Sage Journals: Discover world-class research

Abstract

Political polarization has intensified in the United States since the 1970s. The present research concerned adult attachment style as a potential explanation for ideological divisions. Few studies have examined this, mainly in college student samples, and their results conflict. The present research contributes to the literature because it is the first one to focus on the relationship of adult attachment to ideology in general population samples. Ideology was measured by self-placement on a 7-point scale. Attachment was measured by Carver’s 14-item self-reported Measure of Attachment Qualities (MAQ). Structural equation models were estimated for path analyses. The results for the direct effect of secure attachment style on ideology were mixed and indicated that both liberals and conservatives can have secure attachments. Results also underscored the potential importance of higher socioeconomic status for secure relational attachment style. Higher religiosity and lower trust in mass media contributed to conservatism.

Keywords

adult attachment ideology religiosity media trust socioeconomic status political psychology

Introduction

Scholars of political ideology are increasingly interested in adult attachment theory to understand better the role of ideology in thought processes and behaviors from childhood to adulthood (Fraley, Griffin, Belsky, & Roisman, 2012; Huddy, Feldman, & Weber, 2007; Koleva & Rip, 2009). Different attachment styles may condition different ideological worldviews (Buck, 2014; Roccato, Russo, & Senestro, 2013; Thornhill & Fincher, 2007; Weber & Federico, 2007). Only a few studies have examined the relationship of adult attachment relational styles to ideology, mainly in college student samples in the United States,¹ and their results differ (Dunkel & Decker, 2012; Gillath & Hart, 2010; Koleva & Rip, 2009; Thornhill & Fincher, 2007; Weber & Federico, 2007; Weise et al., 2008). The objective of the present research was to study ideology and attachment in general population samples.

Ideological divisions have deepened in the United States in the past 40 years, making it increasingly difficult for partisan groups to agree or compromise on policy and legislation (Abramowitz, 2010; Abramowitz & Saunders, 2005; Gauchat, 2012). Well-educated citizens tend to be the most politically involved and the most likely to identify with an ideological stance (Abramowitz, 2010; Pew Research Center, 2016). Among other factors contributing to the increasing ideological impasse are intensifying differences among the generations (Pew Research Center, 2016) and increasing divides between the highly religious and the nonreligious (Twenge, Sherman, Exline, & Grubbs, 2016). Being religious is often related to being conservative (Brossard, Scheufele, Kim, & Lewenstein, 2009; Gauchat, 2012; Roccato, 2008). Another factor is citizens’ growing preferences for information sources associated with an ideological orientation. Conservatives tend to view mainstream news media with distrust and to prefer outlets that share their views; liberals are disposed to trust news media, particularly those perceived as congruent with their positions (Center for Political Communication, 2012; Diercks & Landreville, 2017; Gaziano, 2014; Hindman & Yan, 2015; Jamieson & Cappella, 2009; Johansen & Joslyn, 2008; Lee, 2010; Morris, 2005).

The Current Research

The goal was to explore the relationship of attachment and ideology in the context of variables identified as important to trends shaping polarization: formal education, income, age, religiosity, and trust in mass media. Adult attachment was conceptualized as an independent variable because it derives theoretically from infant attachment, a process occurring in the first few years of life. Gender was also considered an independent variable because it is determined before birth.

The earliest relationships that humans know influence their perceptions of power and authority (Ainsworth, Blehar, Waters, & Wall, 1978; Barker & Tinnick, 2006; Feldman & Stenner, 1997; Grossman, Grossmann, & Waters, 2005; Lakoff, 2002; Mikulincer & Shaver, 2003). Political ideological orientations, in part, are based in attitudes toward power and authorities.

Independent Variable: Attachment Theory

Infant attachment

Adult attachment conceptualizations build on infant attachment theory, although there are many differences between these areas. Infant attachment is a process that provides “internal working models” for relationships throughout life and for development of cognitive skills, trust, confidence, and motivation to learn as the child matures (Bowlby, 1969/1982; de Ruiter & van IJzendoorn, 1993; Fraley, 2002; Mikulincer, 1998).

The attachment process develops from the interaction pattern of the primary caregiver and the infant and is defined by how well the caregiver can respond to the emotional needs of the infant. Parents who consistently provide comfort from anxiety and distress tend to foster secure attachments in their babies. Those parents who do not respond to their infants’ emotional needs with dependable regularity are likely to promote insecure attachments. Securely attached infants learn to trust their caregiver as a safe base from which to explore their world (Ainsworth, 1991; Sroufe & Waters, 1977). Insecurely attached children have learned not to trust that their caregiver will comfort their anxieties and provide safety; they tend to feel powerless, have low self-esteem, and approach relationships with trepidation.

Types of attachment

Attachments, therefore, may be secure or insecure, and insecure attachments are divided into three main categories: anxious-ambivalent, anxious-avoidant (Ainsworth et al., 1978), and disorganized (Main & Solomon, 1986). The latter type has seldom been studied in adults and is not discussed here. Ambivalence is characterized by hyperactive attachment behavior in search of security: fear of abandonment, high anxiety, and desire to merge with other people (Ainsworth et al., 1978; Cassidy & Kobak, 1988; Main, 1990). Carver (1997) further distinguished between two anxious-ambivalent styles (ambivalent-worry, or tendency to worry about loss, and ambivalent-merger, or desire for closeness). Avoidance involves lowering anxiety by deactivating the attachment system, restricting close relationships and emotional connections with others, relying on the self, and seeking control over others (Brennan, Clark, & Shaver, 1998; Griffin & Bartholomew, 1994; Mikulincer, 1998).

Research on adult attachment and ideology

Studies of adult attachment as a psychological aspect of political ideology in college students, either in experiments or surveys in the United States, yield conflicting results (Dunkel & Decker, 2012; Gillath & Hart, 2010; Koleva & Rip, 2009; Thornhill & Fincher, 2007; Weber & Federico, 2007; Weise et al., 2008). Student samples diverge enough from nonstudent samples in political interest and knowledge and other characteristics that general population samples should be preferred when studying such constructs (Koleva & Rip, 2009). Peterson (2001) also raised the question of differences found between college student samples and the general population in effect sizes and directionality.

Four student studies of adult attachment style and ideology measured ideology as conservatism/liberalism (Dunkel & Decker, 2012; Thornhill & Fincher, 2007; Weber & Federico, 2007; Weise et al., 2008). Three also included social dominance orientation (SDO) and/or right-wing authoritarianism (RWA), concepts not part of the present research (Thornhill & Fincher, 2007; Weber & Federico, 2007; Weise et al., 2008). A fifth inquiry into these relationships (Gillath & Hart, 2010) looked solely at priming effects, and two analyses (Weber & Federico, 2007; Weise et al., 2008) included primes as part of their research. Primes were unrelated to liberalism/conservativism and are not discussed further. Dunkel and Decker (2012) found that ambivalent and avoidant attachments were related to liberalism, and Thornhill and Fincher (2007) observed that secure attachments were associated with an indicator of conservatism, C-Scale scores. In contrast, Weise et al. (2008) reported that security of attachment was associated with liberal ideology. Weber and Federico (2007) reported a significant positive indirect effect of anxious attachment on left-right placement (a scale combining ideology and partisanship, where high scores indicated greater conservativism/Republicanism), through RWA and a tendency to view the world as dangerous.

Measurement of adult attachment

The student studies varied in how they operationalized attachment. Two relied on the Adult Attachment Questionnaire (AAQ; Dunkel & Decker, 2012; Thornhill & Fincher, 2007), one used the Experiences in Close Relationships (ECR) measure (Weber & Federico, 2007), and one used the Relationship Scales Questionnaire (RSQ; Weise et al., 2008).

The current study used another gauge, the Measure of Attachment Qualities (MAQ; Carver, 1997), selected as the most appropriate measure in a general population survey because it is much shorter than the RSQ or ECR and a little shorter than the AAQ. Brevity was important because the questionnaires were limited to 50 questions. It has been used effectively in previous research (Carver, 1997; Hollist & Miller, 2005; Kim & Carver, 2007; Kim, Carver, Deci, & Kasser, 2008; Roccato, 2008; Roccato & Ricolfi, 2005; Roccato et al., 2013; Segal, Needham, & Coolidge, 2009).

Dependent Variable: Ideology

Jost (2006) defined ideology, following Tedin (1987), as “an interrelated set of moral and political attitudes” that organize people’s values and beliefs, guide and explain political behavior, and are relatively stable over time (p. 653). Conservatives tend to support traditional attitudes and values, maintain caution in innovation or change, and emphasize respect for authority, obedience, and self-discipline in child-rearing. Liberals tend to be more open to social and political change, to support social equality, and to favor self-fulfillment, empathy, and nurturance in child-rearing (Jost, Glaser, Kruglanski, & Sulloway, 2003a; Koleva & Rip, 2009; Lakoff, 2002).²

The present research concerns a liberal–conservative scale of political views that has been a staple of American public opinion polls and related research. The majority of studies of adult attachment relational style and ideology analyzed by Koleva and Rip (2009) measured ideology on a unidimensional scale anchored at each end by liberalism and conservatism.

Various scales measure ideological dimensions, such as RWA (Altemeyer, 1981, 1998), and SDO (Sidanius & Pratto, 1999). According to Jost (2006), however,

Most treatments of political ideology have focused on the left–right (or, especially in the United States, the liberalism–conservatism) distinction (Jost et al.,2003a, 2003b; Knight, 1990). . . Although the left–right distinction is by no means airtight, it has been the single most useful and parsimonious way to classify political attitudes for more than 200 years. (p. 654)

A check of questions measuring political ideology during the past 40 years or more used by major organizations such as the American National Election Studies (ANES), the National Opinion Research Center, the Pew Research Center, and the Roper Center supports this conclusion.

Research Question 1: Is adult attachment a predictor of political ideology in a general population sample?

Intervening Variables

Variables that may influence the relationship between attachment and ideology were measured. These included trust in mass media reporting, two other related types of trust, interest in news and public affairs, religiosity, and demographics.

Trust in mass media and other trust

Three types of trust appear in much political and mass communication research: trust in media to report news accurately and fairly, trust in people, and trust in government to act fairly and democratically. Trust in people and trust in government are vital antecedents to political participation (Moy & Scheufele, 2000). If social segments and news media are segregated ideologically, then media distrust impedes the political process because widespread publication of news is essential in a democratic society (Jones, 2004).

Public affairs interest

This variable was included in initial analyses because attachment in infants sets the course for interest in new experiences and ideas, deriving from a sense of agency and competence (Ainsworth et al., 1978; Suess, Grossmann, & Sroufe, 1992). It may also be connected to adults’ interest in public affairs. Public affairs interest is related to political knowledge and participation (Eveland & Scheufele, 2000; Gaziano, 1997; Liu & Eveland, 2005; Viswanath & Finnegan, 1996).

Religiosity

Several studies demonstrate that people comprehend complicated or controversial issues such as abortion or global warming using their value predispositions, many of which tend to be religious values or heuristic cues from news media, religious institutions, political elites, and other sources (Brossard et al., 2009; Diercks & Landreville, 2017; Hindman, 2009; Ho, Brossard, & Scheufele, 2008). Many researchers operationalize religiosity as evaluation of the importance of religion in respondents’ lives and their frequency of attending religious services (Brossard et al., 2009; Gauchat, 2012; Ho et al., 2008; Malka & Lelkes, 2010; Roccato, 2008; Roccato & Ricolfi, 2005).

Demographics

Education, in particular, as well as household income, age, gender, and race are indicators of ideological divides (Abramowitz, 2010; Diercks & Landreville, 2017; Gauchat, 2012; Gaziano, 2013, 2014, 2017; Hindman, 2009, 2012; Hindman & Yan, 2015; Pew Research Center, 2016; Veenstra, Hossain, & Lyons, 2014).

Research Question 2: Which other variables are related to adult attachment, and how are they related to each other?

Study 1

Method

Participants

A random national general population sample in the United States (N = 330) was obtained from SurveyMonkey at www.surveymonkey.com between September 27 and October 1, 2013. The response rate was 35.1% of individuals invited to participate. The 50-question survey, “Opinions About Topics in the News, Interpersonal, and Social Issues,” took about 10 min to complete. Two additional samples were drawn in 2014 and are reported later. SurveyMonkey is an online survey site that maintains a large, diverse, national panel of people who have completed a profile survey and have agreed to respond to surveys commissioned by SurveyMonkey customers. The company reports that it maintains regular benchmarking surveys to help ensure that panel members are representative of the population. SurveyMonkey audiences do not receive money. They receive credits toward donations to charitable organizations and sweepstakes entries. SurveyMonkey removes any information identifying respondents before the data are given to investigators. Advantages of an Internet survey are cost-effectiveness and speed. Disadvantages are that American adults with Internet access tend to be of higher socioeconomic status (SES) than those who lack access and may be different in other ways from non-Internet samples (see, for example, Zickuhr & Rainie, 2013).

Independent variables

Adult attachment

The MAQ is a self-report measure of adult attachment containing 14 items with four Likert-type categories, ranging from “disagree with the statement a lot” to “agree with the statement a lot” (Carver, 1997; Kim & Carver, 2007). Four subscales containing three to five items assess secure, avoidant, ambivalent-worry, and ambivalent-merger patterns. Examples of items are “It feels relaxing and good to be close to someone” and “I prefer not to be too close to others.”³ Responses to the four subscales were summed according to Carver’s directions. The subscales had adequate internal consistency: security, α = .71; avoidant, α = .80; ambivalent-worry, α = .77; and ambivalent-merger, α = .75. Summary statistics for variables are in the appendix. Because the numbers of statements for each type of attachment were unequal, responses were averaged.

Gender

This is described under “Demographics” section below.

Dependent variable: Ideology

Respondents were asked, “Here is a 7-point scale on which the political views that people might hold are arranged from extremely liberal to extremely conservative. Where would you place yourself on this scale?” Participants also were asked which political party described their views the best on a 7-point scale ranging from “strong Republican” to “strong Democrat,” recoded so that high meant strong Republican; however, party identification was not used in the present analysis because it was highly correlated with ideology (r = .70, p < .001).

Intervening variables

Trust in government, media, and people

The questions were (a) “How much of the time do you think you can trust the federal government in Washington to make decisions in a fair way?” (b) “How much of the time do you think you can trust the media to report the news fairly?” (c) “Generally speaking, how often can you trust other people?” The questions used 5-point scales and were separated in the questionnaire. Answer categories were always, most of the time, about half the time, once in a while, and never (recoded so that high scores denoted high trust).

Interest in public affairs

People were asked how interested they were in what was going on in government and public affairs on a 5-point scale anchored by “extremely interested” and “not interested at all” (recoded so that high scores signified high interest).

Religiosity

Participants rated the importance of religion in their lives and their frequency of attending religious services on 5-point scales. The end points were “extremely important” and “not important at all” for the first measure and “every week” to “never” for the second. Responses were combined to form a religiosity score, averaged, and recoded so that high scores meant highly religious (Cronbach’s α = .86).

Demographics

Participants were asked the year they were born, the last grade they completed in school, household income, gender, and race. Age was measured as continuous, and the others were measured as categories. Compared with census data (Table 1), the sample overrepresented women, the middle-aged, and people with higher education and household incomes. Minority racial groups were not well represented, and race was discarded from additional analysis. A t-test comparison of the sample mean and the mean of a sample from the ANES⁴ in 2012 using the identical question about ideology showed that the present study significantly overrepresented liberals (t = 3.95, df = 1,565, p < .001, two-tailed). Table 1 shows two additional 2014 samples, to be discussed later as Study 2.

Table 1.

Demographic Comparisons of 2013 and 2014 Samples With Census Data.

Variables	2013 general population sample(N = 330)(%)	2014 general population sample(N = 223)(%)	2014 conservative sample(N = 202)(%)	2013 American community survey^a(N = 163,663)(%)
Gender
Males	45.5	44.8	55.4	48.6
Females	54.5	55.2	44.6	51.4
Age
18-24 years	7.9	14.3	7.4	12.8
25-34 years	16.1	17.9	11.4	17.6
35-44 years	22.4	17.5	14.9	16.8
45-54 years	20.3	23.3	16.3	18.3
55-64 years	20.3	16.6	25.7	16.2
65 years or older	13.0	10.3	24.3	18.3
Education
<High school degree	0.6	0.9	0.5	13.6^b
High school degree	9.7	7.6	12.4	28.0
Some college	35.7	34.1	34.7	29.2
College degree	30.0	30.9	28.2	18.2
Some graduate work/graduate degree	23.9	26.5	24.3	11.0
Household income
Less than US$20,000	9.4	13.5	7.9	18.8
US$20,000-US$34,999	11.6	13.0	11.4	16.0
US$35,000-US$49,999	10.6	15.2	14.9	13.7
US$50,000-US$74,999	23.7	13.0	20.3	18.1
US$75,000-US$99,999	17.9	15.7	15.3	11.8
US$100,000-US$149,999	17.6	17.9	13.9	12.4
US$150,000 or more	9.1	11.7	16.3	9.2

Note. Some totals do not add to 100% because of rounding.

Retrieved from https://www.census.gov/acs/www/methodology/sample-size-and-data-quality/sample-size/

Based on adults aged 25 years or older.

Analytical method

Causal modeling with path analysis in a structural equation model with maximum likelihood estimation in AMOS 23.0 was selected to evaluate the potential relationships with attachment and to estimate how well these variables fit the model for each of the three samples. Path analysis allows the consideration of both direct and indirect effects of exogenous variables on dependent variables. Causality is inferred for gender and security of adult attachment, the observed exogenous variables, as those variables precede the others in time order. Observed endogenous variables were trust in mass media to report news fairly, trust in the government to make fair decisions, religiosity, public affairs interest, respondents’ age, education, household income, and ideology. Unobserved exogenous variables were error variances that cause response variation in the observed endogenous variables. Structural equation modeling permits multiple pathways through intervening variables. Figure 1 presents the conceptual model.

Figure 1.

The theoretical model.

Figures 2 to 4, below, report standardized regression coefficients that attained statistical significance. Standardized regression coefficients (β) range from +1.0 to −1.0 and allow direct comparisons of the relative strength of each predictor in a model and comparisons of results across the figures.

Figure 2.

The sample in 2013 with significant paths shown except for the correlation between gender and attachment security (ns).

Figure 3.

General population sample in 2014 with significant paths shown except for the link between gender and attachment security (ns).

Figure 4.

Conservative sample in 2014 with significant paths shown except for the correlation between gender and attachment security (ns).

Study 1 Results

Table 2 provides the correlations among ideology, the four attachment variables, religiosity, the three trust items, public affairs interest, and three demographic variables. Among the attachment measures, highest relationship was between security and avoidance (negative) and between ambivalent-worry and ambivalent-merger (positive). These results were similar to previous reports (Carver, 1997; Kim & Carver, 2007; Kim et al., 2008). Correlations among the other attachment variables were moderate. Security was significantly and negatively related to ideology and was the only attachment item retained for further analysis. (Worry was related to ideology only at r = .10, p < .10, and initial testing in a path model showed that the relationship decreased markedly when other variables were taken into account.) Ideology was significantly and positively correlated with religiosity and significantly and negatively correlated with media trust and government trust but unrelated to trust in people. Media trust was uncorrelated with religiosity, though it might have been expected to be associated because Golan and Day (2010) reported mixed evidence of a relationship in their investigation of media credibility and religiosity. Public affairs interest was not related to any of the other variables at .05 or less, and it was discarded from further tests as was trust in people.

Table 2.

Pearson Correlations for Main Variables in 2013 Survey of General Population (N = 330).

	Ideology^a	Security	Avoidance	Worry	Merger	Religiosity	Trust in media	Trust in government	Trust in other people	Public affairs interest	Education	Household income	Age
Ideology	—
Security	−.15**	—
Avoidance	−.04	−.44***	—
Worry	−.10^†	.01	.30***	—
Merger	−.07	.04	.29***	.49***	—
Religiosity	.40***	.03	−.22***	−.14*	.00	—
Trust in media	−.30***	.22***	−.12*	.06	.05	−.00	—
Trust in government	−.23***	.05	−.14**	−.06	.08	−.01	.44***	—
Trust in other people	−.06	.24***	−.41***	−.19***	−.23***	.13*	.20***	.32***	—
Public affairs interest	−.07	−.02	−.06	−.10^†	.02	−.01	.06	.05	−.10^†	—
Education	−.09^†	.09	−.06	−.05	.05	.05	.11^†	.19***	.11*	.11^†	—
Household income	.05	.13*	−.14**	−.09^†	−.07	−.03	.08	.07	.15**	−.01	.24***	—
Age	.10^†	.05	−.08	−.17**	.01	.11*	.16**	.09	.07	.26***	.16**	.05	—

High = conservative, low = liberal.

†

p = .10. *p = .05. **p = .01. ***p = .001 (two-tailed).

A structural equation model was estimated for hypothesis testing in a path analysis in which attachment security and gender were exogenous variables and were allowed to have direct effects on religiosity, media trust, government trust, age, education, household incomes, and ideology. Then religiosity, trust in media reporting, trust in government, age, education, and income were permitted to have direct effects on ideology.

The resulting path model is given in Figure 2, showing standardized coefficients. The model was a good fit to the data: χ² = 18.866, 19 df, p = .465 (45 distinct sample moments, 26 distinct parameters to be estimated), N = 330, CMIN/df = .993 (chi-square statistic that is the minimum value of the discrepancy between the model and the data, divided by its degrees of freedom), Bentler’s comparative fit index (CFI) = 1.0, Tucker–Lewis Index (TLI) = 1.001, and the root mean square error of approximation (RMSEA) = .001. The criteria considered desirable for goodness of fit statistics were the following: a nonsignificant chi-square, RMSEA less than .05, CFI and TLI of .95 or higher, and CMIN/df of less than 2 (Hox & Bechger, 1998; Hu & Bentler, 1995). The results fit the hypothesized model well.

The first research question asked if adult attachment was related to political ideology. Figure 2 shows that security of attachment had a direct effect of β = –.12 (p < .05) on ideology. In addition, the standardized coefficient for an indirect effect was β = –.03, for a standardized total effect of β = –.15. The negative sign indicated that liberals were higher in security than conservatives. The second research question concerned which other variables were related to adult attachment and how they might relate to each other. Secure attachment also was a significant predictor of trust in media reporting and possessing higher household incomes. Gender had no direct effect on ideology, and its indirect effect was negligible. The only effect of gender was on household income, β = –.12 (p < .05), indicating that men tended more than women to have higher household incomes. Distrust of government and distrust of media were significantly related to conservatism, as were being older, less educated, and having higher household incomes. Religiosity was a significant predictor of conservative ideology at β = .40 (p < .001) and was the most related to ideology, among all the variables.

Discussion

Inspired by a desire to untangle the relations between adult attachment and ideology because of mixed results in previous research, the present study sought to explore this relationship in a general population sample. The primary finding was that secure attachment was a significant predictor of liberal ideology. No known previous research on ideological orientation and attachment style has included media trust or government trust. Secure adult attachment led to trust in media, and media trust predicted liberalism. Security was not significantly related to trust in government, but government trust was correlated with media trust and also predicted liberalism.

Importance of religion predicted conservatism, consonant with other research (Gaziano, 2014). Religion was not a significant predictor in Thornhill and Fincher’s (2007) work, but their operational definition of religiosity may have been too simple to achieve predictive ability. It was a variable coded as “1 = religion” and “0 = no religion.”

Older age, lower education, and higher household incomes also predicted conservatism. Previous research has found conservatives tending to have lower education than liberals but tending to have higher household incomes than liberals (Diercks & Landreville, 2017⁵; Gaziano, 2014; Hindman & Yan, 2015).

Study 2

Method

Participants

A national random sample of U.S. adults 18 years or older (N = 223) and a separate national random sample of relatively conservative adults 18 years or older (N = 202) to improve representation of conservatives in the study were drawn on September 23 to 24, 2014, through SurveyMonkey. The 2014 general population sample and the 2013 general population sample were not significantly different in ideology (t = 1.59, df = 551, p < .11, two-tailed). The 2014 general population sample, however, was significantly more liberal than the ANES 2012 general population sample (t = 5.44, df = 1,458, p < .001, two-tailed).

Dependent variable: Ideology

Ten questions comprising an Ideological Consistency Scale (ICS) used by the Pew Research Center (2014) were included as a check on the validity of the 7-point scale measuring ideology, and a high correlation between them increased confidence in the liberal–conservative scale. Respondents chose the one statement in a pair of statements that came closer to their views, where one statement represented a conservative point of view, and one represented a liberal stance. For example, “Government regulation of business usually does more harm than good” (conservative) OR “Government regulation of business is necessary to protect the public interest” (liberal). Conservative responses were scored “+1,” and liberal responses were scored “–1.” Don’t know/refused/volunteered responses counted as “0.” Responses were summed to form the scale (Cronbach’s α = .78, general population sample; .58, conservative sample). In the general population sample, ideology and the ICS were highly related (r = .75, p < .001), as were the ICS and party identification (r = .71, p < .001), and ideology and partisanship (r = .77, p < .001). Among the conservatives, results were r = .24, ICS and ideology; r = .48, ICS and party identification; and r = .33, ideology and partisanship, all significant at .001.

Independent variable: Adult attachment style

The measures of security, avoidance, ambivalent-worry, and ambivalent-merger were constructed as in Study 1. The internal consistency for the general population sample was security, α = .68; avoidant, α = .83; ambivalent-worry, α = .73; and ambivalent-merger, α = .75, comparable with the 2013 results. Among the conservative sample, results were, respectively, .73, .78, .70, and .75.

Intervening variables

Religiosity

The scale for religiosity was created as in Study 1 with Cronbach’s α = .88, general population sample, and .78, conservative sample. The conservative sample surpassed the general population sample significantly in how religious they were.

Media trust

The general population sample significantly exceeded in trusting the media, compared with the conservative sample. Trust in the government to make good decisions and trust in people were not asked in Study 2.

Demographics

As Table 1 shows, all the samples considerably overrepresented higher SES groups. As with the 2013 sample, the two 2014 samples were significantly more likely to be White and non-Hispanic than the general public. Like the 2013 sample, the 2014 general population sample slightly underrepresented men. The 2014 sample of conservatives was significantly more likely to be male and older than the 2014 and 2013 general population samples. Their SES differences were not significant.

Study 2 Results

Tables 3 and 4 present the intercorrelations for the same measures as used in Study 1. Table 3 shows that in the 2014 general population sample, security correlated inversely and substantially with avoidance at –.53 (p < .001), and security was related to worry at .08 (ns) and merger at .16 (p < .05). Avoidance was associated with worry and with merger both at .20 (p < .01). Worry and merger correlated at .60 (p < .001). In the 2014 conservative sample, results were similar, except that security and worry correlated at –.25 (p < .001) and security and merger showed no significant relationship (Table 4). Ideology and security were inversely related but not significant in the general population sample. Contrary to these results, in the conservative sample, ideology and security were positively and significantly correlated.

Table 3.

Pearson Correlations for Main Variables in 2014 Survey of General Population (N = 223).

	Ideology^a	Security	Avoidance	Worry	Merger	Religiosity	Trust in media	Education	Household income	Age
Ideology	—
Security	−.08	—
Avoidance	.07	−.53***	—
Worry	.08	.08	.20**	—
Merger	.04	.16*	.20**	.60***	—
Religiosity	.41***	.07	−.11	−.01	−.03	—
Trust in media	−.09	.18**	−.15*	−.04	−.09	.08	—
Education	−.20**	.17**	−.08	−.17**	−.14*	.03	−.00	—
Household income	−.03	.11	−.04	−.20**	−.16*	−.01	.17*	.36***	—
Age	.16*	−.06	−.07	−.16*	−.07	.21**	.14*	.14*	.28***	—

High = conservative, low = liberal.

p = .05. **p = .01. ***p = .001 (two-tailed).

Table 4.

Pearson Correlations for Main Variables in 2014 Survey of Conservatives (N = 202).

	Ideology^a	Security	Avoidance	Worry	Merger	Religiosity	Trust in media	Education	Household income	Age
Ideology	—
Security	.15*	—
Avoidance	−.04	−.59***	—
Worry	−.07	−.25***	.30***	—
Merger	−.04	−.08	.30***	.53***	—
Religiosity	.24***	.06	−.15*	−.07	.06	—
Trust in media	−.10	.08	−.05	.03	.06	.09	—
Education	−.02	.06	−.02	−.12^†	−.10	.08	−.07	—
Household income	.04	.18*	−.10	−.06	−.12^†	−.07	−.04	.22**	—
Age	−.00	.04	−.08	−.10	−.11	−.04	.11	.16*	.15*	—

High = conservative, low = liberal.

†

p = .10. *p = .05. **p = .01. ***p = .001 (two-tailed).

Figure 3 illustrates the results of the path analysis in the population at large in 2014, including the same variables as in Figure 2, except for government trust. It had an adequate model fit with χ² = 10.432, df = 12 (36 distinct sample moments, 24 distinct parameters), p = .578, n = 223, CMIN/df = .869, CFI = 1.000, TLI = 1.027, RMSEA = .001. Although the standardized direct effect of security on ideology was –.024, the standardized indirect effect was –.065 for a total standardized effect of –.089. Despite being small and not significant, the direction of relationship mirrored that in the 2013 sample, that is, that security was related to liberalism.

Gender directly affected ideology in a significant negative pattern, indicating that men were more conservative than women. This result was not observed in Figure 2.

The patterns of the relationships of media trust, religiosity, age, and education to ideology were similar to those in Study 1. Media trust was negatively related to conservative ideology, β = –.11 (but only p < .10), and religiosity was positively correlated with conservative ideology, β = .41 (p < .001). The direct effect of security of attachment on income was similarly positive and significant, β = .13 (p < .05). Security also directly affected education (β = .18, p < .01). Age affected ideology as in Study 1, but only at p < .10.

Figure 4 depicts the results for the conservative sample. The model achieved a satisfactory fit to the hypothesized model with χ² = 6.758, df = 8 (21 distinct sample moments, 13 distinct parameters), p = .563, N = 202, CMIN/df = .845, CFI = 1.000, TLI = 1.08, and RMSEA = .001.

Here, secure style of attachment was directly, positively, and significantly related to ideology (β = .14, p < .05), with no indirect effect. The direction of this relationship was opposite of that found in Figures 2 and 3. Security maintained the positive direction of effect on household income, however, β = .18 (p < .01), consistent with results in Figures 2 and 3, but it had no direct effect on education as it had in Figure 3 for the 2014 general population sample. Women tended more than men to have faith in fairness of media reporting, a pattern that did not occur in the other two samples. The conservative sample differed from the general population sample in 2014 in the lack of relationships between security and media trust or between gender and ideology. It was like the other two samples in the relation of low media trust and high religiousness to conservatism.

Limitations

The Internet samples were of significantly higher SES and differed somewhat in age from census data from the 2013 American Community Survey. The 2013 and 2014 general population samples underrepresented men, and the 2014 conservative sample overrepresented men. Both of the general population samples in 2013 and 2014 were significantly more liberal than the 2012 ANES, although a third sample of conservatives was drawn for balance. These differences weaken the ability of this study to generalize to the population at large. Ideally, the research should be repeated with larger, more representative samples. Because the present research and the four previous reports on college students’ ideology and attachment have been with samples in the United States, such work should be carried out in other parts of the globe among different kinds of populations.

In addition, ideology in the present research was measured as self-placement on a single 7-point self-report item. Although numerous polls and surveys have found this type of measure to be reliable and valid, other measures of liberalism and conservatism could be used for a more multidimensional examination of ideology. In 2014, results for ideology were compared with a measure of ideological consistency known to be a valid and reliable indicator of political philosophy over time (Pew Research Center, 2014), increasing confidence that the self-placement question is a justifiable gauge of political orientation. Interest in news of government and public affairs was not a predictor, but it may have been too general. Interest levels tended to be similarly high among liberals and conservatives and lagged among independents. The relation of ideology and interest in specific politically contested public affairs and science issues could be investigated further.

Not including government trust in the 2014 studies can be criticized as this variable was useful in the 2013 research. It could be included in future work.

Discussion and Conclusion

Political polarization is of increasing concern in the United States because it has intensified since the 1970s. The present research concerned adult relational attachment style as a potential contributor to ideological divisions. Only a few studies have examined this, primarily in college student samples, and their results differ. The present research contributes to the literature because it is the first one to focus on the relationship of adult attachment to ideology in general population samples. The main findings are as follows.

Study 1: General Population Sample

Liberals scored significantly higher than conservatives in secure attachment relational style. Secure attachment was significantly related to trust in media reporting and also to having higher household incomes. As would be expected, conservatives were significantly more likely than liberals to be religious and to distrust both government and mass media. Other significant predictors of conservatism were being older and less educated but possessing higher household incomes. Religiosity and media trust were uncorrelated. Gender, conceptualized as an independent variable, made no difference in ideology and was associated significantly only with household income, meaning that men had higher incomes.

Study 2: General Population Sample

Secure attachment was unrelated to ideology, although the combined direct and indirect effects appeared similar to the result in Study 1. The other independent variable, gender, however, was significantly and negatively related to ideology, meaning that men were more conservative than women. Gender had no effects on other variables. Security was positively and significantly associated with education and incomes. Conservatism was linked to lower trust in media but only at p < .10. It was strongly related to religiosity, as well as to lower education. There was a positive association with age, only at p < .10. Again, religiosity and media trust were unrelated.

Study 2: Conservative Sample

Secure attachment style was a significant and positive predictor of conservatism, contrary to the positive relationship in Study 1 and the lack of significant relationship in the Study 2 general population sample. As in Study 1 and Study 2 general population groups, security and household incomes were positively and significantly correlated. Consistent with the other two sample results, being religious was highly related to conservatism. Also congruent with the other two groups’ results, media trust was not connected to being religious. Lack of trust in media fairness of reporting and being religious independently had significant direct impacts on ideology. Neither education nor income had an effect on political ideology. Gender had a positive, significant effect on media trust, meaning that women were more trusting of media, and it was negatively linked to household incomes, meaning that men’s incomes were higher.

Security and ideology

Although most findings were relatively consistent in the three samples, the relationships of secure attachment style and ideology clearly were different in each one. Results for the previous studies of college student samples were mixed, also. These results, taken together, demonstrate that both liberals and conservatives can have secure attachments.

Security and SES

Security was consistently positively and significantly linked to household income in all three samples, and security positively and directly affected education in the 2014 sample of the general public. All the previous studies of college students and the present study concerned primarily more advantaged social groups. Perhaps greater SES promotes more secure attachments, regardless of political leanings.

Ideology and SES

Household income was significantly and positively connected to ideology only in 2013, but education was negatively and significantly related to ideology in both the general population samples. The finding of lower education directly influencing conservatism is consistent with other research that has reported lower education characterizing conservatives (Adorno, Frenkel-Brunswik, Levinson, & Sanford, 1950; Hindman & Yan, 2015; McCutcheon, 1985; Stone, Lederer, & Christie, 1993/2011; Stubager, 2008). It is not unusual to find that conservatives, especially if better educated, have higher household incomes than well-educated liberals, even though conservatives tend to be less well educated overall (Diercks & Landreville, 2017; Gaziano, 2014; Hindman & Yan, 2015). Well-educated liberals tended to be younger and in earlier stages of the life cycle than well-educated conservatives in Gaziano (2014). Better educated liberals were more likely than better educated conservatives to have advanced degrees.

Future Research

At least four areas of future study come to mind. These include analyses of adult attachment relational style in lower SES groups, attachment style and types of information environments, the relationship of attachment style and further elaboration of the models of nurturant parent and strict parent families, and examination of attachment styles in groups with extreme political attitudes. As most of the research so far has been done in the United States, research should be carried out in other countries, as well. Although the models did well enough in this study, they could be refined further and new concepts tested.

First, the results suggest that future research should emphasize the relationship of ideology to attachment in population segments with lower education and incomes. While the current project found security to be more relevant to ideological orientation than avoidance or anxiety, future researchers should concentrate more on forms of insecure attachment in lower SES groups with regard to a range of social and economic issues. The 2016 U.S. election points up the need to better understand the needs and concerns of White working class population segments, as well as other lower and working class ethnic and racial segments. For White working class groups, declining economic and job opportunities was a primary element in their vote decisions (Wan, 2017). Related elements were rising rates of mortality and morbidity among these groups (Achenbach & Keating, 2017; Case & Deaton, 2017). They also were more likely than higher SES groups to feel like strangers in their own country (Cox, Lienesch, & Jones, 2017). In general, voters of various SES groups who supported Donald J. Trump for president tended to support racist attitudes (Wood, 2017), to accept false information and untruths that support one’s views (Sykes, 2017), and to defend vulgarity, crude behavior, and bullying (Martin, 2017). The social conditions that foster such beliefs and attitudes could be delved into with an eye to addressing solutions to these issues.

Factors that promote less secure attachment styles among lower SES groups may help to explain attitudes and beliefs in these groups. Complicating factors in the links between lower SES and insecure attachment is that lower SES families are vulnerable to multiple stresses of poverty, such as poorer health, lack of stable employment, teen parent and/or single-parent status, and living in more crowded conditions, which can drain parents physically and emotionally and consequently constrain their ability to parent as well as they would wish (Bakermans-Kranenburg, van IJzendoorn, & Bradley, 2005; Bavolek, 2001; Conger, Patterson, & Ge, 1995; Hackman, Farah, & Meaney, 2010; Lynch, Kaplan, & Shema, 1997).

Second, security of attachment style and types of information channels may yield insights into the consequences of fragmented media environments and insular discussion networks as political polarization increases. These processes may have implications for knowledge and belief gaps, flows of inaccurate information, and tolerance of differing viewpoints (Duca & Saving, 2017; Garrett, Weeks, & Neo, 2016). Numbers and types of information networks can be explored. Researchers have revealed sociological fault lines in the dissemination of information in mainstream and partisan news sources and ideological blogs (Eveland & Dylko, 2007; Hindman & Yan, 2015; Jamieson & Cappella, 2009). Fragmentation can lead to decreased understanding of conflicting views and diminished tolerance for diversity, particularly among more partisan groups who may be among the easiest to mobilize to support political candidates for office (Baum & Groeling, 2008; Dilliplane, 2011; Iyengar & Hahn, 2009; Shapiro & Bloch-Elkon, 2008; Tewksbury, 2005).

Third, attachment styles could be studied further within the framework of nurturant parent and strict parent families, and further insights could be gained by refining the conceptualization of liberals as coming from nurturant parent families versus conservatives as coming from strict parent families (Barker & Tinnick, 2006; Feldman & Stenner, 1997; Lakoff, 2002). These conceptualizations currently may be too simplistic. One avenue for future research in adults is retrospective recall of perceived parenting styles or longitudinal studies from childhood or adolescence through adulthood. One approach could be contrasting parental demandingness, or control, with parental responsiveness, or warmth (see, for example, Baumrind, 1971, 1996; Maccoby & Martin, 1983; Murray & Mulvaney, 2012). Other insights could be gained from comparing family communication patterns and attachment (Gaziano, 2001).

Fourth, attachment and extreme ideologies should be studied more. An interesting example is that of Roccato and Ricolfi (2005), who examined six nonstudent right-wing and left-wing political militant groups in Turin and Palermo, Italy. While avoidantly attached participants were far less religious than securely attached participants in the present study, the Italian study described one group of activists who were the most overtly antidemocratic, right-wing, and avoidant in attachment style, yet who were also among the most religious of the militants. This suggests that additional insights into attachment and ideology may be gained by studying politically active groups of different types.

Footnotes

Appendix

Descriptive Statistics.

	M	SD	n
2013 study, general population
Ideology	3.74	1.669	330
Political party identification	3.46	1.726	313
Education	3.67	0.989	330
Household income	4.18	1.760	330
Age	46.58	14.767	330
Secure	16.44	3.009	330
Avoidant	10.37	3.465	330
Ambivalent-worry	8.76	3.878	330
Ambivalent-merger	9.58	3.537	330
Religiosity	5.53	2.927	330
Trust in media	2.67	0.884	330
Trust in government	2.37	0.907	330
Trust in people	3.41	0.792	328
Interest in public affairs	3.52	1.061	328
2014 study, general population
Ideology	3.52	1.494	223
Political party identification	3.43	1.606	204
Education	3.74	0.964	223
Household income	4.05	1.955	223
Age	43.67	15.372	223
Secure	16.44	3.108	223
Avoidant	10.60	3.461	223
Ambivalent-worry	9.14	3.730	223
Ambivalent-merger	9.76	3.627	223
Religiosity	5.14	2.870	223
Trust in media	3.01	0.872	223
2014 study, conservatives
Ideology	5.80	0.904	202
Political party identification	5.45	1.319	194
Education	3.63	1.000	202
Household income	4.31	1.849	202
Age	51.31	16.311	202
Secure	17.10	2.740	202
Avoidant	9.97	3.211	202
Ambivalent-worry	8.08	3.404	202
Ambivalent-merger	9.26	3.390	202
Religiosity	7.22	2.747	202
Trust in media	2.63	0.867	202

Declaration of Conflicting Interests

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

Funding

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

Notes

Author Biography

Cecilie Gaziano holds master’s and doctoral degrees in mass communication theory and methods from the University of Minnesota and is a fellow of the Midwest Association for Public Opinion Research. Her research interests include the knowledge gap and belief gap hypotheses, mass media and their audiences, and social science and medical research on parents, children, and families.

References

Abramowitz

A. I.

(2010). The disappearing center: Engaged citizens, polarization, and American democracy. New Haven, CT: Yale University Press.

Abramowitz

A. I.

Saunders

(2005). Why can’t we all just get along? The reality of a polarized America. The Forum, 3(2), Article 1. doi:10.2202/1540-8884.1076

Achenbach

Keating

(2017, March 23). Unnatural causes: Sick and dying in small-town America—A new divide in American death (Introduction to a series of articles). The Washington Post. Retrieved from https://www.washingtonpost.com/unnatural-causes/?utm_term=.8e2899a1141a

Adorno

T. W.

Frenkel-Brunswik

Levinson

D. J.

Sanford

R. N.

(1950). The authoritarian personality. New York, NY: Harper & Row.

Ainsworth

M. D. S.

(1991). Attachments and other affectional bonds across the life cycle. In Parkes

C. M.

Stevenson-Hinde

Marris

(Eds.), Attachment across the life cycle (pp. 33-51). London, England: Routledge.

Ainsworth

M. D. S.

Blehar

M. C.

Waters

Wall

S. N.

(1978). Patterns of attachment: A psychological study of the strange situation. Hillsdale, NJ: Lawrence Erlbaum.

Altemeyer

(1981). Right-wing authoritarianism. Winnipeg, Canada: University of Manitoba Press.

Altemeyer

(1998). The other “authoritarian personality.” Advances in Experimental Social Psychology, 30, 47-92.

Aspelund

Lindeman

Verkasalo

(2013). Political conservatism and left-right orientation in 28 Eastern and Western European countries. Political Psychology, 34, 409-417.

10.

Bakermans-Kranenburg

M. J.

van IJzendoorn

M. H.

Bradley

R. H.

(2005). Those who have, receive: The Matthew effect in early childhood intervention in the home environment. Review of Educational Research, 75, 1-26.

11.

Barker

D. C.

Tinnick

J. D.

III . (2006). Competing visions of parental roles and ideological constraint. American Political Science Review, 100, 249-263.

12.

Baum

M. A.

Groeling

(2008). New media and the polarization of American political discourse. Political Communication, 25, 345-365.

13.

Baumrind

(1971). Current patterns of parental authority. Developmental Psychology Monograph, 4(1, Pt. 2), 1-103.

14.

Baumrind

(1996). The discipline controversy revisited. Family Relations, 45, 405-414.

15.

Bavolek

S. J.

(2001). The Florida report: The impact of gender, race, income and education on parenting attitudes: July 1999 to July 2001. Ashville, NC: Family Development Resources.

16.

Bowlby

(1982). Attachment and loss (Vol. 1). New York, NY: Basic Books. (Original work published 1969)

17.

Brennan

K. A.

Clark

C. L.

Shaver

P. R.

(1998). Self-report measurement of adult attachment: An integrative overview. In Simpson

J. A.

Rholes

W. S.

(Eds.), Attachment theory and close relationships (pp. 46-76). New York, NY: Guilford Press.

18.

Brossard

Scheufele

D. A.

Kim

Lewenstein

B. V.

(2009). Religiosity as a perceptual filter: Examining processes of opinion formation about nanotechnology. Public Understanding of Science, 18, 546-558.

19.

Buck

(2014). Emotional attachment security as the origin of liberal-conservative differences in vigilance to negative features of the environment. Behavioral and Brain Sciences, 37, 308-309.

20.

Carver

C. S.

(1997). Adult attachment and personality: Converging evidence and a new measure. Personality and Social Psychology Bulletin, 23, 865-883.

21.

Case

Deaton

(2017, May 1). Mortality and morbidity in the 21st century. Washington, DC: The Brookings Institution. Retrieved from https://www.brookings.edu/bpea-articles/mortality-and-morbidity-in-the-21st-century/

22.

Cassidy

Kobak

R. R.

(1988). Avoidance and its relation to other defensive processes. In Belsky

Nezworski

(Eds.), Clinical implications of attachment (pp. 300-323). Hillsdale, NJ: Lawrence Erlbaum.

23.

Center for Political Communication. (2012, February 15). National survey shows Fox News viewers are least informed, most negative, about the Occupy Wall Street movement. Newark: University of Delaware.

24.

Conger

R. D.

Patterson

G. R.

(1995). It takes two to replicate: A mediational model for the impact of parents’ stress on adolescent adjustment. Child Development, 66, 80-97.

25.

Cox

Lienesch

Jones

R. P.

(2017, May 9). Beyond economics: Fears of cultural displacement pushed the white working class to Trump: PRRI/The Atlantic report. Retrieved from https://www.prri.org/research/white-working-class-attitudes-economy-trade-immigration-election-donald-trump/

26.

de Ruiter

van IJzendoorn

M. H

. (1993). Attachment and cognition: A review of the literature. International Journal of Educational Research, 19, 525-540.

27.

Diercks

Landreville

K. D.

(2017). The indirect effects of partisanship and partisan media on knowledge about same-sex marriage policy: Exploring the knowledge and belief gap hypotheses. Mass Communication and Society, 20, 192-212.

28.

Dilliplane

(2011). All the news you want to hear: The impact of partisan news exposure on political participation. Public Opinion Quarterly, 75, 287-316.

29.

Duca

J. V.

Saving

J. L.

(2017). Income inequality, media fragmentation, and increased political polarization. Contemporary Economic Policy, 35, 392-413.

30.

Dunkel

C. S.

Decker

(2012). Using identity style and parental identification to predict political orientation. Current Psychology, 31, 65-78.

31.

Eveland

W. P.

Jr. Dylko

(2007). Reading political blogs during the 2004 election campaign: Correlates and political consequences. In Tremayne

(Ed.), Blogging, citizenship, and the future of media (pp. 105-126). New York, NY: Routledge.

32.

Eveland

W. P.

Jr. Scheufele

D. A.

(2000). Connecting news media use with gaps in knowledge and participation. Political Communication, 17, 215-237.

33.

Feldman

Stenner

(1997). Perceived threat and authoritarianism. Political Psychology, 18, 741-770.

34.

Fraley

R. C.

(2002). Attachment stability from infancy to adulthood: Meta-analysis and dynamic modeling of developmental mechanisms. Personality and Social Psychology Review, 6, 123-151.

35.

Fraley

R. C.

Griffin

B. N.

Belsky

Roisman

G. I.

(2012). Developmental antecedents of political ideology: A longitudinal investigation from birth to age 18 years. Psychological Science, 23, 1425-1431.

36.

Garrett

R. K.

Weeks

B. E.

Neo

R. L.

(2016). Driving a wedge between evidence and beliefs: How online ideological news exposure promotes political misperceptions. Journal of Computer-Mediated Communication, 21, 331-348.

37.

Gauchat

(2012). Politicization of science in the public sphere: A study of public trust in the United States, 1974 to 2010. American Sociological Review, 77, 167-187.

38.

Gaziano

(1997). Forecast 2000: Widening knowledge gaps. Journalism & Mass Communication Quarterly, 74, 237-264.

39.

Gaziano

(2001). Toward a broader conceptual framework for research on social stratification, childrearing patterns, and media effects. Mass Communication and Society, 4, 219-244.

40.

Gaziano

(2013). Knowledge gaps, belief gaps, ideology, and culture wars. Open Journal of Political Science, 3, 116-130.

41.

Gaziano

(2014, January 10). Components of the belief gap: Ideology and education. SAGE Open. doi:10.1177/2158244013518052

42.

Gaziano

(2017, January 9). Adult attachment style and beliefs about public affairs and scientific issues. Cogent Psychology. doi:10.1080/23311908.2017.1279423

43.

Gillath

Hart

(2010). The effects of psychological security and insecurity on political attitudes and leadership preferences. European Journal of Social Psychology, 40, 122-134.

44.

Golan

G. J.

Day

A. G.

(2010). In God we trust: Religiosity as a predictor of perceptions of media trust, factuality, and privacy invasion. American Behavioral Scientist, 54, 120-136.

45.

Griffin

D. W.

Bartholomew

(1994). Models of the self and other: Fundamental dimensions underlying measures of adult attachment. Journal of Personality and Social Psychology, 67, 430-445.

46.

Grossman

K. E.

Grossmann

Waters

(2005). Attachment from infancy to adulthood: The major longitudinal studies. New York, NY: Guilford Press.

47.

Hackman

D. A.

Farah

M. J.

Meaney

M. J.

(2010). Socioeconomic status and the brain: Mechanistic insights from human and animal research. Nature Reviews Neuroscience, 11, 651-659.

48.

Hindman

D. B.

(2009). Mass media flow and differential distribution of politically disputed beliefs: The belief gap hypothesis. Journalism & Mass Communication Quarterly, 86, 790-808.

49.

Hindman

D. B.

(2012). Knowledge gaps, belief gaps, and public opinion about health care reform. Journalism & Mass Communication Quarterly, 89, 585-605.

50.

Hindman

D. B.

Yan

(2015). The knowledge gap versus the belief gap and abstinence-only sex education. Journal of Health Communication, 20, 949-957.

51.

S. S.

Brossard

Scheufele

D. A.

(2008). Effects of value predispositions, mass media use, and knowledge on public attitudes toward embryonic stem cell research. International Journal of Public Opinion Research, 20, 171-192.

52.

Hollist

C. S.

Miller

R. B.

(2005). Perceptions of attachment style and marital quality in midlife marriage. Family Relations, 54, 46-57.

53.

Hox

J. J.

Bechger

T. M.

(1998). An introduction to structural equation modeling. Family Science Review, 11, 354-373.

54.

L.-t.

Bentler

P. M.

(1995). Evaluating model fit. In Hoyle

R. H.

(Ed.), Structural equation modeling: Concepts, issues, and applications (pp. 76-99). Thousand Oaks, CA: Sage.

55.

Huddy

Feldman

Weber

(2007). The political consequences of perceived threat and felt insecurity. Annals of the American Academy of Political and Social Science, 614, 131-153.

56.

Iyengar

Hahn

K. S.

(2009). Red media, blue media: Evidence of ideological selectivity in media use. Journal of Communication, 59, 19-39.

57.

Jamieson

K. H.

Cappella

J. N.

(2009). Echo chamber: Rush Limbaugh and the conservative media establishment. New York, NY: Oxford University Press.

58.

Johansen

M. S.

Joslyn

M. R.

(2008). Political persuasion during times of crisis: The effects of education and news media on citizens’ factual information about Iraq. Journalism & Mass Communication Quarterly, 85, 591-608.

59.

Jones

D. A.

(2004). Why Americans don’t trust the media: A preliminary analysis. The International Journal of Press/Politics, 9, 60-75.

60.

Jost

J. T.

(2006). The end of the end of ideology. American Psychologist, 61, 651-670.

61.

Jost

J. T.

Glaser

Kruglanski

A. W.

Sulloway

F. J.

(2003a). Exceptions that prove the rule—Using a theory of motivated social cognition to account for ideological incongruities and political anomalies. Psychological Bulletin, 129, 383-393.

62.

Jost

J. T.

Glaser

Kruglanski

A. W.

Sulloway

F. J.

(2003b). Political conservatism as motivated social cognition. Psychological Bulletin, 129, 339-375.

63.

Kim

Carver

C. S.

(2007). Frequency and difficulty in caregiving among spouses of individuals with cancer: Effects of adult attachment and gender. Psycho-Oncology, 16, 714-723.

64.

Kim

Carver

C. S.

Deci

E. L.

Kasser

(2008). Adult attachment and psychological well-being in cancer caregivers: The mediational role of spouses’ motives for caregiving. Health Psychology, 27, S144-S154.

65.

Knight

(1990). Ideology and public opinion. In Long

(Ed.), Research in micropolitics (Vol. 3, pp. 59-82). Greenwich, CT: JAI Press.

66.

Koleva

S. P.

Rip

(2009). Attachment style and political ideology: A review of contradictory findings. Social Justice Research, 22, 241-258.

67.

Lakoff

(2002). Moral politics: How liberals and conservatives think (2nd ed.). Chicago, IL: The University of Chicago Press.

68.

Lee

T.-T.

(2010). Why they don’t trust the media: An examination of factors predicting trust. American Behavioral Scientist, 54, 8-21.

69.

Liu

Y.-I.

Eveland

W. P.

Jr. (2005). Education, need for cognition, and campaign interest as moderators of news effects on political knowledge: An analysis of the knowledge gap. Journalism & Mass Communication Quarterly, 82, 910-929.

70.

Lynch

J. W.

Kaplan

G. A.

Shema

S. J.

(1997). Cumulative impact of sustained economic hardship on physical, cognitive, psychological, and social functioning. New England Journal of Medicine, 337, 1889-1895.

71.

Maccoby

E. E.

Martin

J. A.

(1983). Socialization in the context of the family: Parent–child interaction. In Mussen

P. H.

(Ed.), Handbook of child psychology (Vol. 4, pp. 1-101). New York, NY: John Wiley.

72.

Main

(1990). Cross-cultural studies of attachment organization: Recent studies, changing methodologies, and the concept of conditional strategies. Human Development, 33, 48-61.

73.

Main

Solomon

(1986). Discovery of a disorganized/disoriented attachment pattern. In Brazelton

T. B.

Yogman

M. W.

(Eds.), Affective development in infancy (pp. 95-124). Westport, CT: Ablex.

74.

Malka

Lelkes

(2010). More than ideology: Conservative–liberal identity and receptivity to political cues. Social Justice Research, 23, 156-188.

75.

Martin

(2017, February 16). Miss Manners on rudeness in the age of Trump. The Atlantic. Retrieved from https://www.theatlantic.com/entertainment/archive/2017/02/alternative-virtues-for-the-trump-era/516498/

76.

McCutcheon

A. L.

(1985). A latent class analysis of tolerance for nonconformity in the American public. Public Opinion Quarterly, 49, 474-488.

77.

Mikulincer

(1998). Attachment working models and the sense of trust: An exploration of interaction goals and affect regulation. Journal of Personality and Social Psychology, 74, 1209-1224.

78.

Mikulincer

Shaver

P. R.

(2003). The attachment behavioral system in adulthood: Activation, psychodynamics, and interpersonal processes. In Zanna

M. P.

(Ed.), Advances in experimental social psychology (Vol. 35, pp. 53-152). New York, NY: Academic Press.

79.

Morris

J. S.

(2005). The Fox News factor. Harvard International Journal of Press/Politics, 10, 56-79.

80.

Moy

Scheufele

D. A.

(2000). Media effects on political and social trust. Journalism & Mass Communication Quarterly, 77, 744-759.

81.

Murray

G. R.

Mulvaney

M. K.

(2012). Parenting styles, socialization, and the transmission of political ideology and partisanship. Politics & Policy, 40, 1106-1130.

82.

Peterson

R. A.

(2001). On the use of college students in social science research: Insights from a second-order meta-analysis. Journal of Consumer Research, 28, 450-461.

83.

Pew Research Center. (2014). Political polarization in the American public: How increasing ideological uniformity and partisan antipathy affect politics, compromise and everyday life. Retrieved from http://www.people-press.org/2014/06/12/political-polarization-in-the-american-public/

84.

Pew Research Center. (2016). A wider ideological gap between more and less educated adults: Political polarization update. Retrieved from http://www.people-press.org/2016/04/26/a-wider-ideological-gap-between-more-and-less-educated-adults/

85.

Piurko

Schwartz

S. H.

Davidov

(2011). Basic personal values and the meaning of left-right political orientations in 20 countries. Political Psychology, 32, 537-561.

86.

Roccato

(2008). Right-wing authoritarianism, social dominance orientation, and attachment: An Italian study. Swiss Journal of Psychology, 67, 219-229.

87.

Roccato

Ricolfi

(2005). On the correlation between right-wing authoritarianism and social dominance orientation. Basic and Applied Social Psychology, 27, 187-200.

88.

Roccato

Russo

Senestro

(2013). Adult attachment styles, life experiences, right-wing authoritarianism, and social dominance orientation. Psicologia Sociale, 8, 249-265.

89.

Segal

D. L.

Needham

T. N.

Coolidge

F. L.

(2009). Age differences in attachment orientations among younger and older adults: Evidence from two self-report measures of attachment. International Journal of Aging and Human Development, 69, 119-132.

90.

Shapiro

R. Y.

Bloch-Elkon

(2008). Do the facts speak for themselves? Partisan disagreement as a challenge to democratic competence. Critical Review, 20, 115-139.

91.

Sidanius

Pratto

(1999). Social dominance: An intergroup theory of social hierarchy and oppression. Cambridge, UK: Cambridge University Press.

92.

Sroufe

L. A.

Waters

(1977). Attachment as an organizational construct. Child Development, 48, 1184-1199.

93.

Stone

W. F.

Lederer

Christie

(2011). Strength and weakness: The authoritarian personality today. New York, NY: Springer-Verlag. (Original work published 1993)

94.

Stubager

(2008). Education effects on authoritarian–libertarian values: A question of socialization. British Journal of Sociology, 59, 327-350.

95.

Suess

G. J.

Grossmann

K. E.

Sroufe

L. A.

(1992). Effects of infant attachment to mother and father on quality of adaptation in preschool: From dyadic to individual organisation of self. International Journal of Behavioral Development, 15, 43-65.

96.

Sykes

C. J.

(2017, February 4). Sunday Review—Opinion: Why nobody cares the President is lying. The New York Times. Retrieved from https://www.nytimes.com/2017/02/04/opinion/sunday/why-nobody-cares-the-president-is-lying.html?_r=0

97.

Tedin

K. L.

(1987). Political ideology and the vote. In Long

(Ed.), Research in micropolitics (Vol. 2, pp. 63-94). Greenwich, CT: JAI Press.

98.

Tewksbury

(2005). The seeds of audience fragmentation: Specialization in the use of online news sites. Journal of Broadcasting & Electronic Media, 49, 332-348.

99.

Thornhill

Fincher

C. L.

(2007). What is the relevance of attachment and life history to political values? Evolution & Human Behavior, 28, 215-222.

100.

Twenge

J. M.

Sherman

R. A.

Exline

J. J.

Grubbs

J. B.

(2016, March 23). Declines in American adults’ religious participation and beliefs, 1972-2014. SAGE Open. doi:10.1177/2158244016638133

101.

Veenstra

A. S.

Hossain

M. D.

Lyons

B. A.

(2014). Partisan media and discussion as enhancers of the belief gap. Mass Communication and Society, 17, 874-897.

102.

Viswanath

Finnegan

J. R.

Jr. (1996). The knowledge gap hypothesis: Twenty-five years later. In Burleson

B. R.

(Ed.), Communication Yearbook 19 (pp. 187-227). Thousand Oaks, CA: Sage.

103.

Wan

(2017, April 5). Democrats are still ignoring the people who could have helped them defeat Trump, Ohio party leaders say. The Washington Post. Retrieved from https://www.washingtonpost.com/national/democrats-are-still-ignoring-the-people-who-could-have-helped-them-defeat-trump-ohio-party-leaders-say/2017/04/04/405295e0-0428-11e7-ad5b-d22680e18d10_story.html?utm_term=.49ec601f5256

104.

Weber

Federico

C. M.

(2007). Interpersonal attachment and patterns of ideological belief. Political Psychology, 28, 389-416.

105.

Weise

D. R.

Pyszczynski

Cox

C. R.

Arndt

Greenberg

Solomon

Kosloff

(2008). Interpersonal politics: The role of terror management and attachment processes in shaping political preferences. Psychological Science, 19, 448-455.

106.

Wood

(2017, April 17). Racism motivated Trump voters more than authoritarianism. The Washington Post. Retrieved from https://www.washingtonpost.com/news/monkey-cage/wp/2017/04/17/racism-motivated-trump-voters-more-than-authoritarianism-or-income-inequality/?utm_term=.c9115880bc10

107.

Zickuhr

Rainie

(2013, November 29). 7 things to know about offline Americans. Washington, DC: Pew Research Center.

Adult Attachment Style and Political Ideology

Abstract

Keywords

Introduction

The Current Research

Independent Variable: Attachment Theory

Infant attachment

Types of attachment

Research on adult attachment and ideology

Measurement of adult attachment

Dependent Variable: Ideology

Intervening Variables

Trust in mass media and other trust

Public affairs interest

Religiosity

Demographics

Study 1

Method

Participants

Independent variables

Adult attachment

Gender

Dependent variable: Ideology

Intervening variables

Trust in government, media, and people

Interest in public affairs

Religiosity

Demographics

Analytical method

Study 1 Results

Discussion

Study 2

Method

Participants

Dependent variable: Ideology

Independent variable: Adult attachment style

Intervening variables

Religiosity

Media trust

Demographics

Study 2 Results

Limitations

Discussion and Conclusion

Study 1: General Population Sample

Study 2: General Population Sample

Study 2: Conservative Sample

Security and ideology

Security and SES

Ideology and SES

Future Research

Footnotes

Appendix

Declaration of Conflicting Interests

Funding

Notes

Author Biography

References