Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2021 Jul 12.
Published in final edited form as: Dev Psychol. 2020 May 7;56(7):1360–1371. doi: 10.1037/dev0000965

Parenting, Temperament, and Attachment Security as Antecedents of Political Orientation: Longitudinal Evidence from Early Childhood to Age 26

Christopher M Wegemer 1, Deborah Lowe Vandell 1
PMCID: PMC8274710  NIHMSID: NIHMS1716709  PMID: 32378919

Abstract

This paper examines early childhood antecedents of adults’ political orientation. Using longitudinal data from the National Institute of Child Health and Human Development Study of Early Child Care and Youth Development, we investigate associations between parenting beliefs and behaviors, child temperament, and attachment security during early childhood in relation to adult political ideology and party affiliation at age 26 years (N = 1,364). Young children’s fearful temperament and anxious attachment security, as well as mothers’ authoritarian parenting beliefs in early childhood, predicted conservative political orientations at age 26. Children’s abilities to focus attention and avoidant attachment security predicted liberal orientations. These findings provide evidence that multiple aspects of early developmental experience – temperament, parenting, and infant-mother attachment – are associated with later political orientations.

Keywords: political development, parenting, temperament, attachment, conservativism, political psychology

Historically, studies have found that developmental experiences prior to adulthood have enduring effects on political orientation across the life span (Sears & Brown, 2013). Questions regarding the nature and persistence of childhood influences on political orientation have resurfaced (Astuto & Ruck, 2010; Holbein, 2017; Patterson et al., 2019; Torney-Purta, 2017). Recent work in political psychology has conceptualized ideology as motivated social cognition, reframing individual political differences in terms of personality, needs, attitudes, and beliefs (Jost, Glaser, Kruglanski, & Sulloway, 2003; Jost, 2017). Clarifying the developmental mechanisms that underlie differences in political orientation remains an important task for developmental psychology. The present study advances research by linking parenting, temperament, and attachment in early childhood to three aspects of political orientation in adulthood: ideological conservatism, Republican party affiliation, and Democratic Party affiliation.

In a series of seminal studies, Adorno and colleagues (1950) used retrospective questionnaires and interviews to explore the relationship between authoritarianism and parenting (particularly in early childhood), concluding that conservatives were more likely than liberals to report that their parents used harsh and punitive parenting practices. A large body of literature subsequently investigated the relationship between parenting practices and beliefs and the political orientation of their offspring (Altemeyer, 1981; Duckitt, 2001; Rokeach, 1960). Conceptual frameworks regarding the enduring effects of parenting have typically relied on personality theory to understand differences in the development of ideologies (Adorno et al., 1950; Altemeyer, 1981; Duckitt, 2001; Pratto, Sidanius, Stallworth, & Malle, 1994; Rokeach, 1960). Across cultures and contexts, liberals consistently score higher on openness to experience and conservatives score higher on conscientiousness (Carney, Jost, Gosling, & Potter, 2008; Mondak, 2010).

Contrary to existing theories (e.g., Altemeyer, 1981), recent studies suggest that individual differences in political ideologies related to authoritarianism begin to emerge in childhood (Reifen Tagar, Federico, Lyons, Ludeke, & Koenig, 2014). Researchers have increasingly advocated for the importance of childhood as a period of political development (Astuto & Ruck, 2010; Patterson et al., 2019; Torney-Purta, 2017). For example, Holbein (2017) found that increases in psychosocial skills in childhood (such as emotional regulation and cooperation with others) were related to greater likelihood of political participation in adulthood. The present study applies a lens of political psychology to established developmental constructs, contributing to a narrative of political development across the lifespan.

To date, very few studies have used longitudinal data to examine links between early childhood temperament, parenting, and adult political orientation. Block and Block (2006) followed a small sample of nursery school children to age 23 (N = 95); children who were fearful, anxious, and rigid were more likely to be conservative in adulthood. Using the NICHD Study of Early Child Care and Youth Development (SECCYD), Fraley, Griffin, Belsky, and Roisman (2012) investigated relations between early childhood temperament and political orientation at age 18. Similar to Block and Block, child fearfulness was related to later conservativism, whereas higher activity levels and attentional focusing in early childhood were related to liberalism at age 18. In addition, Fraley et al. (2012), detected links between parenting styles in early childhood and later political orientations. Children whose mothers endorsed authoritarian parenting beliefs were more likely to be conservative at age 18. The present research extends the previous longitudinal research studies by investigating early childhood attachment security; we test attachment, temperament, and parenting style simultaneously to build evidence that may begin to disentangle mechanisms that underlie the development of ideology.

Comprehensive meta-analyses by Jost and colleagues (2003; 2017) have found that conservativism is characterized by resistance to change related to an underlying preoccupation with fear of uncertainty and threats to security. To the degree that political ideology is motivated by management of underlying security, attachment theory may be implicated. Bowlby (1973) has argued that individuals’ responses to threat are influenced by their early interactions with caregivers, which shape internal working models that persist across the life course and generalize across contexts. These models contain internalized representations of the world that guide interactions aimed at maintaining security (Bretherton & Munholland, 2008). In the current study, children’s attachment security was assessed at an age of 3 years using a modified version of Ainsworth’s classic “Strange Situation” (Cassidy, Marvin, & MacArthur Working Group, 1992). We examined attachment at age 3 because internal working models are better consolidated than in infancy (Bowlby, 1973; Sherman, Rice, & Cassidy, 2015) although analyses using attachment measured at 15 months are included in the appendix of this paper.

Distinct patterns of security management behaviors have been classified into four attachment categories: secure, insecure-anxious, insecure-avoidant, and disorganized (Ainsworth, Blehar, Waters, & Wall, 1978; Main & Solomon, 1990). Securely attached children explore the environment when the mother is present, become distressed when the mother leaves, and are easily comforted when she returns (Ainsworth et al., 1978; Cassidy et al., 1992). In contrast, anxiously attached children become deeply upset when separated from their mother and are unable to derive comfort from maternal interactions upon reunion. Avoidantly attached children appear unaffected by the departure of their mother and remain disengaged after she returns. Children who exhibit disorganized attachment do not have a coherent pattern of interacting with their mother (Main & Solomon, 1990).

Anxious attachment is aligned with Jost et al.’s (2003) definitional features of conservativism: specifically, intense fear of losing security, low tolerance of uncertainty, and reduced exploration. The political congruence of other attachment types is less clear. A small handful of studies have linked attachment types to political ideology, but all relied on retrospective or cross-sectional data and produced conflicting findings (Koleva & Rip, 2009). The present study is the first to investigate longitudinal associations between attachment in early childhood and adult political orientation. Thus far, scholars have established a link between parenting and political outcomes without invoking attachment theory, despite a strong relationship between parenting and attachment (De Wolff & van Ijzendoorn, 1997).

We hypothesize that children’s attachment type and their temperament, as well as mothers’ beliefs and practices in early childhood, will be related to adult political orientation at age 26. We expect that fearful temperament and authoritarian parenting beliefs will be associated with adult political conservatism, consistent with findings at age 18 reported by Fraley et al. (2012) using a subsample of participants who partially overlapped with the present study. Results congruent with Fraley et al.’s findings at age 18 will provide evidence for the enduring effects of early childhood experiences (as opposed to fade out). In addition, based on other cross-sectional and retrospective studies (Koleva & Rip, 2009), we expect that anxious attachment will be related to adult conservativism. Finally, we anticipate that childhood attachment, temperament, and parenting beliefs and practices will be independently related to adult political orientation through separate predictive pathways. The present research will be the first to provide evidence that may disaggregate psychological mechanisms associated with antecedents of adult political orientation.

Method

Participants

The National Institute of Child Health and Human Development Study of Early Child Care and Youth Development (NICHD SECCYD) is a longitudinal study that followed a cohort of children between 1991 and 2018. The investigators approached 1,364 mothers in the hospital shortly after the birth of the study children in 10 locations across the United States: Little Rock, AR; Irvine, CA; Lawrence, KS; Boston, MA; Philadelphia, PA; Pittsburgh, PA; Charlottesville, VA; Seattle, WA; Morganton, NC; and Madison, WI. For a detailed description of recruitment and sample statistics, see the NICHD Early Child Care Research Network (2001); more information is available at http://www.icpsr.umich.edu/icpsrweb/ICPSR/series/00233.

The children’s development and life experiences were the focus of five different waves of investigation through the end of high school. The children, now in adulthood, were participants in the present study. The study was approved by the Institutional Review Board at the University of California, Irvine (HS2017–3847: “Are there meaningful and enduring effects of out-of-school time? Two proposed studies”). A survey was administered to 814 of the original 1,364 study participants at age 26. Participants were primarily White and middle class; sample statistics at recruitment and age 26 are described in Table 1. The sample of the present study consists of all 1,364 participants; we utilized multiple imputation to account for missing data, as described below. (Disciplines vary in their approach to selective attrition and imputation of dependent variables. To account for differing methodological perspectives, we conducted parallel analyses using a sample with data imputed for only the 814 participants who completed the age 26 survey. Results were nearly identical, as shown in tables A1 and A2.)

Table 1.

Descriptive statistics of the NICHD SECCYD sample, at recruitment and at age 26

Recruitment (at 1 month) Age 26 t p
Gender
 Female 48.3% 52.6% −1.93 .05
Race/ethnicity
 White 75.0% 81.0% −2.49 .01
 Black 12.8% 8.6% 2.93 .00
 Hispanic 6.6% 5.4% 0.66 .51
 Other 5.6% 5.0% −0.21 .84
Socioeconomic status
 Avg. maternal education 14.2 14.7 −3.97 < .01
 Avg. income-to-needs ratio 3.4 3.7 −2.76 .01
Geographic location of birth
 Little Rock, AR 11.0% 5.9% 4.02 < .01
 Irvine, CA 9.7% 11.8% −1.56 .12
 Lawrence, KS 9.8% 9.7% 0.03 .97
 Boston, MA 10.3% 10.6% −0.22 .82
 Pittsburgh, PA 9.0% 11.7% −2.00 .05
 Philadelphia, PA 10.0% 9.1% 0.50 .67
 Charlottesville, VA 10.0% 9.3% 0.48 .63
 Seattle, WA 10.2% 11.1% −0.64 .52
 Morganton, NC 10.6% 10.0% 0.45 .65
 Madison, WI 9.6% 10.9% −1.00 .32
Study predictors
 Authoritarian parenting beliefs 60.3 57.9 3.69 < .01
 Observed maternal sensitivity 0.0 0.1 −3.23 < .01
 Temperament, Activity level 4.8 4.7 1.15 .25
 Temperament, Attentional focusing 4.7 4.8 −1.36 .17
 Temperament, Fear 4.1 4.1 0.49 .63
 Temperament, Shyness 3.5 3.5 −0.04 .97
 Attachment, Secure 61.5% 62.1% −0.28 .78
 Attachment, Anxious 17.3% 16.7% 0.34 .74
 Attachment, Avoidant 4.8% 5.0% −0.21 .84
 Attachment, Disorganized 16.4% 16.2% 0.14 .89
N 1,364 814
Open in a new tab

Note. Percentages are displayed for categorical indicators and means are displayed for continuous variables. The total N’s are all possible participants at each wave, although data on all indicators was not available for all participants (see Table 2 for item-specific N’s). Two-sample t-tests were used to identify significant differences between the waves.

Measures

Four constructs were assessed in early childhood: maternal parenting beliefs, maternal behavior, child temperament, and attachment security. Political ideology and party affiliation were measured in adulthood. A rich array of variables that served as covariates were collected, as described below. Descriptive statistics and correlations are presented in Table 2.

Table 2.

Correlations among study variables

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
1. Ideological conservativism 1
2. Republican party affiliation .60* 1
3. Democratic party affiliation −.48* −.31* 1
4. Secure attachment −.04 −.02 .03 1
5. Anxious attachment .13* .10* −.02 −.58* 1
6. Avoidant attachment −.11* −.08* −.04 −.29* −.10* 1
7. Disorganized attachment −.03 −.03 .00 −.56* −.20* −.10* 1
8. Authoritarian parenting beliefs .19* .09* −.19* −.10* .05 −.01 .08* 1
9. Observed maternal sensitivity −.15* −.03 .12* .20* −.06* −.03 −.19* −.55* 1
10. Temperament, activity level .06 .04 −.07 −.05 .03 −.03 .05 .11* −.16* 1
11. Temperament, attentional focusing −.16* −.13* .11* .12* −.08* −.03 −.07* −.30* .33* −.42* 1
12. Temperament, fear .08 .03 −.02 .00 −.01 −.04 .04 .11* .01 .01 −.06 1
13. Temperament, shyness −.05 −.05 .07 −.02 .04 .01 −.02 .02 .02 −.21* .00 .23* 1
14. Female −.08* −.06 .14* −.06 .01 −.05 .09* −.01 .08* −.18* .13* .05 .07* 1
15. White .07 .13* −.08* .05 .02 .00 −.09* −.33* .39* .01 .16* −.07* −.04 −.01 1
16. Black −.05 −.13* .09* −.09* −.02 .04 .12* .33* −.39* .05 −.17* .06 −.02 .00 −.69* 1
17. Hispanic −.06 −.02 .02 .03 −.01 −.04 −.01 .11* −.06* −.05 −.03 .04 .05 .00 −.46* −.10* 1
18. Other race/ethnicity .00 −.05 .01 .00 .00 −.01 .01 .02 −.09* −.04 −.02 .00 .04 .01 −.41* −.09* −.06* 1
19. Maternal education −.14* −.04 .21* .13* −.06* −.06* −.07* −.55* .52* −.15* .28* .02 .03 .04 .23* −.20* −.12* .00 1
20. Income-to-needs ratio −.05 .02 .14* .07* −.04 −.05 −.02 −.42* .42* −.12* .23* .01 .02 .04 .24* −.27* −.09* .03 .55* 1
Mean 2.43 .24 .57 .62 .17 .05 .16 60.34 −0.01 4.79 4.71 4.09 3.52 .48 .76 .13 .06 .05 14.23 3.39
SD 1.16 .54 .81 .49 .38 .21 .37 15.21 0.74 .76 .85 .85 1.10 .50 .43 .33 .24 .22 2.51 2.69
Min 0 0 0 0 0 0 0 0 −3.07 1 1 1 1 0 0 0 0 0 7 0.13
Max 5 2 2 1 1 1 1 110 1.48 7 7 7 7 1 1 1 1 1 21 23.79
N 652 811 811 1,140 1,140 1,140 1,140 1,360 1,210 1,025 1,023 818 1,054 1,364 1,364 1,364 1,364 1,364 1,363 1,355
21. Little Rock, AR .07 .03 −.08* .05 −.05 .04 −.04 .16* −.16* .06* −.01 .01 .02 −.01 −.04 .16* −.08* −.07* −.11* −.14*
22. Irvine, CA .08* .09* −.05 −.01 .02 .02 −.02 −.06* .13* −.05 .03 −.02 −.04 −.01 −.04 −.12* .24* .00 .05 .19*
23. Lawrence, KS .03 .04 −.03 .04 .04 −.05 −.06 −.04 .06* −.02 .00 .02 .01 .00 −.02 −.04 .06* .03 .03 −.08*
24. Boston, MA −.02 −.08* .02 −.01 .03 −.04 .00 −.03 .02 −.04 −.01 .00 .02 −.02 .06* −.09* −.03 .05 −.01 .06*
25. Pittsburgh, PA −.02 .00 −.02 −.05 .05 .00 .02 −.01 .01 −.02 .00 .01 −.03 −.01 −.01 .09* −.07* −.04 .05 .04
26. Philadelphia, PA −.13* −.08* .10* −.05 −.01 .01 .06* −.03 −.03 −.06 .01 .04 −.02 .02 −.07* .13* −.04 −.02 .10* .11*
27. Charlottesville, VA .01 .01 .00 −.02 −.03 −.02 .07* .04 −.03 .06* −.07* −.04 −.03 .00 .01 .02 −.03 −.02 −.06* −.07*
28. Seattle, WA −.13* −.11* .13* .01 .00 −.01 −.01 −.14* .09* −.02 .12* .00 .05 .01 −.03 −.03 .03 .08* .06* .07*
29. Morganton, NC .23* .18* −.13* .01 −.01 −.01 .01 .12* −.10* .09* −.10* .01 −.03 .00 .09* −.05 −.06* −.03 −.12* −.09*
30. Madison, WI −.11* −.09* .03 .02 −.03 .05 −.03 −.03 .03 −.02 .02 −.03 .07* .03 .05 −.06* −.01 .02 .01 −.07*
Mean 2.43 .24 .57 .62 .17 .05 .16 60.34 −.01 4.79 4.71 4.09 3.52 .48 .76 .13 .06 .05 14.23 3.39
SD 1.16 .54 .81 .49 .38 .21 .37 15.21 .74 .76 .85 .85 1.10 .50 .43 .33 .24 .22 2.51 2.69
Min 0 0 0 0 0 0 0 0 −3.07 1 1 1 1 0 0 0 0 0 7 0.13
Max 5 2 2 1 1 1 1 110 1.48 7 7 7 7 1 1 1 1 1 21 23.79
N 652 811 811 1,140 1,140 1,140 1,140 1,360 1,210 1,025 1,023 818 1,054 1,364 1,364 1,364 1,364 1,364 1,363 1,355
Open in a new tab

Note. In the creation of the ideological conservativism measure, 159 participants selected “don’t know, haven’t decided” and were included as missing, reducing the sample size to 652. For dichotomous variables, the mean represents the proportion of participants. Bivariate correlations between sites were omitted because they are not conceptually meaningful. For all geographic sites, N = 1,364.

*

p < .05

Authoritarian parenting beliefs.

The Parental Modernity Inventory (Schaefer & Edgerton, 1985) was completed by mothers when the study child was one month old. Thirty items assessed attitudes regarding parenting practices using a 5-point Likert scale. The inventory consisted of two subscales: traditional parenting attitudes (22 items, α = .90) and progressive parenting attitudes (8 items, α = .60) that were negatively correlated (r = –.38). Congruent with previous research, the subscale of traditional parenting attitudes was used to represent authoritarian parenting beliefs (Fraley et al., 2012). The inventory captured authoritarian parenting beliefs by asking mothers the degree to which they agreed with statements such as “children should not question the authority of their parents” and “children will be bad unless they are taught what is right.” Consistent with the original scale design, response scores were summed, then standardized.

Observed maternal sensitivity.

Maternal parenting behavior was observed when children were 6, 15, 24, 36, and 54 months during semi-structured tasks in the laboratory. Children participated in activities that required the assistance of their mother to complete. At 6, 15, and 24 months, three subscales measured responsiveness to non-distress, intrusiveness (reverse scored), and positive regard. At 36 and 54 months, subscales of respect for autonomy, hostility (reverse scored), and supportive presence were used. Across all time points, inter-rater reliability ranged from .83 to .88. A single measure representing observed maternal sensitivity was created for each wave by combining the respective subscales, yielding good internal consistency (α > .80 for all time points). Measures were standardized and averaged across all waves to create a composite indicator of maternal sensitivity (α = .77).

Child temperament.

When children were 54 months of age, mothers completed 80 items from an abbreviated version of the Children’s Behavior Questionnaire (CBQ; Rothbart, Ahadi, Hershey, & Fisher, 2001). Each item contained a statement that described a child’s behavior; mothers were asked how strongly the specific behavior resembled that of their own child on a seven-point Likert scale from “extremely untrue” to “extremely true.” The inventory consisted of 8 scales. Fraley et al. (2012) tested the association between four categories of temperament and conservativism at age 18. We used four temperament scales that paralleled the categories used by Fraley et al. (2012): shyness (8 items, α = .87, e.g., “sometimes seems nervous when talking to adults s/he has just met”), activity level (10 items, α = .68, e.g., “tends to run rather than walk from a room”), fear (10 items, α = .60, e.g., “is afraid of the dark”), attentional focusing (8 items, α = .74, e.g., “is good at following directions”). All scales were standardized.

Attachment security.

When children were 36 months of age, a 25-minute attachment assessment was videotaped in the laboratory at the 10 research sites. The MacArthur measure, a modified version of the Strange Situation, was employed (Cassidy et al., 1992). All videotapes were scored by three trained coders at a single site; inter-coder agreement (before conferencing) was 77% (k = .50, p < .001). Four types of attachment were classified: secure, insecure-anxious, insecure-avoidant, or disorganized. Of the original 1,364 participants recruited for the study, attachment security was assessed for 1,140 children at 36 months. Overall, 62% of children (n = 701) in the sample were classified as secure, 17% (n = 197) were insecure-anxious, 5% (n = 55) were insecure-avoidant, and 16% (n = 187) were disorganized.

Adult political orientation.

Political ideology and party affiliation at age 26 years were captured with survey questions drawn from the Monitoring the Future Study (Schulenberg et al., 2018). Other nationally representative longitudinal studies, such as the American Freshman survey and the General Social Survey, used similar survey questions (Twenge, Honeycutt, Prislin, & Sherman, 2016). Similar to the analytic strategies employed in these studies, we constructed three variables, one representing political ideology and two representing party affiliation.

First, participants were asked to identify which of seven political categories best described their political ideology. The first five categories presented a progression from “Very conservative” to “Very liberal” with “Moderate” at the midpoint. A sixth category, “Radical,” was given as a response option after “Very Liberal.” There were 159 participants who selected a seventh category, “Don’t know, haven’t decided”; when creating our measure of political ideology, their values were marked as missing. The scale was reverse-coded so that higher values indicated greater conservativism. This measure of conservativism was standardized. A similar measure has been traditionally used to represent political ideology (Hibbing, Smith, & Alford, 2014; Jost, Federico, & Napier, 2009), which was also employed in one of the few longitudinal studies of political antecedents (Block & Block, 2006).

Next, participants were asked to report which of eight choices best described the political party they identified with. The first four categories presented a progression from “Strongly Republican” to “Strongly Democrat,” with “Mildly Republican” and “Mildly Democrat” in between. Other response options were “Independent,” “Other,” “No preference,” and “Don’t know, haven’t decided.” A continuous measure was created for Republican Party affiliation, with a value of 2 representing “Strongly Republican,” a value of 1 representing “Mildly Republican,” and a value of 0 assigned to any other category choice. A similar approach was used to create a score for Democratic Party affiliation: a value of 2 represented “Strongly Democratic,” 1 represented “Mildly Democratic,” and 0 assigned to any other category choice. In total, 152 participants identified as “mildly” or “strongly” Republican, whereas 294 identified as “mildly” or “strongly” Democrat. Both measures of party affiliation were standardized. This item was originally used in the longitudinal Youth in Transition Project and has since been widely adapted (Bachman, 1974).

Covariates.

A number of variables were included as covariates in the analyses because they could potentially account for associations between parenting practices and political outcomes. Maternal education was measured as the number of years of education reported by mothers at 1 month. Values ranged from 7 years to 21 years with an average of 14.2. Family income-to-needs ratio was computed based on financial information provided by mothers at 1, 6, 15, 24, 36 and 54 months. U.S. Census Bureau poverty data was used in conjunction with reported financial information to calculate ratios, which were averaged across the data points (α = .94). Both maternal education and income-to-needs ratios were standardized. Participant’s geographic location of birth was also controlled.

Consistent with political psychology literature, gender was treated as a binary construct. A dummy variable was used to indicate whether or not each participant was female. Lastly, a categorical variable was created to indicate whether the participant was identified as White, Black, Hispanic, or other. Each child’s race/ethnicity was reported by their mother at 1 month.

Missing data

As is common for longitudinal studies, sample attrition occurred. The results of t-tests comparing the participants at 1 month and the most recent survey 26 years later are displayed in Table 1. Participants were more likely to continue in the study if they were female and White with higher maternal education level and income-to-needs ratio. Differences emerged on study predictors. Participants whose mothers expressed authoritarian parenting beliefs were less likely to continue in the study to age 26, whereas those whose mothers displayed sensitivity were more likely to persist.

We used multiple imputation to account for missing data for all participants in the original recruitment sample (N = 1,364). See Table 2 for the un-imputed sample size for each variable. The imputation model included all study variables as well as additional auxiliary variables from the dataset that were either theoretically implicated or at least moderately correlated with study variables. Following established practices, 30 analysis datasets were created using chained equations (MICE; see White, Royston, & Wood, 2011). This approach allowed separate conditional distributions for each imputed variable, which was suitable for our dataset because several variables were not normally distributed. The study analyses involved pooling parameter estimates from statistical tests on each of the imputed datasets to yield a single set of results.

Analytic plan

First, we investigated zero-order correlations between all study variables. Next, we used OLS regressions to examine the relations between early childhood predictors and three political orientation outcomes at age 26: ideological conservativism, Republican Party affiliation, and Democratic Party affiliation. Separate models were estimated for each type of antecedent: maternal beliefs and behaviors, child temperament, and attachment (Tables 3–5). All predictors were then included simultaneously to examine the unique associations of each construct (Table 6).

Table 3.

OLS regression results, maternal beliefs and behavior as predictors of political orientation outcomes at age 26

Ideological conservativism
 Republican Party affiliation
 Democratic Party affiliation
b SE (b) β t p b SE (b) β t p b SE (b) β t p
Authoritarian parenting beliefs 0.01 0.00 0.18 3.61 < .01 0.00 0.00 0.13 2.55 .01 −0.01 0.00 −0.14 −3.22 < .01
Observed maternal sensitivity −0.14 0.08 −0.12 −1.70 .09 −0.03 0.03 −0.06 −0.94 .35 0.05 0.05 0.06 0.93 .35
Open in a new tab

Note. N = 1,364. Standard errors are shown for unstandardized coefficients. Unstandardized coefficients were calculated using unstandardized predictors and outcomes, except observed maternal sensitivity was kept as a standardized predictor because the composite is inherently unitless. Controls included gender, race/ethnicity, maternal education, income-to-needs ratio, and geographic location. The adjusted R2 for the ideological conservativism model was .10, F(17, 847) = 4.79, p < .001. The adjusted R2 for the Republican Party affiliation model was .07, F(17, 961) = 4.08, p < .001. The adjusted R2 for the Democratic Party affiliation model was .12, F(17, 944) = 6.97, p < .001.

Table 5.

OLS regression results, child attachment as a predictor of political orientation outcomes at age 26

Ideological conservativism
 Republican Party affiliation
 Democratic Party affiliation
b SE (b) β t p b SE (b) β t p b SE (b) β t p
Attachment
 Anxious 0.28 0.12 0.24 2.23 .03 0.12 0.05 0.23 2.57 .01 −0.03 0.07 −0.03 −0.36 .72
 Avoidant −0.61 0.22 −0.53 −2.71 < .01 −0.15 0.09 −0.28 −1.60 .11 −0.13 0.14 −0.16 −0.95 .34
 Disorganized 0.01 0.13 0.01 0.10 .92 0.02 0.06 0.05 0.43 .67 −0.09 0.09 −0.11 −0.96 .34
Open in a new tab

Note. N = 1,364. Standard errors are shown for unstandardized coefficients. Unstandardized coefficients were calculated using unstandardized predictors and outcomes, except observed maternal sensitivity was kept as a standardized predictor because the composite is inherently unitless. Controls included gender, race/ethnicity, maternal education, income-to-needs ratio, and geographic location. The adjusted R2 for the ideological conservativism model was .09, F(18, 840) = 4.13, p < .001. The adjusted R2 for the Republican Party affiliation model was .07, F(18, 957) = 3.74, p < .001. The adjusted R2 for the Democratic Party affiliation model was .11, F(18, 936) = 5.68, p < .001.

Table 6.

OLS regression results, all early child characteristics as predictors of political orientation outcomes at age 26

Ideological conservativism
 Republican Party affiliation
 Democratic Party affiliation
b SE (b) β t p b SE (b) β t p b SE (b) β t p
Authoritarian parenting beliefs 0.01 0.00 0.14 2.97 < .01 0.00 0.00 0.11 2.11 .04 −0.01 0.00 −0.13 −3.05 < .01
Observed maternal sensitivity −0.12 0.08 −0.10 −1.44 .16 −0.01 0.03 −0.02 −0.33 .74 0.03 0.05 0.04 0.66 .51
Temperament
 Activity level −0.11 0.07 −0.07 −1.55 .13 −0.03 0.03 −0.05 −1.03 .31 0.04 0.04 0.03 0.82 .41
 Attentional focusing −0.15 0.07 −0.11 −2.24 .03 −0.08 0.03 −0.13 −2.76 < .01 0.05 0.04 0.05 1.19 .24
 Fear 0.12 0.05 0.09 2.32 .02 0.02 0.03 0.03 0.71 .48 −0.05 0.04 −0.05 −1.21 .23
 Shyness −0.06 0.04 −0.06 −1.63 .11 −0.03 0.02 −0.06 −1.71 .09 0.05 0.03 0.06 1.70 .09
Attachment
 Anxious 0.24 0.10 0.21 2.00 .05 0.10 0.05 0.20 2.27 .02 −0.01 0.07 −0.02 −0.19 .85
 Avoidant −0.59 0.20 −0.51 −2.58 .01 −0.15 0.09 −0.27 −1.55 .13 −0.16 0.14 −0.19 −1.11 .27
 Disorganized −0.05 0.12 −0.04 −0.33 .74 0.01 0.06 0.02 0.18 .86 −0.07 0.09 −0.08 −0.73 .47
Open in a new tab

Note. N = 1,364. Standard errors are shown for unstandardized coefficients and t-statistics are shown for standardized coefficients. Unstandardized coefficients were calculated using unstandardized predictors and outcomes, except observed maternal sensitivity was kept as a standardized predictor because the composite is inherently unitless. Controls included gender, race/ethnicity, maternal education, income-to-needs ratio, and geographic location. The adjusted R2 for the ideological conservativism model was .14, F(24, 893) = 4.47, p < .001. The adjusted R2 for the Republican Party affiliation model was .09, F(24, 995) = 3.60, p < .001. The adjusted R2 for the Democratic Party affiliation model was .13, F(24, 997) = 5.02, p < .001.

Results

As shown in Table 2, the child and family covariates were correlated with the early childhood predictors and adult political orientation outcomes. Generally, study participants who were ideologically conservative at age 26 were less likely to be female (r = −.08, p = .032) or raised by mothers with higher education levels (r = −.14, p < .001). Geographic location of birth was typically associated with political orientation in ways that were consistent with voting trends over the last several decades. For instance, participants born in rural North Carolina were more likely to have a conservative ideology (r = .23, p < .001), whereas participants from Seattle were less likely (r = −.13, p < .001).

Using OLS regression models, we first investigated the associations between maternal authoritarian parenting beliefs, observed maternal sensitivity, and political orientation in adulthood (see Table 3), controlling for child gender, race/ethnicity, years of maternal education, income-to-needs ratio, and site. Mothers who held authoritarian parenting beliefs were more likely to have children who were ideologically conservative at age 26 (β = 0.18, p < .001) and affiliated with the Republican Party (β = 0.13, p = .012). Their children were also less likely to be associated with Democratic Party at age 26 (β = −0.14, p = .002).

Next, we examined the relationship between child temperament and adult political orientation using regressions (see Table 4). Holding all covariates constant, a more fearful temperament at 54 months was linked to endorsing ideological conservativism at age 26 (β = 0.10, p = .013). Higher attentional focusing in early childhood was linked to less ideological conservativism (β = −0.14, p = .006) and weaker Republican Party affiliation (β = −0.14, p = .002).

Table 4.

OLS regression results, child temperament characteristics as predictors of political orientation outcomes at age 26

Ideological conservativism
 Republican Party affiliation
 Democratic Party affiliation
b SE (b) β t p b SE (b) β t p b SE (b) β t p
Temperament
 Activity level −0.10 0.07 −0.06 −1.42 .16 −0.03 0.03 −0.04 −0.96 .34 0.04 0.04 0.04 0.89 .37
 Attentional focusing −0.18 0.07 −0.14 −2.82 < .01 −0.09 0.03 −0.14 −3.21 < .01 0.07 0.04 0.07 1.74 .09
 Fear 0.13 0.05 0.10 2.55 .01 0.02 0.03 0.04 0.90 .37 −0.05 0.04 −0.06 −1.45 .15
 Shyness −0.06 0.04 −0.05 −1.52 .13 −0.03 0.02 −0.06 −1.56 .12 0.04 0.03 0.06 1.64 .10
Open in a new tab

Note. N = 1,364. Standard errors are shown for unstandardized coefficients. Unstandardized coefficients were calculated using unstandardized predictors and outcomes, except observed maternal sensitivity was kept as a standardized predictor because the composite is inherently unitless. Controls included gender, race/ethnicity, maternal education, income-to-needs ratio, and geographic location. The adjusted R2 for the ideological conservativism model was .09, F(19, 866) = 3.97, p < .001. The adjusted R2 for the Republican Party affiliation model was .08, F(19, 949) = 3.74, p < .001. The adjusted R2 for the Democratic Party affiliation model was .11, F(19, 947) = 5.68, p < .001.

We then investigated the relationship between child attachment type and adult political orientation, controlling for all covariates (see Table 5). Relative to secure attachment, anxious attachment at 36 months predicted higher ideological conservativism at age 26 (β = 0.24, p =.028) and stronger Republican Party affiliation (β = 0.23, p = .011). In contrast, children who were avoidantly attached (versus securely attached) were less ideologically conservative at age 26 (β = -−0.53, p = .009). Childhood attachment security was not associated with adult Democratic Party affiliation.

Next, we tested all of the childhood predictors (maternal beliefs and behavior, child temperament, and attachment security) simultaneously (see Table 6). All of the aforementioned patterns of associations were unchanged from the previous regression models: adult ideological conservativism was predicted by authoritarian parenting beliefs (β = 0.14, p = .004), fearful temperament (β = 0.09, p = .023), and anxious attachment (β = 0.21, p = .049), whereas avoidant attachment (β = –0.51, p = .012) and attentional focusing (β = −0.11, p = .029) predicted away from ideological conservativism. Republican Party affiliation was predicted by authoritarian parenting beliefs (β = .11, p = .037) and anxious attachment (β = 0.20, p = .024), whereas attentional focusing (β = −0.13, p = .007) predicted away from Republican Party affiliation. Authoritarian parenting beliefs predicted away from Democratic Party affiliation (β = −0.13, p = .003). Using Wald tests, we found that an anxious versus secure attachment predicted ideological conservativism and Republican Party affiliation more strongly than any other study variables, and furthermore, regression coefficients for attachment were statistically as large as any control variables.

Lastly, we conducted two sets of ancillary analyses, presented in the appendix (as previously described). First, we tested regression models paralleling those in Table 6 using a sample constrained to participants who completed the survey at age 26 (N = 814). Each of the study variables for parenting beliefs, temperament, and attachment predicted political outcomes with directions and magnitudes consistent with our primary findings (see Table A2). Second, we conducted analyses using attachment measured at 15 months instead of attachment measured at 3 years (see Table A3). Anxious attachment at 15 months was related to ideological conservativism at age 26 (β = 0.34, p = .014), consistent with the association of anxious attachment at 3 years.

Discussion

This study investigated early childhood antecedents of adult political orientation. Using longitudinal NICHD SECCYD data, we found that mothers’ beliefs and parenting behaviors, children’s temperament, and attachment security during early childhood were related to political ideology and party affiliation at age 26. The results were consistent with existing longitudinal studies of political differences that focused on parenting beliefs and early temperament (Block & Block, 2006; Fraley et al., 2012). In addition, the current study advances the literature by examining attachment security, clarifying potential mechanisms of political development, and demonstrating enduring effects of multiple antecedents well into adulthood.

Congruent with the longitudinal work of Block and Block (2006) and Fraley and colleagues (2012), we found that authoritarian parenting beliefs, temperamental fearfulness, and deficits in attentional control in early childhood predicted adult conservativism. The design of the latter study (Fraley et al., 2012) was similar to our present work and assessed a subset of our current sample at age 18. The predictive strength of authoritarian parenting beliefs, fearfulness, and attentional focusing was comparable between age 18 and age 26, which provides evidence that these patterns are enduring at least into early adulthood. The present study, as well as the work of Fraley et al. (2012), extends the earlier work of Block and Block (2006) to multiple locations across the country, encompassing a diversity of political environments. Similar to adults, child political attitudes differ by geographic location (Patterson et al., 2019); future work will probe the relationship between childhood antecedents of adult political outcomes and geographic location, especially in light of Jost et al.’s (2009) characterization of “elective affinities.”

The present study also is the first to establish longitudinal associations between attachment security in early childhood and adult political orientation. We found that anxious attachment was strongly related to conservative outcomes in adulthood, whereas avoidant attachment strongly predicted away from conservativism. Cross-sectional studies relying primarily on surveys of college students have previously found that anxious attachment in adulthood was related to concurrent conservativism (Koleva & Rip, 2009).

Jost and colleagues (2003) found that fear of uncertainty and threats to security differentiate conservativism from liberalism. Reviewing prospective research on political antecedents (Block & Block, 2006; Fraley et al., 2012), Hibbing et al. (2014) argued that dispositional differences are rooted in negativity bias. Our study found that fearful temperament and anxious attachment were both linked to ideological conservativism. The effect of each predictor was not attenuated when both were included in the same regression models and anxious attachment was not correlated with fearful temperament. These results suggest that fearful dispositions and attachment security are two separate mechanisms that are differentially linked to individual differences in political orientation rather than general negativity bias (consistent with the arguments of Lilienfeld & Latzman, 2014). Future research may clarify the mechanisms and explore potential interactions between fearful temperament and attachment security. Our work adds nuance to prior research by implicating attachment security as a potential origin of differences in threat sensitivity associated with conservativism.

The results regarding attachment support Bowlby’s (1973) early argument that security management strategies developed in early childhood may have enduring effects in other domains later in the lifespan. Patterns of proximity and contact behaviors in early childhood that define attachment are conceptualized as manifestations of internal working models, which explains the persistence of attachment security and its generalizability across contexts (Bretherton & Munholland, 2008). This study found such patterns are related to adult political orientation. Attachment security-seeking strategies in childhood include affect displays, exploration, and searching for the attachment figure, each of which differ depending on attachment type. As a child develops, these security-seeking strategies are transposed across different attachment figures and contexts (Mikulincer & Shaver, 2003), which as the present research suggests, can include the political domain. For instance, avoidant children’s apparent indifference towards caregivers may manifest in adulthood as disregard towards authority or tradition as sources of security. Despite the central role of the quality of maternal interactions in shaping attachment security (Ainsworth et al., 1978), we found that maternal sensitivity during early childhood did not predict adult political outcomes, even though early childhood attachment did. The results suggest that attachment is not simply a mediating process linking maternal behavior and their children’s political orientations. Instead, our findings support the hypothesis that early childhood attachment security shapes enduring psychosocial models that relate to political outcomes through security management.

Some scholars have previously used attachment theory as a framework for understanding individual differences in terror management strategies associated with political ideology (Huddy, Feldman, & Weber, 2007), although such research did not employ longitudinal data. In a series of studies, Mikulincer and Florian (2000), for example, found that attachment types in adulthood were differentially related to mechanisms of terror management; anxiously attached individuals exhibited greater threat sensitivity and defensiveness, whereas avoidantly attached individuals were more likely to employ suppression. Experimental manipulations found that priming with an existential threat caused an increase in conservativism for insecurely (anxiously or avoidantly) attached adults, whereas priming with security increased liberal disposition in all adults regardless of attachment type (Gillath & Hart, 2010; Weise et al., 2008). Our study is aligned with this literature and is the first to examine security management patterns by providing a direct link between childhood attachment types and adult political orientation.

We found that childhood antecedents were differentially related to three aspects of political orientation: ideological conservativism, Republican Party affiliation, and Democratic Party affiliation. This is consistent with nationally representative studies that have found differences in the degree of congruence between political ideology and party affiliation (Twenge et al., 2016). Fearful temperament was associated with ideological conservativism, but not Republican Party affiliation. Similarly, avoidant attachment predicted away from ideological conservativism, but was not negatively related to Republican Party affiliation. Political ideology may be more conceptually proximal to psychological constructs than party affiliation. Interestingly, neither early childhood attachment nor temperament were predictive of Democratic Party affiliation, suggesting that fear and security management are only relevant to conservative orientation.

This research was limited by the absence of data on parents’ politics so we could not control for parent political orientation or assess the role of child characteristics in intergenerational political differences. According to Jennings & Niemi’s (1968) widely invoked family transmission model, socialization processes explain offspring’s tendency to adopt their parent’s political orientations. Parenting beliefs and behaviors have been found to moderate the transmission of political orientation from mothers to their children (Murray & Mulvaney, 2012) and transmission has been found to be dependent on children’s attitudes and behaviors (Ojeda & Hatemi, 2015), as well as genetics (Hatemi et al., 2009). Further examining the processes that underlie longitudinal associations between parent political orientation, parenting beliefs and behavior, child characteristics, and children’s later political orientation will be an important area for future work.

Despite its limitations, the present study advances understanding of the origins of individual political differences by identifying potential developmental mechanisms. The dataset employed is unique in its capability to establish such relationships because of its extensive investigation of early childhood and its temporal span of more than two and a half decades. The findings contribute to a growing body of literature that locates political development prior to adulthood and implicates well-established developmental mechanisms in childhood. Through this work, we promote the importance of developmental perspectives in understanding the origins of political ideology in relation to child psychology.

Acknowledgments

A cooperative agreement (5 U10 HD027040) between the study investigators that included Deborah Lowe Vandell and the Eunice Kennedy Shriver National Institute of Child Health and Human Development supported the design and data collection of the Study of Early Child Care and Youth Development (SECCYD) from birth through age 15 years. The age 26 study design, data collection, and analyses were supported by a grant from the Charles Stewart Mott Foundation (G-2017-00786) to Deborah Lowe Vandell. We are grateful to SECCYD site coordinators (Susan Dell, Wendy Wagner Robeson, Carol Rangel, and Janet Gouge Johns) and the research team at UC Irvine (Andrea Karsh, Stephanie Soto-Lara, Khamia Powell, Olaitan Jimoh, Lea Ibalio, and Michael Mayfield) for their tireless efforts to locate study participants and to coordinate data collection for the age 26 follow-up. Finally, we thank the study participants for their willingness to be a part of this longitudinal follow-up of a project that began shortly after their birth.

Appendix A. Ancillary analyses.

Table A1.

Correlations among study variables, using sample constrained to age 26 survey responses

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
1. Ideological conservativism 1
2. Republican party affiliation .60* 1
3. Democratic party affiliation −.48* −.31* 1
4. Secure attachment −.04 −.02 .03 1
5. Anxious attachment .13* .10* −.02 −.57* 1
6. Avoidant attachment −.11* −.08* −.04 −.30* −.10* 1
7. Disorganized attachment −.03 −.03 .00 −.56* −.20* −.10* 1
8. Authoritarian parenting beliefs .19* .09* −.19* −.06 .07 .00 .01 1
9. Observed maternal sensitivity −.15* −.03 .12* .19* −.08* −.04 −.14* −.48* 1
10. Temperament, activity level .06 .04 −.07 −.03 .02 −.04 .04 .11* −.17* 1
11. Temperament, attentional focusing −.16* −.13* .11* .12* −.07 −.07 −.06 −.26* .32* −.44* 1
12. Temperament, fear .08 .03 −.02 .02 −.02 −.04 .02 .12* .05 .02 −.06 1
13. Temperament, shyness −.05 −.05 .07 −.04 .05 .02 −.01 .00 .04 −.20* −.02 .23* 1
14. Female −.08* −.06 .14* −.07* .02 −.04 .10* .00 .09* −.20* .11* .05 .06 1
15. White .07 .13* −.08* .05 .03 −.03 −.08* −.28* .35* .02 .13* −.07 −.03 .02 1
16. Black −.05 −.13* .09* −.09* −.03 .07 .11* .29* −.30* .03 −.13* .10* .00 −.02 −.63* 1
17. Hispanic −.06 −.02 .02 .03 −.01 −.03 −.02 .10* −.09* −.06 −.04 −.02 .03 −.03 −.49* −.07* 1
18. Other race/ethnicity .00 −.05 .01 −.01 −.02 .00 .03 .03 −.15* −.01 −.02 .02 .02 .02 −.48* −.07* −.06 1
19. Maternal education −.14* −.04 .21* .11* −.06 −.09* −.03 −.52* .48* −.14* .25* .05 .03 .04 .20* −.16* −.11* −.05 1
20. Income-to-needs ratio −.05 .02 .14* .07 −.05 −.09* .01 −.37* .39* −.13* .21* .05 .04 .05 .21* −.21* −.08* −.02 .54* 1
Mean 2.43 .24 .57 .62 .17 .05 .16 57.90 .10 4.74 4.76 4.07 3.52 .53 .81 .09 .05 .05 14.67 3.72
SD 1.16 .56 .81 .49 .37 .22 .37 14.36 .68 .76 .85 .84 1.11 .50 .39 .28 .23 .22 2.42 2.63
Min 0 0 0 0 0 0 0 0 −2.77 1 1 1 1 0 0 0 0 0 7 .17
Max 5 2 2 1 1 1 1 110 1.37 7 7 7 7 1 1 1 1 1 21 23.79
N 652 811 811 755 755 755 755 814 789 742 735 587 757 814 814 814 814 814 814 813
Open in a new tab

Note. All variables are presented unstandardized. In the creation of the ideological conservativism measure, 159 participants selected “don’t know, haven’t decided” and were included as missing, reducing the sample size to 652.

*

p < .05

Table A2.

OLS regression results, all early child characteristics as predictors of political orientation outcomes at age 26, using sample constrained to age 26 survey responses

Ideological conservativism
 Republican Party affiliation
 Democratic Party affiliation
b SE (b) β t p b SE (b) β t p b SE (b) β t p
Authoritarian parenting beliefs 0.01 0.00 0.1 1.94 .05 0.00 0.00 0.09 1.82 .07 −0.01 0.00 −0.12 −2.56 .01
Observed maternal sensitivity −0.13 0.09 −0.11 −1.45 .15 −0.01 0.03 −0.02 −0.30 .76 0.02 0.05 0.02 0.35 .73
Temperament
 Activity level −0.10 0.07 −0.06 −1.39 .16 −0.04 0.03 −0.05 −1.19 .24 0.03 0.04 0.03 0.65 .52
 Attentional focusing −0.14 0.06 −0.1 −2.19 .03 −0.08 0.03 −0.12 −2.72 < .01 0.04 0.04 0.04 0.97 .33
 Fear 0.12 0.06 0.09 2.00 .05 0.01 0.03 0.02 0.40 .69 −0.04 0.04 −0.04 −0.90 .37
 Shyness −0.05 0.04 −0.05 −1.29 .20 −0.02 0.02 −0.04 −1.14 .26 0.04 0.03 0.05 1.47 .14
Attachment
 Anxious 0.34 0.12 0.3 2.83 < .01 0.12 0.05 0.22 2.30 .02 −0.01 0.08 −0.02 −0.17 .87
 Avoidant −0.56 0.23 −0.48 −2.44 .02 −0.18 0.08 −0.34 −2.18 .03 −0.11 0.13 −0.13 −0.85 .40
 Disorganized −0.06 0.13 −0.06 −0.50 .62 −0.01 0.05 −0.02 −0.23 .82 −0.08 0.08 −0.09 −0.94 .35
Open in a new tab

Note. N = 655 for the ideological conservativism model and N = 814 for the two party affiliation models, constrained by survey responses. Standard errors are shown for unstandardized coefficients and t-statistics are shown for standardized coefficients. Unstandardized coefficients were calculated using unstandardized predictors and outcomes, except observed maternal sensitivity was kept as a standardized predictor because the composite is inherently unitless. Secure is the reference category for attachment coefficients. Controls included gender, race/ethnicity, maternal education, income-to-needs ratio, and geographic location. The adjusted R2 for the ideological conservativism model was .16, F(24, 626) = 5.64, p < .001. The adjusted R2 for the Republican Party affiliation model was .10, F(24, 783) = 4.55, p < .001. The adjusted R2 for the Democratic Party affiliation model was .11, F(24, 783) = 4.97, p < .001.

Table A3.

OLS regression results, all early child characteristics as predictors of political orientation outcomes at age 26, using attachment at 15 months

Ideological conservativism
 Republican Party affiliation
 Democratic Party affiliation
b SE (b) β t p b SE (b) β t p b SE (b) β t p
Authoritarian parenting beliefs 0.01 0.00 0.15 3.22 < .01 0.00 0.00 0.11 2.25 .03 −0.01 0.00 −0.13 −3.06 < .01
Observed maternal sensitivity −0.13 0.08 −0.12 −1.63 .11 −0.02 0.03 −0.03 −0.53 .60 0.04 0.05 0.04 0.73 .47
Temperament
 Activity level −0.09 0.07 −0.06 −1.38 .17 −0.03 0.03 −0.04 −0.92 .36 0.04 0.04 0.04 0.95 .34
 Attentional focusing −0.13 0.07 −0.11 −2.17 .03 −0.08 0.03 −0.13 −2.79 .01 0.05 0.04 0.05 1.20 .23
 Fear 0.11 0.05 0.08 2.14 .04 0.02 0.03 0.03 0.66 .51 −0.05 0.04 −0.05 −1.25 .22
 Shyness −0.06 0.04 −0.05 −1.49 .14 −0.03 0.02 −0.06 −1.59 .11 0.05 0.03 0.06 1.70 .09
Attachment at 15 months
 Anxious 0.39 0.15 0.34 2.51 .01 0.07 0.07 0.14 1.04 .30 0.07 0.10 0.08 0.71 .48
 Avoidant −0.06 0.15 −0.05 −0.42 .68 0.00 0.06 0.01 0.07 .94 −0.07 0.10 −0.08 −0.68 .50
 Disorganized −0.08 0.13 −0.07 −0.61 .55 −0.04 0.06 −0.07 −0.68 .50 0.11 0.08 0.13 1.32 .19
Open in a new tab

Note. N = 1,364. Standard errors are shown for unstandardized coefficients and t-statistics are shown for standardized coefficients. Unstandardized coefficients were calculated using unstandardized predictors and outcomes, except observed maternal sensitivity was kept as a standardized predictor because the composite is inherently unitless. Controls included gender, race/ethnicity, maternal education, income-to-needs ratio, and geographic location. The adjusted R2 for the ideological conservativism model was .13, F(24, 894) = 4.11, p < .001. The adjusted R2 for the Republican Party affiliation model was .09, F(24, 986) = 3.27, p < .001. The adjusted R2 for the Democratic Party affiliation model was .13, F(24, 998) = 5.11, p < .001.

Footnotes

The authors have no conflicts of interest to disclose.

Survey instruments and data sets for the current study are available from http://www.icpsr.umich.edu/icpsrweb/ICPSR/series/00233

References

  1. Adorno TW, Frenkel-Brunswik E, Levinson DJ, & Sanford RN (1950). The authoritarian personality. New York: Harper and Row. [Google Scholar]
  2. Ainsworth MDS, Blehar MC, Waters E, & Wall SN (1978). Patterns of attachment: A psychological study of the strange situation. Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. [DOI] [Google Scholar]
  3. Altemeyer B (1981). Right-wing authoritarianism. Winnipeg, Manitoba, Canada: University of Manitoba Press. 10.1017/s0003055400189488 [DOI] [Google Scholar]
  4. Astuto J, & Ruck MD (2010). Early childhood as a foundation for civic engagement. In Sherrod LR, Torney-Purta J, Flanagan CA, Sherrod LR, Torney-Purta J, & Flanagan CA (Eds.), Handbook of research on civic engagement in youth (pp. 249–275). Hoboken, NJ: Wiley. 10.1002/9780470767603.ch10 [DOI] [Google Scholar]
  5. Bachman JG (1974). Young men in high school and beyond: A summary of findings of the youth in transition project, 1966–1974. Ann Arbor, MI: Inter-university Consortium for Political and Social Research. 10.3886/icpsr03505 [DOI] [Google Scholar]
  6. Block J, & Block JH (2006). Nursery school personality and political orientation two decades later. Journal of Research in Personality, 40(5), 734–749. 10.1016/j.jrp.2005.09.005 [DOI] [Google Scholar]
  7. Bowlby J (1973). Separation, anxiety and anger (Attachment and loss, Vol. 2). New York: Basic Books. [Google Scholar]
  8. Bretherton I, & Munholland KA (2008). Internal working models in attachment relationships: Elaborating a central construct in attachment theory. In Cassidy J & Shaver PR (Eds.), Handbook of attachment: Theory, research, and clinical applications (2nd ed., pp. 102–127). New York, NY: Guilford Press. [Google Scholar]
  9. Carney DR, Jost JT, Gosling SD, & Potter J (2008). The secret lives of liberals and conservatives: Personality profiles, interaction styles, and the things they leave behind. Political Psychology, 29(6), 807–840. 10.1111/j.1467-9221.2008.00668.x [DOI] [Google Scholar]
  10. Cassidy K, & Marvin RS; the MacArthur Working Group on Attachment of the John D. and Catherine T. MacArthur Foundation Network on the Transition From Infancy to Early Childhood. (1992). Attachment organization in three-and four-year olds: Procedures and coding manual. Unpublished manuscript, University of Virginia. [Google Scholar]
  11. De Wolff MS, & van Ijzendoorn MH (1997). Sensitivity and attachment: A meta-analysis on parental antecedents of infant attachment. Child Development, 68(4), 571–591. 10.2307/1132107 [DOI] [PubMed] [Google Scholar]
  12. Duckitt J (2001). A dual-process cognitive-motivational theory of ideology and prejudice. Advances in Experimental Social Psychology, 33, 41–113. 10.1016/s00652601(01)80004-6 [DOI] [Google Scholar]
  13. Fraley RC, Griffin BN, Belsky J, & Roisman GI (2012). Developmental antecedents of political ideology: A longitudinal investigation from birth to age 18 years. Psychological Science, 23(11), 1425–1431. 10.1177/0956797612440102 [DOI] [PubMed] [Google Scholar]
  14. Gillath O, & Hart J (2010). The effects of psychological security and insecurity on political attitudes and leadership preferences. European Journal of Social Psychology, 40(1), 122–134. 10.1002/ejsp.614 [DOI] [Google Scholar]
  15. 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. 10.1017/s0022381609090938 [DOI] [Google Scholar]
  16. Hibbing JR, Smith KB, & Alford JR (2014). Differences in negativity bias underlie variations in political ideology. Behavioral and Brain Sciences, 37, 297–307. [DOI] [PubMed] [Google Scholar]
  17. Holbein JB (2017). Childhood skill development and adult political participation. American Political Science Review, 111(3), 572–583. 10.1017/s0003055417000119 [DOI] [Google Scholar]
  18. Huddy L, Feldman S, & Weber C (2007). The political consequences of perceived threat and felt insecurity. The Annals of the American Academy of Political and Social Science, 614, 131–153. 10.1177/0002716207305951 [DOI] [Google Scholar]
  19. Jennings MK, & Niemi RG (1968). The transmission of political values from parent to child. American Political Science Review, 62(1), 169–184. 10.2307/1953332 [DOI] [Google Scholar]
  20. Jost JT (2017). Ideological asymmetries and the essence of political psychology. Political Psychology, 38(2), 167–208. 10.1111/pops.12407 [DOI] [Google Scholar]
  21. Jost JT, Federico CM, & Napier JL (2009). Political ideology: Its structure, functions, and elective affinities. Annual Review of Psychology, 60, 307–337. 10.1146/annurev.psych.60.110707.163600 [DOI] [PubMed] [Google Scholar]
  22. Jost JT, Glaser J, Kruglanski AW, & Sulloway FJ (2003). Political conservatism as motivated social cognition. Psychological Bulletin, 129(3), 339. 10.1037/0033-2909.129.3.339 [DOI] [PubMed] [Google Scholar]
  23. Koleva SP, & Rip B (2009). Attachment style and political ideology: A review of contradictory findings. Social Justice Research, 22(2–3), 241–258. 10.1007/s11211-009-0099-y [DOI] [Google Scholar]
  24. Lilienfeld SO, & Latzman RD (2014). Threat bias, not negativity bias, underpins differences in political ideology. Behavioral and Brain Sciences, 37(3), 318–319. [DOI] [PubMed] [Google Scholar]
  25. Main M, & Solomon J (1990). Procedures for identifying infants as disorganized/disoriented during the Ainsworth Strange Situation. In Greenberg MT, Ciccetti D, & Cummings EM (Eds.), Attachment in the preschool years (pp. 121–160). Chicago, IL: University of Chicago Press. [Google Scholar]
  26. Mikulincer M, & Florian V (2000). Exploring individual differences in reactions to mortality salience: Does attachment style regulate terror management mechanisms?. Journal of personality and social psychology, 79(2), 260. 10.1037//0022-3514.79.2.260 [DOI] [PubMed] [Google Scholar]
  27. Mikulincer M, & Shaver PR (2003). The attachment behavioral system in adulthood: Activation, psychodynamics, and interpersonal processes. Advances in Experimental Social Psychology, 35, 56–152. 10.1016/s0065-2601(03)01002-5 [DOI] [Google Scholar]
  28. Mondak JJ (2010). Personality and the Foundations of Political Behavior. Cambridge, UK: Cambridge University Press. 10.1017/cbo9780511761515 [DOI] [Google Scholar]
  29. Murray RG, & Mulvaney MK (2012). Parenting styles, socialization, and the transmission of political ideology and partisanship. Politics & Policy, 40(6), 1106–1130. 10.1111/j.1747-1346.2012.00395.x [DOI] [Google Scholar]
  30. NICHD Early Child Care Research Network. (2001). Nonmaternal care and family factors in early development: An overview of the NICHD Study of Early Child Care. Journal of Applied Developmental Psychology, 22(5), 457–492. 10.1016/s0193-3973(01)00092-2 [DOI] [Google Scholar]
  31. Ojeda C, & Hatemi PK (2015). Accounting for the child in the transmission of party identification. American Sociological Review, 80(6), 1150–1174. 10.1177/0003122415606101 [DOI] [Google Scholar]
  32. Patterson MM, Bigler RS, Pahlke E, Brown CS, Hayes AR, Ramirez MC, & Nelson A (2019). Toward a developmental science of politics. Monographs of the Society for Research in Child Development, 84(3), 7–185. 10.1111/mono.12410 [DOI] [PubMed] [Google Scholar]
  33. Pratto F, Sidanius J, Stallworth LM, & Malle BF (1994). Social dominance orientation: A personality variable predicting social and political attitudes. Journal of Personality and Social Psychology, 67(4), 741. 10.1037//0022-3514.67.4.741 [DOI] [Google Scholar]
  34. Reifen Tagar M, Federico CM, Lyons KE, Ludeke S, & Koenig MA (2014). Heralding the authoritarian? Orientation toward authority in early childhood. Psychological science, 25(4), 883–892. 10.1177/0956797613516470 [DOI] [PubMed] [Google Scholar]
  35. Rokeach M (1960). The open and closed mind: Investigations into the nature of belief systems and personality systems. New York: Basic Books. [Google Scholar]
  36. Rothbart MK, Ahadi SA, Hershey KL, & Fisher P (2001). Investigations of temperament at three to seven years: The Children’s Behavior Questionnaire. Child Development, 72(5), 1394–1408. 10.1111/1467-8624.00355 [DOI] [PubMed] [Google Scholar]
  37. Schaefer ES, & Edgerton M (1985). Parent and child correlates of parental modernity. In Sigel IE (Ed.), Parental belief systems: The psychological consequences for children (pp. 287–318). Hillsdale, NJ: Erlbaum. [Google Scholar]
  38. Schulenberg JE, Johnston LD, O’Malley PM, Bachman JG, Miech RA, & Patrick ME (2018). Monitoring the future national survey results on drug use, 1975–2017: Volume II, College students & adults ages 19–55. Ann Arbor, MI: Institute for Social Research, The University of Michigan. 10.3998/2027.42/148123 [DOI] [Google Scholar]
  39. Sears DO, & Brown C (2013). Childhood and adult political development. In Huddy L, Sears D, & Levy J (Eds.), The Oxford handbook of political psychology (2nd ed., pp. 60–109). Oxford University Press. 10.1093/oxfordhb/9780199760107.013.0003 [DOI] [Google Scholar]
  40. Sherman LJ, Rice K, & Cassidy J (2015). Infant capacities related to building internal working models of attachment figures: A theoretical and empirical review. Developmental Review, 37, 109–141. 10.1016/j.dr.2015.06.001 [DOI] [Google Scholar]
  41. Torney-Purta JV (2017). The development of political attitudes in children. Routledge. 10.4324/9781315131726 [DOI] [Google Scholar]
  42. Twenge JM, Honeycutt N, Prislin R, & Sherman RA (2016). More polarized but more independent: Political party identification and ideological self-categorization among US adults, college students, and late adolescents, 1970–2015. Personality and Social Psychology Bulletin, 42(10), 1364–1383. 10.1177/0146167216660058 [DOI] [PubMed] [Google Scholar]
  43. Weise DR, Pyszczynski T, Cox CR, Arndt J, Greenberg J, Solomon S, & Kosloff S (2008). Interpersonal politics: The role of terror management and attachment processes in shaping political preferences. Psychological Science, 19(5), 448–455. 10.1111/j.1467-9280.2008.02108.x [DOI] [PubMed] [Google Scholar]
  44. White IR, Royston P, & Wood AM (2011). Multiple imputation using chained equations: Issues and guidance for practice. Statistics in Medicine, 30(4), 377–399. 10.1002/sim.4067 [DOI] [PubMed] [Google Scholar]

RESOURCES