Political difference and polarization in the family: The role of (non)accommodating communication for navigating identity differences

Abstract

This study joins the relatively modest literature examining the effects of political disagreement in the family. We consider the effects of communication accommodation on shared family identity in the context political disagreement. To do this, we utilize survey responses from a quota-stratified sample of participants in an online panel (N = 833) taken immediately after the contentious 2016 presidential election. We find that more disagreement and more affective polarization are associated with less communication accommodation and that shared family identity suffers as a result. Furthermore, our findings reveal that respecting divergent values is the most influential communication accommodation strategy and is also among the most adversely affected by political differences in the family. We conclude that political disagreement in the family reduces the likelihood of communication that is respectful of differences in political values, but that this accommodation strategy is crucial to reduce the deleterious consequences that political differences can have on family relationships.

Keywords

Affective polarization communication accommodation theory political communication in the family political difference shared family identity

Pew Research Center found 33% of U.S. adults reported that only a few or none of their family shared their views (Oliphant, 2018); only 28% of individuals in this group were willing to discuss politics. More generally, 40% of adults try to avoid political conversations with their family members. These data provide evidence that political differences can constrain family dialog. Such constrained conversations can have deleterious effects on family relationships. After the hotly contested and wildly divisive 2016 presidential election, scholars found that families with political disagreement shortened their Thanksgiving trips by, on average, more than 40 minutes compared to 2015—a trend that resulted in an estimated loss of almost 74 million hours of Thanksgiving time with the family (Chen & Rohla, 2018). Family scholars are just beginning to investigate the effect of these political differences on family relationships. Afifi and colleagues (2020) found that married/cohabitating individuals who voted differently from their partners felt less relational closeness, exhibited less relational maintenance, and experienced increased conflict and stress. The cumulative effect of these negative interactions resulted in decreased relationship resilience.

Contentious political conversations may pose risks for familial bonds. Political disagreements have a tendency to become disrespectful and personal, and people usually withdraw from those with whom they disagree (Cowan & Baldassarri, 2018). Our knowledge of political disagreements, however, is largely based in political communication research. For example, scholars of political communication have adapted politeness norms (Brown & Levinson, 1987) to understand political incivility (Muddiman, 2017; Mutz, 2015). Their focus, however, has been on the political discourse of politicians and the news media or political interactions online (e.g., Coe et al., 2014; Frimer & Skitka, 2018; Muddiman & Stroud, 2017), and the effects of incivility have generally been studied with regard to their consequences on normatively desirable democratic behaviors (Brooks & Geer, 2007). Although family communication scholars and practitioners recognize complicated family structures and the implications thereof, family scholarship lacks an explicit focus on differences in families, especially as it relates to salient social identities such as race-ethnicity, religion, and—important to the current study—political affiliation (Soliz & Phillips, 2018; Turner & West, 2018). Because identity differences can cause relational difficulty (Soliz, & Rittenour, 2012), it is important to understand how families communicatively navigate these differences.

Given the long-term and involuntary nature of family relationships, it can be difficult to evade a family member with a political difference. Thus, differences must be managed through communication. How families navigate political differences may determine whether these differences ultimately undermine the collective family bond. The current study investigates the nature of familial conversations about political differences by assessing perceptions of (non)accommodation between family members in which political differences are most pronounced. We introduce communication accommodation theory (CAT) to explicate the role of accommodating political communication in promoting relational solidarity in the face of political differences. We apply literature about political polarization, intergroup theorizing, and CAT to theorize about political differences in family relationships. The result is a model of family communication in which perceptions of family members’ (non)accommodating communication mediates the effect of political difference on shared family identification.

Political polarization and the family

Political polarization is defined as an increase in the difference between how favorably people feel toward members of their own political party and how unfavorably they feel toward members of the other political party—a trend that has been labeled affective (Iyengar et al., 2012) or social (Mason, 2018) polarization by political scientists. There has been a steady decline in evaluations of political outgroups in recent decades, with partisans increasingly disliking or even hating members of the opposing political group (Iyengar et al., 2012). This outgroup animus is so intense that it results in stronger implicit bias, more hostile feelings, and more frequent discriminatory behaviors when compared to racial prejudice (Iyengar & Westwood, 2015). In fact, political scientists now consider partisanship the most significant social cleavage in the United States (Iyengar et al., 2018). It should therefore be no surprise that affective polarization has influenced non-political social interactions. People are less willing to hire those who do not share their political views (Gift & Gift, 2015; Iyengar & Westwood, 2015) and increasingly select romantic mates who agree with them about politics (Huber & Malhotra, 2017).

One consequence of increased affective polarization is greater social sorting—people prefer to live, marry, and work in spaces surrounded by those who share their views (Iyengar et al., 2018; Motyl et al., 2014). In this way, political polarization can be perpetuated within the family. Families exert considerable influence on political beliefs on their members (Dalton, 1980; Jennings & Niemi, 1968) primarily through political discussion (Ledbetter, 2015; McDevitt & Chaffee, 2002; Warner & Colaner, 2016). As a result, there is often a strong correspondence in political beliefs within a family (Zuckerman et al., 2007).

Though there are good reasons to expect familial agreement to be the norm, families are sprawling, interconnected networks with a variety of extended kin relationships. Families are one of the most likely social settings to generate contact with individuals who possess different social identities, including political affiliation (Colaner & Soliz, 2020). Young adults are increasingly individualizing their values in ways that create fractures in collective family values (Arnett & Jensen, 2002). Millennials are more progressive with regard to environmental, financial, and social policies (Pew Research Center, 2018). As a result, families with adult children are experiencing substantial differences in beliefs between generations. This evolution of individual thought can create discord in family relationships.

We assess polarization within the family by examining participants’ relationship with the family member with whom they disagree the most. This relationship offers the clearest site of potential discord in the family. Naturally, relationships in which there is little disagreement require little accommodation with regard to political conversations. Greater identity differences prompt more elaborate accommodation behaviors to balance ingroup family membership with outgroup political identification. Given that political division is associated with a host of hostile feelings (Iyengar et al., 2012; Iyengar & Westwood, 2015), the relationship with the greatest political difference may suffer from a lack of closeness or relational dissatisfaction. Importantly, participants report on the relationship that has the highest perceived difference. It is possible that these are the relationships that are marked by contentious political conversations that amplify feelings of difference, even if there are others in the family with a greater difference that is minimized through accommodative communication. Perhaps participants select the family member who has the most nonaccommodative communication rather than the greatest political differences. Either way, we focus on relationships in which non-accommodated political differences become salient. This allows us to generate important insight into the relationship between perceived political difference, perceived (non)accommodation, and feelings of shared family identity.

Family relationships are difficult if not impossible to dissolve, thus social sorting within family bonds is not an option. Families must navigate political polarization in structurally non-negotiable relationships. When political differences do emerge in the family, families have to balance collective group membership as a family with divergent individual social identities. Intergroup theorizing provides important insight into this process.

Intergroup theorizing

Intergroup theorizing presumes that social identities (such as race, religion, and political affiliation) guide our interactions and shape how we see ourselves and others (Giles et al., 2010). Research in social identity theory demonstrates that humans sort into ingroups, wherein individuals share salient identities, and outgroups, in which individuals possess a different identity (Tajfel & Turner, 1979). These groupings form the basis of our sense of self and influence how we interact with members of our ingroup as well as the outgroup.

Families are often one of the most important sources of belonging and identity, or as some refer to it, “the ultimate ingroup” (Rittenour, 2020). Shared family identity (SFI) refers to the degree to which family relationships function as an ingroup for individuals (Soliz & Harwood, 2006). Individuals tend to report high levels of identification with their family group members (Soliz & Rittenour, 2012). Family membership is a core source of identity, providing a sense of home, support, and security (Colaner & Soliz, 2020). SFI is an important indicator of relational quality because it reflects closeness in a variety of relationships such as grandparent-grandchild relationships (Soliz & Harwood, 2006), step-grandparent relationships (Soliz, 2007), in-law relationships (Rittenour & Soliz, 2009), multiracial families (Soliz et al., 2009), and adoptive families (Colaner et al., 2018). SFI is positively related to relational satisfaction (Rittenour & Soliz, 2009). Comparing the constructs, however, SFI may better indicate relational quality than relational satisfaction (Rittenour & Soliz, 2009), as one may feel very satisfied with a relationship characterized by low levels of closeness. In this case, the relationship may not be important to an individual as they are happy with that lack of closeness. SFI, however, illustrates the degree to which one feels a sense of belonging with another as co-members of a family ingroup.

Differences on other (non-familial) identity categories pose a challenge to SFI, especially as individual family members increasingly have within-group differences (Galvin, 2013). These social identity outgroups can clash with ingroup family identification (Soliz & Rittenour, 2012), positioning the family as a likely site of intergroup contact (Harwood, 2006). Harwood urges scholars to examine modern family relationships from an intergroup perspective, saying, “A social identity approach will be crucial in understanding when and how these individuals categorize as ingroup (family) versus outgroup (different races, faiths, ages, etc.) members and the influence of such categorization on family communication” (p. 87). With regard to the current study, political differences create an intergroup encounter in which opposing political identities create outgroups that place family members against one another at the polls and in everyday conversations.

Communicatively navigating political differences

Political disagreements within the family make cross-pressures more salient when people realize that their kin do not share their values. People often minimize political discussions in which they expect disagreement and conflict (Cowan & Baldassarri, 2018; Gerber et al., 2012; Huckfeldt, 2007). However, conversations about political differences often become uncivil in tone. Given that politics is aligned with religious, social, cultural, racial, and economic identities, people have much at stake when they disagree (Mason, 2018). Political disagreement often signals a more fundamental divergence in world views. For example, a post-election article in the New York Times profiled families with highly contentious interactions. One individual who voted for Trump lamented that members of her family accused her of being racist because of her vote. Another woman, this time a Democrat who became a naturalized U.S. citizen after immigrating from Ghana, described a vote for Trump as “a rejection of everyone who looks like [her]” (Tavernise & Seelye, 2016, para. 6). She elected to move her wedding to Italy to discourage her Republican family members from attending, a move reflective of the family discord heightened by the 2016 election.

Given the severity of political differences and the potential for relationship disruption, we explore how these differences can be communicatively managed in order to maintain harmonious family relationships. Communication accommodation theory (CAT) is a guiding theory of the communication behaviors related to intergroup contact in various contexts (Gallois et al., 2005). CAT outlines two communication processes available when social difference is salient: accommodative communication, where communication functions as a catalyst for interpersonal affiliation, and nonaccommodative communication, which serves to amplify or maintain distinctiveness and social distance.

Accommodative communication

Accommodative behaviors acknowledge and respect the conversational partner’s perceived interpersonal and divergent group characteristics through linguistic, paralinguistic, discursive, and nonlinguistic adaptation (Gallois et al., 2005; Harwood, 2000). Accommodative communication occurs in numerous ways, such as selecting appropriate topics of conversation, using inclusive body language, minimizing interruptions, and increasing interpretability of the interaction by adapting to the context and skill of the conversational partner (Harwood et al., 2006). In the present study, we target two accommodation behaviors demonstrated to be important in other family research taking an intergroup approach (e.g., Colaner et al., 2014; Harwood, 2000; Rittenour & Soliz, 2009; Soliz & Harwood, 2006; Soliz et al., 2009): supportive communication, or communication that signals support for an individual’s autonomy in developing views divergent from the family, and respecting divergent values, or communication that emphasizes a respectful disagreement about important issues.

With regard to supportive communication, incorporating elements of immediacy, receptiveness, and equality conveys care and concern for the conversational partner (Burleson & MacGeorge, 2002). Supportiveness has the potential to personalize interactions, reducing the negative effects of intergroup distinctions by emphasizing personal characteristics (Soliz et al., 2009). Supportive communication about political differences prioritizes the other’s values even when those values differ from one’s own. Supportive communication allows the partner to have distinct views without pressure to conform to one’s own opinions. Supportive communication likely plays an important role in contentious conversations about politics by demonstrating acceptance of an individual as a valued family member despite a divergent political affiliation.

Respecting divergent values, the second accommodative communication behavior in the current study, involves adapting communication in order to avoid offending, belittling, or disregarding another’s beliefs. Softening threatening statements, avoiding direct judgments, and expressing tolerance for another’s political affiliation can convey respect for another’s beliefs. For example, censoring insults about a partner’s preferred political candidate conveys respect for the partner to make their own political decisions. Respectfully disagreeing about politics gives space for divergent opinions by honoring personal attributes over political affiliation. Due to the highly personal nature of political values and the important relationship between respect of one’s social identity and individual well-being (De Cremer & Tyler, 2005), affirmation of one’s core political values is likely important in promoting family connectedness in the face of political difference.

Nonaccommodative communication

Whereas accommodative communication attempts to meet the conversational partner’s needs, nonaccommodative behaviors fail, intentionally or unwittingly, to incorporate the partner’s identity and communicative needs into the interaction (Gallois et al., 2005; Harwood, 2000). Again, two behaviors specific to political talk within families are highlighted in the present study based on previous intergroup research (e.g., Colaner et al., 2014; Harwood, 2000; Rittenour & Soliz, 2009; Soliz & Harwood, 2006; Soliz et al., 2009): inappropriate self-disclosure, or excessive communication about areas of disagreement, and emphasizing divergent values, or communication that brings unwanted attention to areas of disagreement.

Inappropriate self-disclosure occurs when individuals offer excessive and unwanted information about their political attitudes. Inappropriate self-disclosure is considered nonaccommodative because it fails to acknowledge the conversational partner’s needs and preferences (Harwood, 2000). Extended monologue on one’s own political attitudes can be polarizing to the conversational partner, potentially leading to communication dissatisfaction, discomfort, and dread over political interactions. The difference between appropriate and inappropriate self-disclosure depends upon individual preferences and subjective evaluation. Idiosyncratic assessments of disclosure stem from communication standards individuals have about how much political discussion feels appropriate. Thus, assessments of appropriateness are in the “eye of the beholder,” based upon the goodness of fit between how much disclosure an individual desires and how much disclosure an individual receives. Disclosures about political beliefs that are disproportionate to the expectations of the conversational partner will likely be considered inappropriate given that they highlight political differences and devalue family connectedness.

The second nonaccommodative communication behavior—emphasizing divergent values—aims to accentuate differences of salient social identities (Rittenour & Soliz, 2009). Individuals can draw attention to differing political identities indirectly by selecting topics that are known to be controversial or directly by openly criticizing another’s belief. Argumentative and aggressive communication within political conversations can prioritize individual political group membership over commonly shared features of the relationship.

Hypothesized model

The use of (non)accommodative communication has implications for family relationships. Across numerous studies, accommodative communication is related to relational solidarity and nonaccommodative communication is related to dissatisfying relationships in contexts such as mother-in-law/daughter-in-law relationships (Rittenour & Soliz, 2009), families with a member identifying as gay (Soliz et al., 2010), grandparent-grandchild relationships (Harwood, 2000; Soliz, 2007; Soliz & Harwood, 2006), multiethnic families (Soliz et al., 2009), and interfaith families (Colaner et al., 2014). We expect political differences to function similarly.

We focus in the current study on individuals’ perceptions of their family members’ (non)accommodating communication (as opposed to an individual’s own communication accommodation behaviors). This focus turns attention to the communication behaviors that amplify or minimize differences and the subsequent effect of these behaviors on relationships, illuminating how family members communicatively manage their identity differences. A main tenet of CAT is that there is a difference between actual and perceived communication, as Soliz and Colaner (2017) explain, “Regardless of what is actually—or objectively—occurring in the interaction, individuals in the interaction will have subjective perceptions, or evaluations, of these interactions” (p. 77). Thus, perceptions of others’ communication likely have an important impact on feelings of relational solidarity. We expect less perceived accommodating and more perceived non-accommodating communication to be associated with adverse family relationship outcomes as indicated by less shared family identity.

Additionally, because hostility to members of the political outgroup affects interpersonal interactions (Iyengar et al., 2018; Iyengar & Westwood, 2015), people in relationships that are characterized by more political difference and partisan animus should engage in less accommodating and more non-accommodating ways. These theoretical expectations are summarized in the model presented in Figure 1—which represents the following hypotheses:

H1: Familial political difference will be associated with less perceived accommodating political conversation in the family as indicated by (a) less supportive communication, (b) less respecting of divergent values, (c) more inappropriate self-disclosure, and (d) more emphasis of divergent values.

H2: Greater political animus will be associated with less perceived accommodating political conversation in the family as indicated by (a) less supportive communication, (b) less respecting of divergent values, (c) more inappropriate self-disclosure, and (d) more emphasis of divergent values.

H3a: More perceived accommodating political communication (supportive communication and respecting of divergent values) in the family will be associated with more shared family identity.

H3b: More perceived nonaccommodating political communication (inappropriate self-disclosure, and emphasis of divergent values) in the family will be associated with less shared family identity.

Figure 1.

Hypothesized path model.

As implied by our theoretical model, we expect (non)accommodating communication behaviors to be the mechanism that explains why political differences often undermine a sense of family membership. Accommodating communication has the potential to allow individuals to transcend differences (Colaner et al., 2014), minimizing the effect of social identity differences by elevating commonalities embedded in the family identity. Nonaccommodating communication, however, has the potential to amplify and reinforce these differences, which could have detrimental consequences for feelings of relational solidarity. Thus, (non)accommodation serves as the mechanism through which political differences likely relate to family membership. Thus, we pose the following mediation hypothesis:

H4: Familial political agreement and political animus will be indirectly associated with shared family identity through perceived accommodating political communication.

Method

Procedure

To test the hypotheses, a survey was conducted in a blend of large online opt-in panels after the 2016 U.S. presidential election. Variables for this study were included in the third wave and final wave of a larger 2016 election survey project. Participants were initially contacted in September, re-interviewed in November, and completed the final interview sometime from December 5th to 14th. Qualtrics was contracted to recruit participants from over 20 actively managed market research panels. In the first wave of interviews, the sample was quota-stratified to match the most recent U.S. Census data on gender, race, and age. In total, 2014 respondents participated in the initial survey, 833 of whom completed the third interview for this study. There were 452 males (54%) and 381 females (46%). The majority of the participants identified as White (n = 603; 72%), followed by African-American (n = 97; 12%), Hispanic/Latinx (n = 78; 9%), Asian (n = 39; 5%), Native American (n = 5; 0.6%), and Pacific Islander (n = 1; 0.4%). The participants’ average age was 51.19 (SD = 15.55). More respondents identified as Democrats (n = 382, 48%) than Republicans (n = 287, 36%) or independents (n = 133, 17%). Eighty-five percent of our sample told us that they voted (49% Clinton, 45% Trump). Attrition from the first wave of data collected resulted in a sample that was somewhat more male (up from 49%), older (up from an average age of 46.61), and whiter (up from 69%) compared to the quota targets.

Measures

Participants were first asked to think of a member of their family who has political views that are the most different from their own. They were then asked to indicate of whom they were thinking. This response was used to auto-fill the language in subsequent questions. In other words, the survey was programmed to substitute “mother” or “son” or “cousin” (etc.), depending on their response, for [family member] in each of the question wordings that follow. A majority of the sample (59%) indicated the greatest political difference with an extended member of the family (e.g., in-law, grandparent, aunt/uncle, cousin, etc.). The remaining two-fifths of the sample indicated that they primarily disagreed with either a parent (15%), child (8%), or sibling (17%). Unless otherwise specified, all items were measured on a 5-point agreement scale ranging from 1 = Strongly disagree to 5 = Strongly agree.

Familial political difference was measured with a scale adapted from Colaner et al. (2014). The original scale measured religious differences. Given the focus of the current project, wording was adapted to refect political rather than religious differences. The scale contained the following 5 items: (a) I feel like my political ideas are very similar to my [family member]’s political ideas (reversed), (b) My political beliefs are not at all like my [family member]’s beliefs (c) I basically believe the same things that my [family member] does (reversed), (d) I really identify with my [family member]’s political affiliation (reversed), and (e) I do not consider myself to share any political beliefs with my [family member] (M = 3.58, SD = 1.06, α = 0.89).

Political animus was measured using the feeling thermometer ratings commonly used in political communication research on polarization (e.g., Garrett et al., 2014; Iyengar et al., 2012; Warner, 2017). Following the recommendations of Druckman and Levendusky (2019), we asked participants about members (supporters) of political parties rather than about the parties themselves. Given the proximity to the 2016 election and evidence that the election was a point of contention in families (Chen & Rohla, 2018; Tavernise & Seelye, 2016), we asked about individual supporters of the two candidates for U.S. president. Respondents were asked to rate the following three groups on a 101-point feeling thermometer in which 0 indicated least favorable (coldest) evaluation and 100 indicated most favorable (warmest) evaluation: [Trump/Clinton] supporters, people who attended [Trump/Clinton] rallies, and people who voted for [Trump/Clinton]. People were only asked about the candidate representing the political outgroup such that Clinton voters were asked about Trump and vice versa. We elected to focus on outgroup evaluations rather than the difference between ingroup and outgroup evaluations because, first, it is declining outgroup evaluations that explain increased polarization (Iyengar et al., 2012) and, second, the effect of political disagreement on familial connectedness is expected to derive from a family member’s support for an undesirable outgroup.

People who did not report voting for either Clinton or Trump were provided a follow-up question which asked who they would have voted for if forced to choose. In general, people felt quite cold toward supporters of the out-group candidate (M = 22.7, SD = 24.02). Because affective polarization assumes more hostility (i.e., colder feelings) toward the outgroup, this variable was inverted by multiplying the feeling thermometer score by −1.

Communication accommodation was measured with an adapted version of the scale developed and validated by Colaner et al. (2014), which measures two accommodating behaviors and two nonaccommodating behaviors. The original scale assessed communication with regard to religious differences. Question items were adapted to reflect political difference. Otherwise, item wording remained consistent with previous research to maintain the established validity of these measures. Measures consisted of supportive communication, respecting divergent values, inappropriate self-disclosure, and emphasizing divergent value. Participants were asked to indicate the degree to which they perceived their family member engaged in the following (non)accommodative communication behaviors.

Supportive communication was measured by asking participants to respond to the following questions: (a) My [family member] lets me know that they support my right to choose my own political candidate, (b) My [family member] helps me think through my political choices without pressuring me to conform to their political party, and (c) It is difficult to talk to my [family member] about my political beliefs because they think my beliefs are wrong (reversed) (M = 2.99, SD = 0.99, α = 0.81).

Respecting divergent values was made up of 3 items: (a) My [family member] is respectful of my political opinion in our conversations, (b) In our interactions, my [family member] takes my political views and opinions into account, and (c) My [family member] is generally respectful of my political beliefs when we talk about our opinions (M = 3.23, SD = 1.11, α = 0.92).

Inappropriate self-disclosure was measured with the following 3 items: (a) My [family member] tells me too much about their political ideas, (b) I wish my [family member] would not talk with me about their political ideas as much as they do, (c) I feel uncomfortable sometimes with the amount of information my [family member] gives me about their political ideas. (M = 2.96, SD = 1.08, α = 0.85).

Emphasizing divergent values was measured with the following 3 items: (a) My [family member] often brings up their political views with me even though they know I do not agree with them, (b) I feel as though my [family member] tries to convince me that my personal beliefs are wrong, (c) My [family member] expresses disapproval over my political choices (M = 2.98, SD = 1.10, α = 0.84).

Shared family identity was measured with the 6-item scale developed and validated by Soliz and Harwood (2006). Participants responded to the following 6 items: (a) I am proud to be in the same family as my [family member], (b) My shared family membership with my [family member] is not that important to me (reversed). (c) Above all else, I think of my [family member] as a member of my family, (d) My [family member] is an important part of my family, (e) I feel as if we are members of one family, and (f) I feel as if we are members of separate groups (reversed) (M = 3.80, SD = 1.01, α = 0.91).

Results

The hypothesized model was tested via structural equation modeling (SEM) using lavaan which is an SEM package developed for r, an open source statistics and programing software (Rosseel, 2012). Variables measured by more than three indicators were parceled to minimize correlated error terms and preserve just identification (see Little et al., 2013).

As implied by the hypothesized model depicted in Figure 1, the dependent variable (shared family identity) was regressed on all four mediators (perceptions of supportive communication, respecting of divergent values, inappropriate self-disclosure, and emphasis of divergent values) and both independent variables (affective polarization and perceived political difference from the family member). All four mediators were regressed on both independent variables. The initial theoretical model was tested using a robust Maximum Likelihood estimator (MLR). Model fit was acceptable, χ²(168) = 661.84, p < .001, RMSEA = .065 (.060–.070), CFI = .954, NNFI/TLI = .943, SRMR = .059. Mediation was assessed following the bootstrapping method outlined by Preacher and Hayes (2008) using ML estimation. The results are depicted in Figure 2.

Figure 2.

Observed relationships between theoretical variables. Note: Item loadings, residual variances, and latent covariances were estimated but are omitted from the model to improve the interpretability of the figure. All variables with more than 3 items were parceled to preserve just identification. Model fit was appropriate, χ²(168) = 661.84, p < .001, RMSEA = .065 (.060–.070), CFI = .954, NNFI/TLI = .943, SRMR = .059.

The first hypothesis predicted that more perceived political disagreement with a family member would be associated with less perceived accommodating and more perceived nonaccommodating communication. Though no statistical controls were included in the final model, robustness checks were performed controlling for partisanship (given the likelihood that winning vs. losing the election might have different effects on family dynamics) and closeness of family member (given that it may be easier to distance from an extended member of the family than to a parent or child). The inclusion of these variables did not alter any of the relationships reported below.

Though all participants were asked to think about the family member with political views most divergent from their own, we still expected individuals who reported smaller overall differences to also perceive more accommodating communication from their family member. As illustrated in Table 1, people who reported a greater sense of political difference with their family members also reported perceiving less supportive communication, less respect of divergent values, more inappropriate self-disclosure, and more communicative emphasis on divergent values. Furthermore, the largest effects were on perceived accommodating communication (supportive communication, β = −.636, and respecting divergent values, β = −.580), which were double the size (in absolute terms) of the effects on perceived nonaccommodating communication (inappropriate self-disclosure and emphasizing divergent values). For every one additional unit in political disagreement reported, there was a reduction of almost two-thirds of a standard deviation in accommodating communication.

Table 1.

Regression estimates from structural equation model.

	B	S.E.	β	p
Supportive Communication
Political Difference	−.833	.069	−.636	<.001
Affective Polarization	−.065	.042	−.050	.118
Respecting Divergent Values
Political Difference	−.723	.053	−.580	<.001
Affective Polarization	−.102	.039	−.082	.009
Inappropriate Self-Disclosure
Political Difference	.234	.05	.228	<.001
Affective Polarization	.038	.04	.037	.345
Emphasizing Divergent Values
Political Difference	.327	.052	.311	<.001
Affective Polarization	−.002	.041	−.002	.961
Shared Family Identity
Political Difference	.173	.06	.139	.004
Affective Polarization	−.170	.037	−.136	<.001
Supportive Communication	−.095	.172	−.100	.581
Respecting Divergent Values	.641	.173	.640	<.001
Inappropriate Self-Disclosure	−.239	.089	−.197	.007
Emphasizing Divergent Values	.083	.085	.070	.331

Note: All estimates generated using a robust Maximum Likelihood estimation procedure. Model fit was appropriate, χ²(168) = 661.84, p < .001, RMSEA = .065 (.060–.070), CFI = .954, NNFI/TLI = .943, SRMR = .059.

The second hypothesis predicted that people with more negative feelings toward supporters of the political outgroup would perceive less accommodating communication from their family member. Support for this hypothesis was limited. Political animus had no effect on perceived supportive communication, inappropriate self-disclosure, or emphasizing divergent values, though it was associated with perceptions of respecting divergent values. However, the effect was relatively modest compared to the influence of perceived similarity (β = −.082).

The third hypothesis predicted that perceived accommodating communication would be associated with more shared family identity. Support for this hypothesis was mixed. There was no effect of perceived supportive communication or emphasizing divergent values on shared family identity. There was, however, a sizable effect of perceptions of respecting divergent values (β = .640) and a modest effect of perceived inappropriate self-disclosures (β = −.197) on shared family identity. The nature of the effect was such that people were likely to report higher shared family identity with family members who they perceived to communicate in ways that emphasized respect for divergent values. Similarly, when participants perceived that their family member frequently over-discloses about their political beliefs, lower shared family identify was reported.

Though no direct effects of perceived difference and political animus on shared family identity were hypothesized, each had a residual direct effect. People who expressed more negative views of the outgroup also reported less sense of shared family identity (β = −.136). Given the relatively few effects of political animus on perceived communication accommodation, this suggests that affective polarization did undermine shared family identity though more as a result of negative attitudes than perceived non-accommodative communication behaviors. Strangely, people who perceived more political difference also reported more shared family identity (β = .139), despite the fact that the bivariate correlation between the two variables was negative (r = −.221, p < .001). This suppression effect suggests that a greater sense of perceived political difference only reduces shared family identity if it is associated with less perceived accommodating communication.

To further interrogate the suppression effect on the association between political difference and shared family identity, we fit four post-hoc simple mediation models. Our objective was to identify which (non)accommodative communication behavior explained the suppression effect. The results illustrated that the residual effect of political difference on shared family only became positive after accounting for the negative relationship between difference and either accommodative strategy (supportive communication and respecting divergent values). In other words, though larger political disagreements were associated with less accommodating communication, political differences were associated with more shared family identity in families where accommodation strategies were present. The suppression effect was not evident in either model fit for non-accommodative communication (i.e., the residual effect of political difference remained negative).

The final hypothesis predicted that the effects of political difference and affective polarization on shared family identity would be mediated by perceived communication accommodation. Political difference was indirectly associated with lower shared family identity through both less perceived respect for divergent values, 95% CI [−.234, −.801], and more perceived inappropriate self-disclosure, 95% CI [.15, .01]. Affective polarization was indirectly associated with less shared family identity through less respect for divergent values, 95% CI [.014, .112]. These findings are consistent with the hypotheses. Because perceived supportive communication and emphasizing divergent values were not associated with shared family identity they were not tested as mediators. Similarly, because affective polarization was not associated with perceived inappropriate self-disclosure, this indirect relationship was not tested. All bivariate associations between theoretical variables are displayed in Table 2.

Table 2.

Correlation matrix.

	1	2	3	4	5	6	7	8
1. Political Difference	1.00
2. Partisan Animus	0.11***	1.00
3. Supportive Communication	−0.58***	−0.12**	1.00
4. Respecting Divergent Values	−0.55***	−0.15***	0.81***	1.00
5. Inappropriate Self-Disclosure	0.25***	0.06 ^#	−0.42***	−0.43***	1.00
6. Emphasizing Divergent Values	0.31***	0.04	−0.40***	−0.45***	0.65***	1.00
7. Shared Family Identity	−0.17***	−0.18***	0.40***	0.48***	−0.40***	−0.35***	1.00
8. Partisanship	0.04	0.12***	0.05	0.07^#	0.07*	0.07*	0.07^#	1

Note: ***p < .001; **p < .01; *p < .01; ^#p < .1.

Discussion

Results from this study illustrate the challenges posed by political differences in the family. People perceived less accommodating communication from family members with whom they reported the greatest political disagreement. They also perceived more nonaccommodating communication from these family members. The result was less shared family identity, a finding that affirms the view that political differences can strain family relationships.

Our findings also suggest that one accommodation strategy is especially important when navigating political differences in the family. Those who reported that their family member expressed respect for their divergent political values reported much more (two-thirds of a standard deviation) shared identity with this family member. Furthermore, this effect was 200% larger than the (negative) effect of perceived inappropriate self-disclosure, which was the only other communication accommodation strategy to significantly influence shared family identity. Thus, our findings demonstrate the significance of political differences for family relationships, as well as the value of respecting divergent political values when communicating about these differences. That these relationships were observed in the political climate of December 2016, after the unexpected and contentious election of President Trump, illustrates that communicating with respect for divergent values can preserve important relationships even in the most challenging of circumstances.

These findings advance family communication and CAT research by outlining conditions in which accommodative communication becomes more challenging in family relationships.

Research has already demonstrated the implications of (non)accommodative communication on family relationships (e.g., Colaner et al., 2014; Harwood, 2000; Rittenour & Soliz, 2009; Soliz & Harwood, 2006; Soliz et al., 2009), with accommodation relating to positive relational outcomes and nonaccommodation relating to relationship difficulty. We extend this research by examining two factors that make these communication behaviors more difficult: increased identity differences and affective polarization. Because family communication research has seldom focused on motivations for (non)accommodative communication, the identification of these two antecedents of (non)accommodative communication behaviors represent an important contribution to CAT.

Findings were stronger for accommodative (relative to non-accommodative) communication. Communication marked by perceived support and respect were less likely when families experienced greater political identity differences. Given the highly personal nature of political identification, it may be difficult to bracket these attitudes when engaging with a family member’s opposing ideas. Political differences were also associated with nonaccommodative communication, though these effects were smaller. Namely, participants reported more perceived inappropriate self-disclosure and emphasizing divergent values when they perceived greater political differences with their family member. Nonaccommodative political communication serves to advance the position of the speaker rather than the conversational partner, in this case aiming to discredit another’s beliefs or convince a partner to change their opinions. Individuals may perceive that their family member feels more loyalty to their political group than their relationship with their family member. This sentiment can prompt feelings of detachment, as participants pull back from their shared family membership in response to communication that centers political difference over family identification.

The current study also introduces a second factor that we theorized would influence (non)accommodative communication: affective polarization. Specifically, we tested the association between partisan animus and (non)accommodative communication. To date, polarization research has largely focused on public political discourse, such as with politicians, news media, or online political interactions (e.g., Coe et al., 2014; Frimer & Skitka, 2018; Muddiman & Stroud, 2017). However, scholars have also linked polarization to problematic interpersonal behaviors such as workplace discrimination (Gift & Gift, 2015; Iyengar & Westwood, 2015). We found only a minimal influence of outgroup animus on communication behaviors. People with colder feelings toward the political outgroup reported less perceived respect of divergent values and thus less shared family identity, but the effects were modest and the other three facets of CAT were unaffected by partisan animus.

The modest effects of partisan animus (relative to political difference) challenge some of our theoretical assumptions. Work from the intergroup contact tradition (Allport, 1954; Pettigrew & Tropp, 2006; Warner & Villamil, 2017) suggests that outgroup attitudes should influence interactions with members of the outgroup, yet we found that it was the interpersonal variable (amount of political disagreement) that dictated the levels of communicated (non)accommodation. It may be the case that, even among people with the strongest dislike for members of the political outgroup, the way their outgroup family member talks about politics has more influence on the health of this relationship than the political disagreements themselves. In other words, political disagreement need not undermine familial bonds so long as people adopt accommodative strategies when discussing politics.

This interpretation is bolstered by the suppression effect regarding the association of family political difference and shared family identity. As noted above, the bivariate association between political disagreement and SFI was negative—more disagreement corresponded with weaker familial ties. However, this appears to be entirely attributable to non-accommodative communication practices. Once the role of communication accommodation was accounted for, the relationship between political difference and SFI became positive. Families that respectfully navigate their differences actually see their relationships strengthened—perhaps because due to a family culture that invites members to productively navigate identity differences (cf. Horstman et al., 2018; Ledbetter, 2015).

Our findings also illustrate that family relationships can be a source of personal growth and transformation as individuals learn to accommodate divergent identities for the sake of collective ingroup identification. Inappropriate self-disclosure, for example, illustrates the potential for growth. The amount of disclosure that a person deems to be appropriate is their own subjective judgment (Soliz & Colaner, 2017). It is possible that there is no amount of political disclosure that is deemed appropriate in some families. Family members must communicate to establish a mutually agreeable level of disclosure. If an individual perceives that their family member is excessively sharing divergent political views, there is an opportunity to respectfully express a desire to limit such conversation. Establishing appropriate boundaries for political discussions across partisan lines can thus increase family solidarity. This calibration of communication accommodation can promote civility and connection in the face of social difference.

Limitations

As with any cross-sectional study, the relationships observed here cannot confirm the causal relationships presumed by our theoretical hypotheses. It may be the case, for example, that a stronger sense of shared family identity fosters greater correspondence in political beliefs within the family. Indeed, Ledbetter (2015) found that the level of conformity in family communication patterns can influence political ideology. Given recent advances in the theorizing of family communication patterns, it is worth considering how political disagreement relates to various facets in the expanded conformity orientation scale (Horstman et al., 2018). Similarly, though we expected affective polarization to result in less accommodating forms of family communication, it is also reasonable to assume that the level of communicated disagreement in a family can serve as either an antecedent to affective polarization if family members reinforce one-anothers’ views (e.g., Huckfeldt et al., 2004; Mutz, 2006) or a remedy for affective polarization if the family unit serves as a space for respectful contact with members of the political outgroup (i.e., Allport, 1954; Pettigrew & Tropp, 2006; Warner & Villamil, 2017). Importantly, the current study examined shared family identity with one person, specifically the person with which they feel the greatest level of political difference. It is possible that the participant has a higher level of identification with the family as a whole than with the selected family member. Such would be the case if the selected family member was estranged from the family. Future research can fill in this gap by comparing shared family with one individual with family identification with the family group as a whole.

Conclusion

Though we hope that these results will make enduring contributions to the study of family communication, we think they are especially timely given the present political moment. The 2020 presidential election has laid bare how partisanship can guide thinking on matters of science, health, racial justice, environmental crisis, and a variety of other deeply held values tied to our identities. Families across the nation have been collectively navigating a global pandemic and a crisis of racial injustice stacked atop what already promised to be perhaps the most contentious election of a generation. We hope that our findings can speak to this moment. People will naturally feel compelled to weigh in on the essential issues of the day and may believe that reaching out to certain members of their family is a natural starting point. Our paper does not warn against these conversations. Instead, we urge that accommodating communication strategies accompany these difficult discussions. These strategies are best positioned to preserve (and may even improve) important relationships with cross-partisan family members.

Footnotes

Authors’ note

A version of this paper was presented at the 2020 National Communication Association convention.

Declaration of conflicting interests

The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: Dr. Colleen Warner Colaner is a family educator in the community.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded by a grant from the University of Missouri’s Reynold’s Journalism Institute.

ORCID iD

Colleen Warner Colaner

Open research statement

As part of IARR’s encouragement of open research practices, the author(s) have provided the following information: This research was not pre-registered. The data used in the research are not available. The materials used in the research are available. The materials can be obtained by emailing: warnerbe@missouri.edu.

References

Afifi

T. D.

Zamanzadeh

Harrison

Torrez

D. P.

(2020). Explaining the impact of differences in voting patterns on resilience and relational load in romantic relationships during the transition to the Trump presidency. Journal of Social and Personal Relationships, 37, 3–26. https://doi.org/10.1177/0265407519846566

Allport

G. W.

(1954). The nature of prejudice. Perseus Books.

Arnett

J. J.

Jensen

L. A.

(2002). A congregation of one: Individualized religious beliefs among emerging adults. Journal of Adolescent Research, 17, 451–467. https://doi.org/10.1177/0743558402175002

Brooks

D. J.

Geer

J. G.

(2007). Beyond negativity: The effects of incivility on the electorate. American Journal of Political Science, 51, 1–16. https://doi.org/10.1111/j.1540-5907.2007.00233.x

Brown

Levinson

S. C.

(1987). Politeness: Some universals in language usage. Cambridge University Press.

Burleson

B. R.

MacGeorge

E. L.

(2002). Supportive communication. In Knapp

M. L.

Daly

J. A.

(Eds.), Handbook of interpersonal communication (pp. 374–424). SAGE.

Chen

M. K.

Rohla

(2018). The effect of partisanship and political advertising on close family ties. Science, 360, 1020–1024. https://doi.org/10.1126/science.aaq1433

Coe

Kenski

Rains

S. A.

(2014). Online and uncivil? Patterns and determinants of incivility in newspaper website comments. Journal of Communication, 64, 658–679. https://doi.org/10.1111/jcom.12104

Colaner

C. W.

Horstman

H. K.

Rittenour

C. E.

(2018). Negotiating adoptive and birth shared family identity: A social identity complexity approach. Western Journal of Communication, 82, 393–415. https://doi.org/10.1080/10570314.2017.1384564

10.

Colaner

C. W.

Soliz

(2020). Preface. In Soliz

Colaner

C. W.

(Eds.), Navigating relationships in the modern family: Communication, identity, and difference (pp. ix–xx). Peter Lang.

11.

Colaner

C. W.

Soliz

Nelson

L. R.

(2014). Communicatively managing religious identity difference in parent-child relationships: The role of accommodative and nonaccommodative communication. Journal of Family Communication, 14, 310–327. https://doi.org/10.1080/15267431.2014.945700

12.

Cowan

S. K.

Baldassarri

(2018). “It could turn ugly”: Selective disclosure of attitudes in political discussion networks. Social Networks, 52, 1–17. https://doi.org/10.1016/j.socnet.2017.04.002

13.

Dalton

R. J.

(1980). Reassessing parental socialization: Indicator unreliability versus generational transfer. American Political Science Review, 74, 421–431. https://doi.org/10.2307/1960637

14.

De Cremer

Tyler

T. R.

(2005). A matter of intragroup status: The importance of respect for the viability of groups. In Thomas-Hunt

(Ed.), Status and groups (Vol. 7, pp. 1–23). Emerald Group Publishing. https://doi.org/10.1016/S1534-0856(05)07001-5

15.

Druckman

J. N.

Levendusky

M. S.

(2019). What do we measure when we measure affective polarization? Public Opinion Quarterly, 81, 114–122. https://doi.org/10.1093/poq/nfz003

16.

Frimer

J. A.

Skitka

L. J.

(2018). The Montagu principle: Incivility decreases politicians’ public approval, even with their political base. Journal of Personality and Social Psychology, 115, 854–866. https://doi.org/10.1037/pspi0000140

17.

Gallois

Ogay

Giles

(2005). Communication accommodation theory. In Gudykunst

W. B.

(Ed.), Theorizing about intercultural communication (pp. 121–148). SAGE.

18.

Galvin

(2013). The family of the future: What do we face? In Vangelisti

A. L.

(Ed.), The handbook of family communication (pp. 675–697). Lawrence Erlbaum.

19.

Garrett

R. K.

Gvirsman

S. D.

Johnson

B. K.

Tsfati

Neo

Dal

(2014). Implications of pro- and counterattitudinal information exposure for affective polarization. Human Communication Research, 40, 309–332. https://doi.org/10.1111/hcre.12028

20.

Gerber

A. S.

Huber

G. A.

Doherty

Dowling

C. M.

(2012). Disagreement and the avoidance of political discussion: Aggregate relationships and differences across personality traits. American Journal of Political Science, 56, 849–874. https://doi.org/10.1111/j.1540-5907.2011.00571.x

21.

Gift

(2015). Does politics influence hiring? Evidence from a randomized experiment. Political Behavior, 37, 653–675. https://doi.org/10.1007/s11109-014-9286-0

22.

Giles

Reid

Harwood

(2010). The dynamics of intergroup communication. Peter Lang Publishing.

23.

Harwood

(2000). Communicative predictors of solidarity in the grandparent-grandchild relationship. Journal of Social and Personal Relationships, 17, 743–766. https://doi.org/10.1177/0265407500176003

24.

Harwood

(2006). Communication as social identity. In Gregory

G. J.

Shepherd

John

J. St.

Striphas

(Eds.), Communication as…: Perspectives on Theory (pp. 84–91). SAGE.

25.

Harwood

Soliz

Lin

M. C.

(2006). Communication accommodation theory: An intergroup approach to family relationships. In Braithwaite

D. O.

Baxter

L. A.

(Eds.), Engaging theories in family communication: Multiple perspectives (pp. 19–34). SAGE.

26.

Horstman

Schrodt

Warner

B. E.

Hays

Maliski

Koerner

Colaner

C. W.

(2018). Expanding the conceptual and empirical boundaries of family communication patterns: The development and validation of an expanded conformity orientation scale. Communication Monographs, 85, 157–180. https://doi.org/10.1080/03637751.2018.1428354

27.

Huber

G. A.

Malhotra

(2017). Political homophily in social relationships: Evidence from online dating behavior. The Journal of Politics, 79, 269–283. https://doi.org/10.1086/687533

28.

Huckfeldt

(2007). Unanimity, discord, and the communication of public opinion. American Journal of Political Science, 51, 978–995. https://doi.org/10.1111/j.1540-5907.2007.00292.x

29.

Huckfeldt

Johnson

P. E.

Johnson

P. E.

Sprague

(2004). Political disagreement: The survival of diverse opinions within communication networks. Cambridge University Press.

30.

Iyengar

Konitzer

Tedin

(2018). The home as a political fortress: Family agreement in an era of polarization. The Journal of Politics, 80, 1326–1338. https://doi.org/10.1086/698929

31.

Iyengar

Sood

Lelkes

(2012). Affect, not ideology: A social identity perspective on polarization. Public Opinion Quarterly, 76, 405–431. https://doi.org/10.1093/poq/nfs038

32.

Iyengar

Westwood

S. J.

(2015). Fear and loathing across party lines: New evidence on group polarization. American Journal of Political Science, 59, 690–707. https://doi.org/10.1111/ajps.12152

33.

Jennings

M. K.

Niemi

R. G.

(1968). The transmission of political values from parent to child. American political science review, 62, 169–184. https://doi.org/10.1017/S0003055400115709

34.

Ledbetter

A. M.

(2015). Political philosophy as a (partial) mediator of the association between family communication patterns and perception of candidate credibility in the 2012 U.S. presidential election. Journal of Family Communication, 15, 214–231. https://doi.org/10.1080/15267431.2015.1043432

35.

Little

T. D.

Rhemtulla

Gibson

Schoemann

A. M.

(2013). Why the items versus parcels controversy needn’t be one. Psychological Methods, 18, 285–300. https://doi.org/10.1037/a0033266

36.

Mason

(2018). Uncivil agreement: How politics became our identity. University of Chicago Press.

37.

McDevitt

Chaffee

(2002). From top-down to trickle-up influence: Revisiting assumptions about the family in political socialization. Political Communication, 19, 281–301. https://doi.org/10.1080/01957470290055501

38.

Motyl

Iyer

Oishi

Trawalter

Nosek

B. A.

(2014). How ideological migration geographically segregates groups. Journal of Experimental Social Psychology, 51, 1–14. https://doi.org/10.1016/j.jesp.2013.10.010

39.

Muddiman

(2017). Personal and public levels of political incivility. International Journal of Communication, 11, 3182–3202. https://ijoc.org/index.php/ijoc/article/view/6137/2106

40.

Muddiman

Stroud

N. J.

(2017). News values, cognitive biases, and partisan incivility in comment sections. Journal of Communication, 67, 586–609. https://doi.org/10.1111/jcom.12312

41.

Mutz

D. C.

(2006). Hearing the other side: Deliberative versus participatory democracy. Cambridge University Press.

42.

Mutz

D. C.

(2015). In-your-face politics: The consequences of uncivil media. Princeton University Press.

43.

Oliphant

J. B.

(2018, November 20). Most say their family is OK with discussing politics—but it helps if they family agrees. Pew Research Center. https://www.pewresearch.org/fact-tank/2018/11/20/most-say-their-family-is-ok-with-discussing-politics-but-it-helps-if-the-family-agrees/

44.

Pettigrew

T. F.

Tropp

L. R.

(2006). A meta-analytic test of intergroup contact theory. Journal of Personality and Social Psychology, 90, 751. https://doi.org/10.1037/0022-3514.90.5.751

45.

Pew Research Center (2018, March 1). The generation gap in American politics. https://www.people-press.org/2018/03/01/the-generation-gap-in-american-politics/

46.

Preacher

K. J.

Hayes

A. F.

(2008). Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behavior Research Methods, 40, 879–891. https://doi.org/10.3758/BRM.40.3.879

47.

Rittenour

C. E.

(2020). Family socialization of “otherness.” In Soliz

Colaner

C. W.

(Eds.), Navigating relationships in the modern family: Communication, identity, and difference (pp. 227–244). Peter Lang.

48.

Rittenour

C. E.

Soliz

(2009). Communicative and relational dimensions of shared family identity and relational intentions in mother-in-law/daughter-in-law relationships: Developing a conceptual model for mother-in-law/daughter-in-law research. Western Journal of Communication, 73, 67–90. https://doi.org/10.1080/10570310802636334

49.

Rosseel

(2012). Lavaan: An R package for structural equation modeling. Journal of Statistical Software, 48, 1–36. https://www.jstatsoft.org/v48/i02/

50.

Soliz

(2007). Communicative predictors of a shared family identity: Comparison of grandchildren’s perceptions of family-of-origin grandparents and stepgrandparents. Journal of Family Communication, 7, 177–190. https://doi.org/10.1080/15267430701221636

51.

Soliz

Colaner

C. W.

(2017). Communication accommodation theory and communication theory of identity: Theories of communication and identity. In Braithwaite

Suter

Floyd

(Eds.), Engaging theories in family communication: Multiple perspectives (2nd ed., pp. 75–86). Routledge.

52.

Soliz

Harwood

(2006). Shared family identity, age salience, and intergroup contact: Investigation of the grandparent–grandchild relationship. Communication Monographs, 73, 87–107. https://doi.org/10.1080/03637750500534388

53.

Soliz

Phillips

K. E.

(2018). Toward a more expansive understanding of family communication: Considerations for inclusion of ethnic-racial and global diversity. Journal of Family Communication, 18, 5–12. https://doi.org/10.1080/15267431.2017.1399890

54.

Soliz

Ribarsky

Harrigan

M. M.

Tye-Williams

(2010). Perceptions of communication with gay and lesbian family members: Predictors of relational satisfaction and implications for outgroup attitudes. Communication Quarterly, 58, 77–95. https://doi.org/10.1080/01463370903538622

55.

Soliz

Rittenour

C. E.

(2012). Family as an intergroup area. The Handbook of Intergroup Communication, 63, 331–334.

56.

Soliz

Thorson

A. R.

Rittenour

C. E.

(2009). Communicative correlates of satisfaction, family identity, and group salience in multiracial/ethnic families. Journal of Marriage and Family, 71, 819–832. https://doi.org/10.1111/j.1741-37372009.00637.x

57.

Tajfel

Turner

J. C.

(1979). An integrative theory of intergroup conflict. In Austin

W. C.

Worchel

(Eds.), The social psychology of intergroup relations (pp. 205–255). Cambridge University Press.

58.

Tavernise

Seelye

K. Q.

(2016, November 15). Political divide splits relationships—and Thanksgiving, too. New York Times. https://www.nytimes.com/2016/11/16/us/political-divide-splits-relationships-and-thanksgiving-too.html

59.

Turner

L. H.

West

(2018). Investigating family voices from the margins. Journal of Family Communication, 18, 85–91. https://doi.org/10.1080/15267431.2018.1435548

60.

Warner

B. R.

(2017). Modeling partisan media effects in the 2014 U.S. midterm elections. Journalism & Mass Communication Quarterly, 11. https://doi.org/10.1177/1077699017712991

61.

Warner

B. R.

Colaner

C. W.

(2016). Talking politics at the dinner table: The effects of family communication styles on young citizens’ normative political attitudes. In McKinney

M. S.

Thorson

Shah

(Eds.), Political socialization of young citizens (pp. 195–211). Peter Lang.

62.

Warner

B. R.

Villamil

(2017). A test of imagined contact as a means to improve cross-partisan feelings and reduce attribution of malevolence and acceptance of political violence. Communication Monographs, 84, 447–465. https://doi.org/10.1080/03637751.2017.1336779

63.

Zuckerman

A. S.

Dasovic

Fitzgerald

(2007). Partisan families. Cambridge University Press.