(1):1–19. https://doi.org/10.1111/j.1467-8624.1992.tb03591.x
Scarr S (1993) Biological and Cultural Diversity: the legacy of Darwin for Development. Child Development 64(5):1333. https://doi.org/10.2307/1131538
Scarr S, McCartney K (1983) How People Make Their Own Environments: A Theory of Genotype --> Environment Effects. Child Development 54(2):424. https://doi.org/10.2307/1129703
Schwarz G (1978) Estimating the dimension of a model. Ann Statist 6(2). https://doi.org/10.1214/aos/1176344136
Steiger JH (2006) Understanding the limitations of global fit assessment in structural equation modeling. Pers Individ Dif 42(5):893–898. https://doi.org/10.1016/j.paid.2006.09.017
Steiger JH, Lind JC (1980). Statistically-based tests for the number of common factors. Annual meeting of the Psychometric Society, Iowa City, IA, 1980.
Zubielevitch E, Osborne D, Milojev P, Sibley CG (2023) Social dominance orientation and right-wing authoritarianism across the adult lifespan: An examination of aging and cohort effects. J Pers Soc Psychol 124(3):544–566. https://doi.org/10.1037/pspi0000400

No competing interests reported.

Download PDF

Journal Publication

published 03 Nov, 2025

Read the published version in Behavior Genetics →

Editorial decision: Revision requested
14 Jul, 2025
Reviews received at journal
08 Jul, 2025
Reviews received at journal
08 Jul, 2025
Reviews received at journal
18 Jun, 2025
Reviewers agreed at journal
12 Jun, 2025
Reviewers agreed at journal
09 Jun, 2025
Reviewers agreed at journal
04 Jun, 2025
Reviews received at journal
10 Apr, 2025
Reviewers agreed at journal
03 Apr, 2025
Reviewers invited by journal
28 Mar, 2025
Editor assigned by journal
20 Mar, 2025
Submission checks completed at journal
18 Mar, 2025
First submitted to journal
03 Mar, 2025

You are reading this latest preprint version

Sources of Differences in Right-Wing Authoritarianism and Social Dominance Orientation from Adolescence to Adulthood: A Multi-Cohort Twin Family Study

Status:

Journal Publication

Version 1

Abstract

Figures

Introduction

Different Sources for RWA and SDO?

Different Sources for Different Ages?

The Current Study

Method

Study Design and Data Collection Procedures

Sample

RWA and SDO Measures

Analyses

Family Correlation Analyses

Nuclear Twin Family Model Analyses

Multi-Cohort Nuclear Twin Family Model Analyses

Results

Family Correlations

Nuclear Twin Family Modeling

Multi-Cohort Nuclear Twin Family Modeling

Discussion

Different Sources for RWA and SDO?

Different Sources for Different Ages?

Limitations and Future Directions

Conclusion

Declarations

Author Contribution

Acknowledgments.

Data Availability

References

Additional Declarations

Status:

Journal Publication

Version 1