While Democrats are early adopters of progressive views, Republicans adopt the same views at a slower pace, which can be easily (mis)interpreted as a sign of polarization

D. Baldassarri, B. Park “Was there a Culture War? Partisan Polarization and Secular Trends in US Public Opinion”, Journal of Politics, forthcoming. Dec 2019. http://deliabaldassarri.org/research/2018/12/1/h29cyp99ttt2wuru2arp4n2u53vale

ABSTRACT: According to many scholars of public opinion, most of the fast-growing divide between Democrats and Republicans over the last few decades has taken place on moral issues. We find that the process of issue partisanship -- the sorting of political preferences along partisan lines -- properly accounts for public opinion dynamics in the economic and civil rights domains: on these issues Democrats as a whole have become more liberal and Republicans more conservative. However, when it comes to moral issues, the prominent change is a partisan secular trend, in which both Democrats and Republicans are adopting more progressive views on moral issues, although at a different rate. While Democrats are early adopters of progressive views, Republicans adopt the same views at a slower pace. This secular change can be easily (mis)interpreted as a sign of polarization because, at the onset of the process, the gap between party supporters broadens due to faster pace at which Democrats adopt progressive views, and only toward the end, the gap between partisan supporters decreases.

---
Furthermore, opinion change, especially on topics concerning gender roles, sexuality, marijuana legalization, and gay rights, has occurred too quickly to be accounted for by generational replacement or demographic shifts (Loftus 2001; Andersen and Fetner 2008; Fischer and Hout 2006). Instead, the secular trend observed for these issues is consistent with a social diffusion dynamic, which operates within generations. In addition, the lag in the adoption curves between Democrats and Republicans (and its persistence even when controlling for the demographic sorting of the electorate) points to the importance of partisan groups in the diffusion process.
Although we cannot provide empirical evidence of the micro-level dynamic of social influence, here we offer some considerations concerning some of the potential underlying mechanisms. In particular, as already anticipated in our general discussion of partisan secular trends,
media partisanship, the homophily of political discussion networks, and the mechanism of
selective disclosure may contribute to the social diffusion process documented in this paper.
Namely, progressive ‘innovators’, who are likely to be Democrats, will spread their views, first,
among like-minded Democrats, both because their political discussion networks are disproportionally composed by fellow Democrats, and because they selectively disclose their opinions
on salient issues to others who they anticipate will agree with them. However, since discussion networks are far from being perfectly homogeneous (Huckfeldt et al. 1995), progressive
opinions would eventually spread among Republicans as well.
In addition, some specific considerations concerning the structural position of gays and lesbians in social networks may partially explain why Republicans have so speedily embraced
secular views on LGBT issues when their elites chose to do otherwise. As with other minority
groups, knowing gays or lesbians has been shown to increase the support for gay rights among
almost all subgroups defined by socio-demographics, ideology, and partisanship (Lewis 2011).
However, compared to racial and socioeconomic cleavages that heavily segregate the interaction patterns of Americans, recent research suggests that interactions with gays and lesbians are
relatively uniformly distributed across the population (DiPrete et al. 2011).14 This is especially
true for social ties within the family, which tend to be “strong ties” both in the emotional and
structural sense (Granovetter 1973). The implication is that, at least in recent years, both Republicans and Democrats have similar probabilities of knowing someone in their close social
circles who is gay or lesbian. Thus, the social influence process might have operated ‘close
to home’, from within the family outwards. This may explain why Republicans have turned
towards more progressive views so easily on these issues. After all, even if party identification
operates as a perceptual screen of the political world (Campbell et al. 1960), it seems safe to
expect that the coming out of a family member or close friend will be a much more persuasive
message than the partisan cues of the political elite.
In sum, empirical evidence supports the claim that public opinion changes on moral issues
have followed a secular trend in which both Democrats and Republicans have adopted more
progressive opinions, although at a different pace. We speculate about the micro-level processes that bring about such outcomes: the partisan lag in the diffusion curves is likely due to
mechanisms of homophily and selective disclosure in political discussion networks. Whereas
the extraordinary pace of the change on gay rights may be due to the diminished segregation
and increased visibility of openly gays and lesbians in people’s social networks. Taken together,
these considerations point to the different nature of public opinion change in the moral domain,
and the possible primacy of social diffusion processes over more classic, party-driven models
of opinion change.
Why is public opinion in the moral domain evolving differently from other domains? Although we cannot provide a fully satisfactory answer here, we mention a couple of possible reasons. First, not all issues are created equal. While the political debate in Western democracies
is usually organized around economic and eventually, geographic or ethnic cleavages (Lipset
and Rokkan 1967), most positions on moral issues do not logically follow from core political
ideologies. This is certainly true in most Western European countries, where the church-state
separation is often written in the Constitution, and the political debate is mainly organized
around issues of redistribution, welfare state, and taxation. However, even in the US, the logical
link between parties’ core ideologies and their stand on moral issues is weak at most (Converse
1964; Baldassarri and Goldberg 2014): for example, it is quite difficult to reconcile Republicans’ laissez faire economic agenda and their opposition to any form of gun control with their
heavily regulatory stand on reproductive issues. Second, moral issues are often concerned with
whether a certain behavior – e.g., women employment in the workforce, smoking marijuana,
gay raising kids – is considered acceptable. As a long tradition in sociology has demonstrated,
what is considered ‘moral’ often coincides with what is considered normal, or ‘average behavior’ in a society: thus the social norm that most people follow in a given space-time (Durkheim
1906). If this is the case, it is understandable why opinions on this type of issues are more
subjected to bottom-up dynamics of diffusion and social influence.