Public Support for Government Intervention in Health Care in the United States from 1984 to 2016

Abstract

Research on public opinion regarding government’s role in health care has paid little attention to how public opinion has changed among different age groups over time and to how the intersection of age, birth year, political affiliation, and historical time shape public opinion. In this article, the authors ask, Who supports governmental spending on health care, and how has this changed over time? The authors propose a life-course perspective to study political polarization in the health care domain using General Social Survey 1984 to 2016 data. The results indicate that the growing political polarization in support for government intervention in health care across the 32 years studied occurred among middle-aged adults. The findings of this study contribute new understandings of how age and party membership interact in contributing to political polarization regarding government’s role in health care over time.

Keywords

life course political affiliation public opinion government intervention health care

In the United States, government involvement in health care is a focal issue for partisan disagreement. Public opinion of government’s role in the health care system not only reflects the general support for health care related policies, but it can also influence health care policy when it translates into votes for a candidate (Hunter 2008). In this article we assess the impact of the unique political culture in the United States on public opinion of government’s role in health care over time.

In the present study, we use a life-course perspective to analyze how people’s political affiliations shape their beliefs about governmental involvement in health care. In Children of the Great Depression, Elder (1999) conceptualized the life course as “a pattern of socially defined, age-graded events and roles which is subject to historical change in culture and social structure” (p. 302). Elder’s research findings underscore the importance of historical context, specifically that one’s life course is situated in and shaped by the historical times and events one encounters over one’s lifetime, and the importance of timing in individual lives, specifically that the significance of personal and historical events varies depending on an individual’s life stage. Therefore, both historical period and individual aging are key considerations in investigating public opinion about the role of government. On the basis of the framework of the life-course perspective, we ask two research questions: (1) How does the association between political affiliation and support for government intervention in health care vary by age and cohort? and (2) How does the association between political affiliation and support for government intervention in health care vary across time periods?

Many previous studies of the effect of political polarization on the health care system have focused on elites in Congress (Hacker and Pierson 2018; Quadagno 2014). Others have focused specifically on public support for the Clinton and Obama health care plans (Brodie et al. 2019; Kriner and Reeves 2014; McIntyre et al. 2020; Morone 2016; Skocpol 1996). The findings from previous research prompt an investigation of the effects of political polarization not only among elites but also among the public. There is also a need to look beyond a short period of time to consider the longer term changing political climate on public opinion of government’s role in health care.

This research bridges the literatures on (1) changing political attitudes with age and over time in life-course sociology (Alwin, Cohen, and Newcomb 1991; Street and Cossman 2006b), (2) political polarization and support for government spending in political sociology (Johnson and Schwadel 2019b; Pacheco, Haselswerdt, and Michener 2020), and (3) public opinion on health care in medical sociology (Brodie et al. 2019; Corman and Levin 2016; Dalen, Waterbrook, and Alpert 2015) and advances scholarship on how political affiliation contributes to public opinion on government intervention in health care and how this association varies by age and over time. We know there has been a growing political divide in the United States (see, e.g., Anderson, Lytle, and Schwadel 2017; McCright, Xiao, and Dunlap 2014), but life-course theory of social change has the potential to shed light on the age dynamics of the partisan divide. Although there is a rich research literature on age-based interests in social welfare policies (Kitschelt and Rehm 2006; Rehm 2009), we know little about how these age effects interact with party identification. This research has the potential to provide new understandings of how (i.e., among which age groups) the partisan divide over social welfare policies has grown in the recent past.

Background

The History of the Partisan Divide over Government’s Role in Health Care

Partisan conflict appears particularly acute in the health care domain. Looking back across the twentieth century, according to Garfinkel, Rainwater, and Smeeding (2010), there were times the federal government made efforts to develop a universal health care system. For example, after World War II, President Truman proposed a national health insurance program. In the 1960s, during the presidential campaign, John F. Kennedy supported national insurance for the aged, which was also known as Medicare. Congress has continuously worked on health care reform since the 1980s to address increasing costs and the rising numbers of uninsured (Hoffman 2012; Quadagno 2006). Changes in presidential administrations and in political support for government spending on health care provide the historical context for changes in public support for government intervention in health care.

The 1980s, from 1981 to 1989, were the Reagan era, a famously conservative historical period. Cutting spending on Medicaid was one of the first actions initiated by the Republican-dominated Congress in 1981 (Hoffman 2012). The 1990s started with President George H. W. Bush, whose administration proposed several health reform acts that represented bipartisan attempts to expand the coverage of health care in the United States, including the Comprehensive Health Reform Act of 1992, which was followed after Bush left office by HEART, the Health Equity and Access Reform Today Act of 1993. When President Clinton won the presidential election in 1992, the Clinton administration began actively making efforts to overhaul the health care system in the United States through the Clinton health plan. Even though the Clinton administration failed in passing its health plan, the public conversation surrounding health care reform successfully led to expansions in health coverage. Then, in the early 2000s, there was more political attention to limiting government intervention because President George W. Bush opposed a national health care plan. However, the aftermath of the terrorist attacks of 9/11 in 2001 and Hurricane Katrina in 2005 brought the inequalities of the health care system to the public’s attention. Congress passed the Medicare Modernization Act of 2003 to offer a temporary Medicaid expansion to cover survivors of 9/11 but did nothing to help victims of Katrina (Hoffman 2012). In the 2008 presidential campaign, candidates directly discussed health care as a right of citizenship. Following President Obama’s election in 2008, the Obama administration campaigned for the passage of the Affordable Care Act (ACA). When the ACA passed in 2010, it was with only one Republican vote in the House and none in the Senate (Morone 2016). The passage of this legislation and the ongoing partisan debate over the ACA has incited further partisan division, with Democrats more in favor of government intervention in health care and Republicans less in favor of government intervention.

Although each of these periods is an important part of the political landscape influencing popular attitudes toward government spending on health care, it is the Clinton administration that is pointed to as having the most significant impact on the political polarization of the health care debate (Skocpol 1995, 1996). Political advertising in opposition to the Clinton plan in the 1990s led to growing public skepticism surrounding government’s role in health care and even government intervention in citizens’ lives more generally. As Skocpol (1995) explained,

by midsummer 1994 and on through the November election, many middle-class citizens . . . had come to perceive the Clinton plan as a misconceived “big government” effort that might threaten the quality of U.S. health care for people like themselves. (p. 76)

The values gap in the United States between Republicans and Democrats now is wider than age, race, class, or gender divides (Pew Research Center 2012). There is a growing body of research investigating the effect of political polarization on public opinion in America (see Baldassarri and Bearman 2007; Fiorina and Abrams 2008; McCright et al. 2014). Researchers have observed a growing gap between Democrats and Republicans on several social issues. For example, analyzing the 38-year data from the General Social Survey (GSS), McCright et al. (2014) found a modest political divide on environmental issues among the public in the 1970s and 1980s but a widened political divide since the 1990s. Similarly, on the basis of 40 years of data from the GSS, Anderson et al. (2017) found that the gap in support for the death penalty between Republicans and non-Republicans was modest in the 1980s and 1990s but has become increasingly widened in the twenty-first century.

Understanding Political Polarization in Health Care through the Life-Course Perspective

Even though the political climate in the United States is constantly shifting over time, the partisan divergence has been one of the major reasons why the United States is the only country without a universal health care system among the industrialized countries in the world (Olafsdottir and Pescosolido 2010; Quadagno 2006, 2010). Therefore, it is important to investigate the changing effect of political views on attitudes toward government intervention in health care.

Many scholars have pointed out that political views affect public opinion on welfare policy in America (see Garfinkel et al. 2010; Hasenfeld and Rafferty 1989; Kluegel and Miyano 1995; Larsen 2008; Margalit 2013; Quadagno and Pederson 2012). In particular, partisan conflict appears prominently in health care policy (Henderson and Hillygus 2011; Morone 2016; Olafsdottir and Pescosolido 2010; Quadagno 2010). A long-standing framework in studies of political behavior argues that group identities are critically important predictors of individual political attitudes and public policy preferences (e.g., Conover 1984; Gelman, Lee, and Ghitza 2010; Kriner and Reeves 2014). The responsive partisanship framework (Kriner and Reeves 2014) or partisan loyalty (Bartels 2002; Lenz 2009) stands out in the context of health care. According to Kriner and Reeves (2014), the partisan discourse among elites affects the level and stability of support or opposition to health care–related policies and laws among partisan identifiers in the mass public. That is, individual-level change in political attitudes will be accordant with the views of their preferred party advocates. However, questions remain regarding how political attitudes change through individual-level aging processes and through population turnover as the mechanisms of social change.

The sociology of aging and the life course provides an analytical framework for understanding the interplay between human lives and changing social structures. Life-course scholars examine the interdependence between aging over the life course as a social process and societies and groups as stratified by age, with the succession of cohorts as the link connecting the two. Life-course theories of social change emphasize the role of age, period, and cohort effects in motivating social change in beliefs and attitudes over time (Alwin and McCammon 2003).

Age effects¹ are due to the aging process, which proceeds in biological, psychological, and social ways (Glenn 1974). Aging affects individuals’ values and preferences as evidenced in political activities such as voting (Tilley and Evans 2014), opinions of government’s role in society (Meuleman and Chung 2012), and attitudes toward social issues such as the inclusion of historically subordinate groups in society and civil liberties (Danigelis, Hardy, and Cutler 2007). However, different hypotheses exist regarding the association between age and political attitudes. One hypothesis is that preferences regarding social spending and government intervention are caused by age-based self-interest, meaning that individuals’ preferences in the social welfare domain are based on whether social welfare spending is going to benefit them or not. For example, older citizens are more likely to support social policies that increase their own well-being (e.g., Social Security and Medicare) (Kitschelt and Rehm 2006; Rehm 2009).

In addition to age-based self-interest, individuals’ preferences for social spending are also related to their opinions of government’s role in society. Values and ideological beliefs about the government play a central role too. The aging process model proposes that an individual’s position in the life cycle (i.e., age) is associated with conservatism (Glenn 1974). This aging process model posits that people tend to be more conservative when they are older because when they age, they tend to be more invested in the status quo and less likely to adopt liberal or progressive views. Some empirical studies have found that openness to new ideas declines at older ages (Roberts, Walton, and Viechtbauer 2006) and also that older people are more conservative and Republican than younger people (see Ghitza and Gelman 2014; Twenge, Carter, and Campbell 2015). Of course, life-course scholars have noted that what is sometimes identified as an age effect on political attitudes is actually attributable to cohort differences (Alwin, Cohen, and Newcomb 1991; Riley 1987).

When specifically studying political party polarization in attitudes toward government intervention in health care, it is possible that aging may decrease political party–based differences in attitudes toward government spending on health care. Life-course scholars have pointed to old age as the “great leveler” (Ferraro and Kelley-Moore 2003) in that older adults of different social statuses (defined by gender, race, class, etc.) share more similar life experiences and outcomes in later life than earlier in life. This is due to the shared changes associated with aging (e.g., loss of loved ones, declining health, increased reliance on government programs such as Medicare and Social Security) that all individuals experience. Scholars have pointed to how this convergence in life conditions can account for declines in health disparities, for example, in old age (Ferraro and Kelley-Moore 2003). We argue that this logic may also extend to political attitudes with older adults reporting more similar views across party lines on government intervention in health care than younger adults.

Cohort effects are a critical contributor to broad social change. Ryder (1985) noted that each new cohort is a possible vehicle for social change. Alwin and McCammon (2003:34) explored four assumptions of the theory of cohort replacement. Two relevant assumptions are discussed as follows: (1) According to the impressionable youth assumption, youth is an impressionable stage in the life course in which individuals are most open to the socialization influences of the social environment maximally. It suggests that late adolescence and early adulthood are the phases of the life course in which attitudes and values are most likely to be formed and crystallized. (2) The cohort effects assumption suggests that unique cohort experiences are formed by the distinctive influence of historical events and experiences. Thus, successive birth cohorts can be distinguished from each other on the basis of the effects of the historical period in which each cohort reaches young adulthood, leaving a distinctive imprint on their worldview. Shapiro and Young (1989) pointed out that the context of early adulthood is critical in establishing opinions on political issues, including the role of government, government spending, taxes, and other issues. For example, in 1970s Britain, a higher level of collective support for welfare state programs, compared with their American counterparts, existed partially because younger cohorts transitioned to adulthood in Britain during a period when welfare provision was an uncontroversial issue (Shapiro and Young 1989:66). In their study of public opinion on social spending, Street and Cossman (2006a) found that earlier born cohorts were less supportive of social spending orientations than more recently born cohorts. They explained that this may be because earlier cohorts had lower expectations of government support for education, health, and pensions. These expectations were formed before the maturation of the U.S. welfare state (Street and Cossman 2006a:92). A more recent survey conducted by the Pew Research Center (2020) suggested that recent cohorts of Republicans, namely millennial and generation Z Republicans, are more likely to agree that the federal government is doing too little on climate change than previous cohorts of Republicans. Therefore, we argue that changes in attitudes about government intervention in health care can be affected by cohort differences, which can be confounded with age effects.

Summary of Literature, Research Questions, and Hypotheses

Although much of the previous research analyzed either one point in time or the impacts of the political views of elites, little is known about how public opinion of government’s role in health care changes across a longer period of time in history and with age. We argue that support for government intervention in health care is dynamic rather than static. It is affected by the intersection of age, birth year, political affiliation, and historical time. We bridge the life-course sociology literature on the mechanisms of social change, the political sociology literature on the growth in political polarization in the United States, and the medical sociology literature on public attitudes about government intervention in health care in asking how the association between political affiliation and support for government intervention in health care changes with age, across birth cohorts, and across historical periods. By bridging these disparate literatures, we shed new light on how the partisan divide in attitudes about government intervention in health care has emerged over the past three decades. To assess these questions, we have the following expectations to guide our analysis.

The first set of hypotheses relates to age effects and the interactions between age and political affiliations. Hypotheses 1A and 1B make predictions about how respondents’ support for government intervention in health care varies by age. On the basis of the self-interest hypothesis, hypothesis 1A predicts that younger adults are less supportive of government intervention in health care than older adults. In contrast, on the basis of the aging process model, hypothesis 1B predicts that younger adults are more supportive of government intervention in health care than older adults. We examine these two competing hypotheses and test the age main effects on the outcome variable. We then examine age and political party interactions. Hypothesis 1C is that, consistent with the notion of old age as the “great leveler,” differences between Republicans and Democrats in support for government spending on health care will narrow in old age. This is because the self-interest involved in the full funding of programs such as Medicare may trump political divides that differentiate members of different political parties at younger ages.

The second set of hypotheses is about cohort effects. We expect to find differences across birth cohorts in support for government intervention in health care. Following Street and Cossman (2006a), hypothesis 2A is that more recent birth cohorts will be more supportive of government intervention than cohorts born earlier in the twentieth century. For hypotheses 2B and 2C, we propose competing expectations regarding cohort effects on political polarization in attitudes toward government intervention in health care. On the basis of recent empirical data (Pew Research Center 2020), we predict in hypothesis 2B that more recent cohorts of Republicans in particular will show more support for government spending on health care, narrowing the gap between political parties in more recent cohorts. Alternatively, for hypothesis 2C, we predict that it is also possible that more recent cohorts who were born and raised in a more polarized political context may instead show a larger gap between political parties in their support for government spending on health care, compared with cohorts born earlier in the twentieth century.

Third, health care has been one of the recurring topics for presidential debates since the Reagan era, yet it was during the Obama era that an unprecedented expansion of health care coverage was written into the law at the national level. Hypothesis 3A predicts that the overall support for government intervention in the health care system will increase significantly from the Reagan era to the Obama era. Furthermore, changes in support for government intervention in health care between Democrats and Republicans will be responsive to the increasing polarization in the discussions of health care–related issues between the two parties. Hypothesis 3B proposes that the gap in support for government intervention in health care among Democratic and Republicans increased from the Reagan era to the Obama era, with a marked increase during the Clinton era.

Last, we also investigate the interaction of age, period, and political affiliation in our analysis. We do not have a priori expectations about how age patterns in the gap between Democrats and Republicans in support for government intervention will vary across historical periods. Instead, we intend to investigate how differences between political parties in the patterns of change with age in support for government intervention may have contributed to growing polarization over time.

Data and Method

Data

Data come from the 1984 to 2016 U.S. GSS (Smith et al. 2018). Each survey uses multistage stratified probability sampling and includes a nationally representative sample of noninstitutionalized adults ages 18 and older. The GSS has measured the attitudes and beliefs of adults in the United States since 1972, making it ideal for research that focuses on the overall changing trends in public attitudes, beliefs, and well-being at the population level (Marsden, Smith, and Hout 2020). Most of the key variables for our study were included in each GSS survey conducted between 1984 and 2016. Data from 1987 were dropped because of oversampling. Cases that had any missing values for variables in the analysis were dropped from the analysis. Listwise deletion results in a final sample of respondents who have valid data on all of the variables included in this analysis. The total number of respondents available in the final analytical sample is 13,219.² The sample size by year is available in Appendix I.

Dependent Variable

Public support for government intervention in health care is measured by one question on the GSS. The question asks respondents if the government should help pay for medical care.³ The dependent variable is dichotomized into those who felt the government should help compared with all others. Those who think government should not help pay for medical care or who are ambivalent regarding government intervention are the reference category (government should help = 1, otherwise = 0).

Independent Variables

Key independent variables include age, period (year of survey), and political affiliation. Age is measured in years. As prior findings (see Anderson et al. 2017; Yang 2008) have shown curvilinear age effects, age squared is added to the analysis. The 32-year period from 1984 to 2016 is divided into five periods on the basis of presidential administration: Ronald Reagan (1981–1988), George H. W. Bush (1989–1992), Bill Clinton (1993–2000), George W. Bush (2001–2008), and Barack Obama (2009–2016).⁴

Party affiliation is categorized into three groups: Democrat (strong Democrat, not strong Democrat, and independent but near Democrat = 1), independent and other (independent and other party = 2), and Republican (strong Republican, not strong Republican, and independent but near Republican = 3).

Previous studies of public opinion of government’s role in social spending suggest the inclusion of several other control variables. Therefore, in addition to the aforementioned key independent variables, the following control variables are included in this study: sex (female = 1, male = 0), race (nonwhite = 1, white = 0), education (less than high school = 1, high school = 2, some college or college = 3, more than college = 4), employment status (full-time = 1, part-time= 2, other = 3), income (converted to 1986 dollars, adjusted for family size, and then broken into quartiles: first quartile = 1, second quartile = 2, third quartile = 3, fourth quartile = 4), self-rated health (from 1 = excellent to 4 = poor), rural residence (1= yes, 0 = no), and religious attendance (frequency of religious service attendance ranging from 1 = never to 9 = more than once a week). Table 1 contains the descriptive statistics of the variables included in this analysis.

Table 1.

Descriptive Statistics of Study Variables (Total n = 13,219).

	Mean/Percentage	SD	Minimum	Maximum
Government should help pay for medical care
Yes	49.63%
No	50.37%
Political affiliation
Democrat	46.93%
Independent/other	17.02%
Republican	36.14%
Age (years)	45.66	16.88	18.00	89.00
Period
Reagan (1981–1988)	12.99%
Bush (1989–1992)	9.64%
Clinton (1993–2000)	39.92%
Bush (2001–2008)	16.95%
Obama (2009–2016)	20.51%
Employment status
Full-time job	53.47%
Part-time job	10.60%
Not working/retired/other	35.93%
Gender
Male	45.38%
Female	54.62%	.50	.00	1.00
Race
White	80.20%	.40	.00	1.00
Nonwhite	19.80%	.40	.00	1.00
Education
Less than high school	15.98%
High school	52.04%
Some college or college	23.39%
More than college	8.59%
Income(quartiles)	2.51	1.11	1.00	4.00
Self-rated health	1.99	.82	1.00	4.00
Religious attendance	4.67	2.73	1.00	9.00
Living in rural area
Yes	11.26%
No	88.74%

Analytical Strategy

To answer our research questions, we analyze our sample in the following steps. First, as many prior studies have suggested (Anderson et al. 2017; Lu et al. 2021; Schwadel and Garneau 2014; Shi, Lu, and Pickett 2020), temporal change in public opinion may be attributed to age, period, or cohort variation. Therefore, our first step is to examine if there are significant age, period, and cohort effects that contribute to the variation in attitudes toward government intervention in health care with an overall model specification test. If all of the three factors are statistically significant, we will use age-period-cohort (APC) modeling techniques for the following analyses. If only two of the three factors are statistically significant, we will use logistic regressions for the other analysis.

We use the APC-interaction (APC-I) approach developed by Luo and Hodges (2020) to examine overall contributions of the age, period, and cohort factors to the model. Modeling age, period, and cohort effects simultaneously is challenging because of the identification problem resulting from the linear dependency among the three factors (cohort = period – age). Scholars have proposed solutions, but many of these solutions are subject to criticism (see review in Fosse and Winship 2019). The APC-I approach we used in the present analysis treats cohort effects as age-by-period interactions and thus is immune to the identification problem (see review in Luo and Hodges 2020).

Specifically, to determine if there is cohort variation in our data, we analyze a full APC-I model and compare its overall model fit with a reduced model with age and period main effects only (AP model). Results of the model specification test reveal that the full APC-I model is not significantly different from the reduced model with age and period effects only (deviance = 91.426, df = 72, chi-square test p = .061). Therefore, we conclude that there is no significant variance in our data that is uniquely attributed to cohort differences and proceed with a simple model that includes age and period effects only. In addition, to examine the robustness of this conclusion, we have conducted a supplemental analysis with an alternative model specification using hierarchical APC-cohort models (Yang and Land 2008). In the supplemental analysis, we also test if the effects of party affiliation interact with period and cohort-level variance with a hierarchical APC approach. Results are consistent across model specifications and suggest very trivial cohort-level variation (see Appendix II). This refutes hypotheses 2A, 2B, and 2C regarding the effects of cohort, as we find no meaningful variation across cohorts.

Next, we use logistic regression models with age and period factors included (AP model) to examine the effect of party affiliation on public opinion of government intervention in health care and its change across time and ages. We first run a model with the main effects of party affiliation, age, period, and all control variables.

Then we introduce interaction terms into the model: we analyze a model with party affiliation interacted with age and with period, followed by a final model that contains the three-way interaction effect of political affiliation, period, and age. To present our interaction effects, we also analyze the predicted probabilities of the main results and present them with predicted probabilities plots. As suggested by Mustillo, Lizardo, and McVeigh (2018; see also Mize 2019), it is problematic to draw conclusions about the significance of statistical interactions in categorical models such as logit, probit, or Poisson models. Therefore, our interpretations of the results of the interaction effects will focus on the predicted probabilities and marginal effects of different subgroups following the procedures suggested by Mize (2019).

Results

Results of our regression models are presented in Table 2. First, we analyze a full model with all control variables examining the main effects of party affiliation, age, and period (model 1). Compared with respondents who self-identified as Democrats, those who identified as independents and Republicans have 39 percent (e^–.500 – 1) and 63 percent (e^–1.001 – 1) lower probabilities of support for government intervention in health care, respectively.

Table 2.

Logistic Regression Models of Public Support for Government Intervention in Health Care.

	Model 1		Model 2
	Coefficient	SE	Coefficient	SE
Intercept	.592	.167***	−.144	.227
Political affiliation (reference Democrat)
Independent/other	−.500	.052***	−.320	.409
Republican	−1.001	.042***	.978	.327**
Age	.008	.006	.033	.009***
Age squared^a	−.002	.001**	−.004	.001***
Period (reference 1984−1988 [Reagan])
1989–1992 (Bush)	.488	.078***	.395	.111***
1993–2000 (Clinton)	.223	.059***	.331	.081***
2001–2008 (Bush)	.291	.069***	.464	.097***
2009–2016 (Obama)	.088	.067	.456	.092***
Race (reference white)	.183	.049***	.168	.049**
Gender (reference male)	.102	.038**	.102	.038**
Education (reference less than high school)
High school	−.259	.056***	−.269	.056***
College	−.157	.067*	−.167	.067*
More than college	.179	.086*	.155	.087
Workforce status (reference full-time job)
Part-time job	.126	.063*	.127	.063*
Not in the labor force	.108	.048*	.104	.048*
Income (reference first quartile)
Second quartile	−.092	.054	−.098	.054
Third quartile	−.339	.056***	−.342	.056***
Fourth quartile	−.343	.062***	−.349	.062***
Church Religious attendance	−.047	.007***	−.045	.007***
Health status (1–4 = excellent to poor)	.113	.024***	.112	.025***
Rural area	−.092	.059	−.092	.059
Age × political affiliation (reference Democrat)
Age × independent/other			−.001	.017
Age × Republican			−.069	.014***
Age squared × independent/other^a			.000	.002
Age squared × Republican^a			.007	.001***
Period × party identification (reference Democrat)
Independent/other × Bush period			.148	.245
Independent/other × Clinton period			−.049	.181
Independent/other × Bush period			−.115	.200
Independent/other × Obama period			−.266	.194
Republican × Bush period			.074	.165
Republican × Clinton period			−.291	.126*
Republican × Bush period			−.430	.149**
Republican × Obama period			−1.039	.150***
Sample size	13,219		13,219
χ²	1,222.590***		1,330.900***

Coefficients are multiplied by 10.

p < .05. **p < .01. ***p < .001.

The linear component of age is not significant, but the quadratic term for age is statistically significant. Respondents in the oldest age groups are less likely to support government intervention in health care. This supports hypothesis 1B (the aging process model). Model 1 also reveals significant variation across the five study periods. Holding all the individual-level effects constant, respondents are significantly more likely to support government intervention in health care in the 1990s and 2000s compared with in the early 1980s. However, support declines in the 2010s to a level that is not significantly different from the level reported in the early 1980s. This partially confirms our hypothesis 3A (which predicted growing support for government intervention over time). In addition, model 1 also reveals that respondents who are nonwhite, are female, are from low-income families, are working part-time jobs or have no job, have less than a high school education or have a graduate degree, have infrequent religious attendance, have poor health status, and are from metropolitan or urban areas are more likely to support government involvement in health care.

We next include interaction effects between party and age as well as interaction effects between party and period in model 2, holding all the individual characteristics constant. The results reveal significant interactions with age and Republican party identification as well as with period and Republican party identification. Interpretation of interaction effects can be challenging, as our model involves multiple interaction terms for party affiliations. To better demonstrate the interaction effects within and across population subgroups, we estimate predicted probabilities across age, period, and party groups on the basis of model 2 in Table 2 in Figure 1.

Table 3.

Results for How Age and Period Associated with Support for Health Care: Tests of Average Marginal Effects.

	Effect of Age		Effect of Period
Average Margin Effects	Age 40 vs. 20 Years	Age 70 vs. 50 Years	Obama (2009–2016) vs. Reagan (1981–1988)
Democrat	.028 (.018)	−.186*** (.023)	.107*** (.022)
Independent/other	.018 (.031)	−.209*** (.042)	.048 (.043)
Republican	−.114*** (.022)	−.056* (.022)	−.113*** (.023)

Note: Values in parentheses are standard errors.

p < .05. ***p < .001.

Figure 1.

Predicted support for government intervention in health care by party, age, and period.

The top panel of Figure 1 demonstrates the predicted support for health care intervention across the three political affiliation groups from ages 18 to 90 years. The shaded areas of each predicted curve represent the 95 percent confidence intervals of the predicted probabilities. Overlapping shaded areas indicate no significant differences in the predicted probabilities between two parties at a specific age. Overall, older respondents are less likely to support government intervention in health care, and this trend is consistent across parties, but there is some variation across the three age curves.

Democrats (blue solid line) are significantly more likely to support government intervention in health care than the other two parties at all ages. The predicted probabilities of supporting health care are very stable for Democrats at young and middle ages but decline substantially after age 60. Independents and respondents who identified with other parties (green dotted line) also share a very similar shape of the age curve with Democrats, with almost no variation at young and middle ages but a rapid decline after age 60. Republicans (red dashed line), however, have a slightly different age curve than the other two groups. The predicted probabilities of supporting health care among Republicans decline gradually between the early 20s and late 50s, but the decline slows thereafter and remains relatively stable for older Republicans. In addition, there are no significant differences in the predicted probabilities between the youngest and oldest Republicans and independents. The interaction effects between age and party show that the gap in attitudes toward government intervention in health care across parties is larger when respondents are between 25 and 60 years of age than when respondents are in their early 20s or after retirement (older than 60). The results provide qualified support for hypothesis 1C, which predicts that the gap in support for government intervention in health care between Democrats and Republicans is narrower among older adults because old age is the great leveler. Surprisingly, however, a convergence in attitudes between Democrats and Republicans is also present in youth.

The bottom panel of Figure 1 presents the predicted support for government intervention in health care in five presidential administrations across the three parties. Different from the age graph that shows consistent declining age trends across the three parties, period trends of the three parties demonstrate very different patterns. Democrats’ support for health care intervention increases slightly across time, but Republicans’ support for health care intervention declines gradually across time. With opposite trends observed among these two parties, the gap between Democrats and Republicans is much wider in the Obama era (2009–2016) than it was in the Reagan era (1981–1988), consistent with hypothesis 3B. Independent respondents also demonstrate a unique pattern, with increasing support for government intervention from the Reagan era (1981–1988) to the George H. W. Bush era (1989–1992), but the period trend remains stable afterward. It is worth noting that independents are statistically indistinguishable from the other two parties during the Reagan era (1981–1988) and the George H. W. Bush era (1989–1992), but there is no overlap from the Clinton (1993–2000) era onward. In addition, it is during the Clinton era that we see statistically significant differences in the period effects for Democrats and Republicans, with a positive period effect for Democrats and a negative effect for Republicans. This is also consistent with hypothesis 3B.

Next, to further examine period and age effects within the same party group, we report the predicted probabilities at specific ages and periods in Table 3. Holding other variables constant at the mean, there is no significant difference between Democrats at age 40 and Democrats at age 20 in terms of their support of government intervention in health care. Democrats at age 70, however, have significantly lower predicted probabilities (–18.6 percent) of supporting health care than Democrats at age 50. Similarly, independents and others also demonstrate no significant difference in their support for health care at young and middle ages, but the differences become significant when comparing age 70 with age 50 (–20.9 percent). In contrast, Republicans’ support for health care declines significantly with age, but the differences in predicted probabilities are greater when comparing age 40 with age 20 (–11.4 percent) than when comparing age 70 with age 50 (–5.6 percent). All of these age differences reported in Table 3 are consistent with the age curves presented in Figure 1. The last column of Table 3 presents the differences in predicted probabilities between the Obama period and the Reagan period within each party group. Democrats reported significantly greater levels of support for health care after 2009 compared with the early 1980s (10.7 percent), but Republicans reported significantly lower levels of support for health care in the most recent period compared with the early 1980s (–11.3 percent). There is no significant difference between the two periods among independents and others.

Last, we further examine the growing polarization over time by testing the three-way interaction effects of political affiliation, age, and period. The predicted probabilities in support of government intervention in health care across the life course of respondents with different party identifications in four different eras are depicted in Figure 2.⁵ Model statistics with three-way interaction effects are included in Appendix III. Although the age curves of each party demonstrate consistent patterns across the four periods, the gaps between Democrats and Republicans become increasingly wider over time, particularly among the middle-aged groups. During the Reagan and George H. W. Bush administrations (Figure 2a), the difference in support for government intervention in health care between Democrats and Republicans is negligible among younger and older age groups, but there are significant differences found among middle-aged Democrats and middle-aged Republicans.

Figure 2.

Predicted support for government intervention in health care at different ages by party and period.

At age 50, for instance, the predicted support for health care intervention is 15 percent higher for Democrats (48 percent) than for Republicans (33 percent) in the Reagan and George H. W. Bush era. This Democrat-Republican gap at age 50 increased to 25 percent (54 percent for Democrats and 29 percent for Republicans) in the Clinton era and 40 percent (59 percent for Democrats and 19 percent for Republicans) during the Obama administration. At age 50, for instance, the predicted support for health care intervention is 15 percent higher for Democrats (48 percent) than for Republicans (33 percent) in the Reagan and George H.W. Bush era. This Democrat-Republican gap at age 50 increased to 25 percent (54 percent for Democrats and 29 percent for Republicans) in the Clinton era and 40 percent (59 percent for Democrats and 19 percent for Republicans) during the Obama administration. For the younger and older age groups (i.e., under age 20 or older than age 80), the predicted party differences also increase overtime, but the magnitude of change is not as substantial as those observed among the middle-aged groups. Overall, the growing polarization by party identification across time seems to be largely attributed to the widening party gap observed among the middle-aged groups.

In summary, consistent with our hypotheses, our results reveal that political affiliation is significantly associated with respondents’ perceptions of government intervention in health care, but the relationship varies across ages and across periods. The party differences are relatively small when respondents are young, but the gap widens with age and then narrows again when respondents are older than 60. The three parties also demonstrate different period trends with a widening gap observed across time, beginning with the Clinton administration.

Supplemental Analysis

We include a series of supplemental analyses to examine the robustness of our findings. First, we examine our findings with alternative model specifications. Using a hierarchical logistic regression model, we treat period as a level 2 random intercept and include random slopes for party identification (i.e., allowing party affiliation coefficients to vary across periods). Results with a hierarchical model are consistent with our present findings (see Appendix IV). Second, we also conduct a subsample analysis by running models within each party group—another way to illustrate the interaction effects between party and age, and party and period. Results of this analysis demonstrate a pattern consistent with our main findings (Appendix V). Last, instead of using listwise deletion, we use multiple imputation to estimate the missing values of the independent and control variables in our sample. The model estimated with the multiple imputation technique is presented in Appendix VI, and the predicted probabilities are presented in Appendix VII. The results based on imputed data are consistent with our main analysis.

Discussion

In this study, we set out to investigate the connections between growing political polarization in the United States and trends in public support for government intervention in health care. Drawing on life-course theories of social change (see, e.g., Alwin and McCammon 2003), we asked how the association between political party affiliation and support for government investment in health care varies by age, period, and cohort. We connected three different research literatures to accomplish our research aims: the literatures on (1) changing political attitudes with age and historical time in life-course sociology, (2) political polarization and support for government spending in political sociology, and (3) public opinion on health care in medical sociology. Overall, we found a significant association between party affiliation and support for government intervention in health care, but this association varies across different ages and historical periods. These findings shed light on how the partisan divide over government intervention in health care has grown in the recent past among middle-aged but not young or older individuals.

More specifically, we identified three sets of hypotheses regarding the associations among party affiliation, age, historical period, and cohort membership in predicting support for government intervention in health care. Using APC models, we determined that, despite our expectation of differences across cohort groups in support for government intervention and in the association between party affiliation and support for government intervention (hypotheses 2A, 2B, and 2C), there were no significant cohort effects. Given the rich life-course literature identifying cohort differences in political orientation attributable to the different historical contexts in which members of succeeding cohorts reached the critical life stage of early adulthood (Alwin, Newcomb, and Cohen 1991), we expected to find measurable differences among cohorts in their attitudes about health care. Instead, we found no meaningful variation across cohorts and a much larger role played by age and period effects. This is consistent with what Johnson and Schwadel (2019a) found with respect to cohort effects in support of federal spending on protecting the environment since the 1970s. It is likely that part of the reason why we do not find significant cohort effects is that we are analyzing attitudes toward government spending on health care. Although other beliefs and attitudes (e.g., toward gender roles or race) may crystallize in early adulthood and remain relatively stable thereafter, attitudes toward government spending are more likely to change as we age. Individual interest in government spending on health care changes first as individuals reach middle adulthood and start paying a larger proportion of their income in taxes and then again as they approach old age and face growing health concerns. For this reason, we suspect that opinions on government intervention in health care are more likely to change with age compared with other beliefs and attitudes. Using the same GSS data set, other researchers have reported mixed findings for age and cohort effects (see Street and Cossman 2006a, 2006b; Eisenstein, Clark, and Jelen 2017; Twenge et al. 2015). It is possible that the mixed findings in the literature are due to the lack of attention to political party differences. Our research uncovers different patterns of change by age in opinions about government intervention for Republicans and Democrats that would be obscured if both groups were examined together.

After narrowing our analytic focus to the effects of age and period on attitudes about government intervention in health care, we estimated our remaining models using logistic regression. We found support for the aging process model in our analysis, with younger adults more supportive of government intervention in health care than older adults (hypothesis 1B). However, when interactions between age and party were added in model 2, we found a somewhat more nuanced pattern of change in attitudes by age and political party. Consistent with the notion of old age as the “great leveler” (Ferraro and Kelley-Moore 2003; House, Lantz, and Herd 2005), we saw a decline in later life in political polarization regarding support for government intervention in health care. This is consistent with the argument that the shared experiences of later life (e.g., loss of loved ones, health problems, transitions out of full-time employment) reduce health disparities and perhaps even differences in political attitudes in later life. But surprisingly, we also found less political polarization in support for government intervention in early adulthood, perhaps suggesting that youth is also a “leveler.” It is possible that the uncertainty of young adulthood (Silva 2012), the lower levels of household wealth (Deloitte 2018), the likelihood that many individuals in this youngest age group are still completing their education and have not yet transitioned to full time work, have a similar “converging” effect on the political attitudes of young people from different political parties (providing qualified support for hypothesis 1C). We think it is particularly important to note that although the numbers of uninsured young adults have declined over time, young adults still have considerably lower levels of health insurance coverage than other age groups (U.S. Census Bureau 2018). This may produce greater similarity across party lines in views on health care in particular in this age group.

We also found evidence of change across historical periods in support for government intervention in health care. Specifically, we found more support for government intervention in the 1990s and 2000s (compared with the 1980s), consistent with hypothesis 3A, but that support declined in the 2010s back to the level of the 1980s, challenging the predictions of hypothesis 3A. When we examined period trends separately by party, however, a more nuanced understanding of the period trends emerged. Indeed, opposite trends were observed between the two parties, with Democrats overall showing increasing support for government intervention in health care and Republicans showing declining support. Consistent with the argument that political attitudes have become increasingly polarized over time, we found that the gap between Democrats and Republicans was much wider in the Obama era (2009–2016) than it was in the Reagan era (1981–1988), supporting hypothesis 3B. This is consistent with other research finding growing political polarization on environmental issues since 1992 (McCright et al. 2014). In addition, it is during the Clinton era that we first see statistically significant differences in the period effects for Democrats and Republicans, with a positive period effect for Democrats and a negative effect for Republicans, again supporting hypothesis 3B. This supports the argument that there was a “political turn” in the mid-1990s that led to skepticism regarding state-provided health benefits (Street and Cossman 2006a). We add to this past research an attention to how this political turn played out differently among Democrats and Republicans.

Finally, in our last model (Figure 2, Appendix III), we include a three-way interaction of age, party identification and period. These results show that the growing political polarization in support for government intervention in health care across historical periods occurred among middle age adults. Although we observe some overlap in the attitudes toward health care among the youngest and oldest Democrats and Republicans in the most recent period (i.e., the Obama era), it is in midlife that we see the greatest divergence in political attitudes by party. Again, we point to the common interests shared by older adults as well as by younger adults as underlying the greater similarity in political attitudes in these life stages. It is in midlife when individuals are typically established in their adult roles and, on average, earning more and paying higher income taxes than either in early adulthood or later life, that we see the greatest polarization between political parties in support for government intervention in health care.

Limitations

It is important to acknowledge the limitations of this analysis. First, it is possible that the age patterns in support for government intervention in health care could also be due in part to selection effects. More specifically, if those with the most extreme political views (either Democrat or Republican) face higher mortality risk, then the trends observed may in part be due to compositional changes in the population, rather than attributable to individuals adopting more centrist views in old age.

Second, although we evaluated the unique contributions of age, period, and cohort to social change in attitudes about government intervention and found no statistically significant effect of cohort differences in the data, we caution that future investigations should test our conclusions using panel data. Conceptually, we know that different birth cohorts experience aging differently because of changes in the sociohistorical context (Riley 1987). We found that in this span of time and for these particular birth cohorts, cohort did not have explanatory power in accounting for the social change in attitudes about government intervention in health care that has occurred. However, this does not mean that cohort metabolism plays no role in motivating broader social changes, just that it did not play a statistically significant role in this analysis of support for government intervention in health care. Panel data would allow an analysis of within-individual change in attitudes with age.

Conclusion

We set out to investigate how attitudes about government support for health care have changed since the 1980s. We draw on life-course theories of social change and on past research investigating age, period, and cohort effects in motivating changing attitudes about government programs (Street and Cossman 2006a) as well as research showing growing political polarization over time in public support for environmental issues in particular (McCright et al. 2014) in developing our research on support for government intervention in health care. Our results contribute to these literatures by shedding light on the mechanisms underlying political polarization. By introducing a life-course analysis of age, period, and cohort effects to the study of political polarization, we gain new understandings of how political polarization between Democrats and Republicans has emerged primarily among midlife individuals, not younger or older adults. Indeed, even during the polarized Obama administration, it is among midlife adults, not younger or older adults, that we see evidence of the growing divide between left and right. We contribute an explanation for this trend on the basis of the theory of old age as the great “leveler,” a theory previously used to explain declining health disparities. Our research extends this theory by positing that (1) youth may also serve as a “leveler” compared with midlife and (2) this theory seems to explain patterns of change not only in health disparities by age but also in political attitudes. Finally, to the medical sociology literature on attitudes toward government intervention in health care, we contribute an understanding of how specific periods (i.e., presidential administrations) contributed to growing polarization in attitudes among midlife adults in particular.

Supplemental Material

sj-docx-1-srd-10.1177_23780231211072394 – Supplemental material for Public Support for Government Intervention in Health Care in the United States from 1984 to 2016

Supplemental material, sj-docx-1-srd-10.1177_23780231211072394 for Public Support for Government Intervention in Health Care in the United States from 1984 to 2016 by Yaqi Yuan, Kristen Schultz Lee and Yunmei Lu in Socius

Footnotes

Acknowledgements

We would like to thank the editors of Socius and two anonymous reviewers. We are also grateful to Drs. Hanna Grol-Prokopczyk, Jessica Houston Su, Alexandra Brewer, and Catherine Harnois for helpful comments on this project.

ORCID iDs

Yaqi Yuan

Kristen Schultz Lee

Supplemental Material

Supplemental material for this article is available online.

Notes

Author Biographies

Yaqi Yuan, PhD, is a visiting assistant professor of sociology at Wake Forest University. She received her doctorate in sociology from the University at Buffalo, SUNY. Her research interests include cross-national studies, public opinion on health care system/s, aging and the life course, socio-legal studies, and mixed methods. Her work has appeared in Families, Relationships and Societies and Health Policy. Her current research is focused mainly on the impact of the coronavirus disease 2019 pandemic on older adults in Singapore and in the United States.

Kristen Schultz Lee, PhD, is an associate professor of sociology at the University at Buffalo, SUNY. After completing her doctorate in sociology at Cornell University, she was a postdoctoral fellow at the Population Research Institute at Penn State University. Her research focuses on gender, aging, happiness, and the life course and has appeared in Social Forces, Social Problems, the Journal of Gerontology: Social Sciences, the Journal of Marriage and Family, and Gender & Society. Her current research includes intergenerational analyses of health in mid- and later life and of gender differences in how educational attainment and happiness are related.

Yunmei Lu, PhD, is an assistant professor of sociology at the University at Buffalo, SUNY. Her research interests include social change and crime, the life course, APC analysis, and cross-cultural studies. Her work has appeared in Criminology, The Sociological Quarterly, Rural Sociology, and the Journal of Quantitative Criminology. Her current research explores the impacts of social change on crime and perceptions of crime among different population subgroups.

References

Alwin

Duane F.

Cohen

Ronald Lee

Newcomb

Theodore Mead

. 1991. Political Attitudes over the Life Span: The Bennington Women after Fifty Years Univ of Wisconsin Press.

Alwin

Duane F.

McCammon

Ryan J.

2003. “Generations, Cohorts, and Social Change.” Pp. 23–49 in Handbook of the Life Course, edited by Mortimer

J. T.

Shanahan

M. J.

New York: Springer.

Anderson

Amy L.

Lytle

Robert

Schwadel

Philip

. 2017. “Age, Period, and Cohort Effects on Death Penalty Attitudes in the United States, 1974–2014.” Criminology 55(4):833–68.

Baldassarri

Delia

Bearman

Peter

. 2007. “Dynamics of Political Polarization.” American Sociological Review 72(5):784–811.

Bartels

Larry M.

2002. “Beyond the Running Tally: Partisan Bias in Political Perceptions.” Political Behavior 24(2):117–50.

Brodie

Mollyann

Hamel

Elizabeth C.

Kirzinger

Ashley

DiJulio

Bianca

. 2019. “Partisanship, Polling, and the Affordable Care Act.” Public Opinion Quarterly 83(2):423–49.

Conover

Pamela Johnston

. 1984. “The Influence of Group Identifications on Political Perception and Evaluation.” Journal of Politics 46(3):760–85.

Corman

Juliane

Levin

David

. 2016. “The Polls-Trends Support for Government Provision of Health Care and the Patient Protection and Affordable Care Act.” Public Opinion Quarterly 80(1):114–79.

Dalen

James E.

Waterbrook

Keith

Alpert

Joseph S.

2015. “Why Do So Many Americans Oppose the Affordable Care Act?” American Journal of Medicine 128(8):807–10.

10.

Danigelis

Nicholas L.

Hardy

Melissa

Cutler

Stephen J.

2007. “Population Aging, Intracohort Aging, and Sociopolitical Attitudes.” American Sociological Review 72(5):812–30.

11.

Deloitte. 2018. “Are We Headed for a Poorer United States? Growing Wealth Inequality by Age Puts Youngers Households Behind.” Retrieved June 9, 2021. https://www2.deloitte.com/us/en/insights/economy/issues-by-the-numbers/march-2018/us-average-wealth-inequality-by-age.html.

12.

Eisenstein

Marie A

Clark

April K

Jelen

Ted G

. 2017. “Political Tolerance and Religion: An Age-Period-Cohort Analysis, 1984–2014.” Review of Religious Research 59(3):395–418.

13.

Elder

Glen H.

1999. Children of the Great Depression. 25th anniv. ed. New York: Taylor & Francis.

14.

Ferraro

Kenneth F.

Kelley-Moore

Jessica A.

2003. “Cumulative Disadvantage and Health: Long-Term Consequences of Obesity?” American Sociological Review 68(5):707–29.

15.

Fiorina

Morris P.

Abrams

Samuel J

. 2008. “Political Polarization in the American Public.” Annual Review of Political Science 11:563–88.

16.

Fosse

Ethan

Winship

Christopher

. 2019. “Analyzing Age-Period-Cohort Data: A Review and Critique.” Annual Review of Sociology 45:467–92.

17.

Garfinkel

Irwin

Rainwater

Lee

Smeeding

Timothy

. 2010. Wealth and Welfare States: Is America a Laggard or Leader? New York: Oxford University Press.

18.

Gelman

Andrew

Lee

Daniel

Ghitza

Yair

. 2010. “Public Opinion on Health Care Reform.” The Forum 8(1):Article 8.

19.

Ghitza

Yair

Gelman

Andrew

. 2014. “The Great Society, Reagan’s Revolution, and Generations of Presidential Voting.” Retrieved December 26, 2021. http://www.stat.columbia.edu/~gelman/research/unpublished/cohort_voting_20191017.pdf.

20.

Glenn

Norval D.

1974. “Aging and Conservatism.” Annals of the American Academy of Political and Social Science 415(1):176–86.

21.

Hacker

Jacob S.

Pierson

Paul

. 2018. “The Dog That Almost Barked: What the ACA Repeal Fight Says about the Resilience of the American Welfare State.” Journal of Health Politics, Policy and Law 43(4):551–77.

22.

Hasenfeld

Yeheskel

Rafferty

Jane A.

1989. “The Determinants of Public Attitudes toward the Welfare State.” Social Forces 67(4):1027–48.

23.

Henderson

Michael

Hillygus

D. Sunshine

. 2011. “The Dynamics of Health Care Opinion, 2008–2010: Partisanship, Self-Interest, and Racial Resentment.” Journal of Health Politics, Policy and Law 36(6):945–60.

24.

Hoffman

Beatrix

. 2012. Health Care for Some: Rights and Rationing in the United States since 1930. Chicago: University of Chicago Press.

25.

House

James S.

Lantz

Paula M.

Herd

Pamela

. 2005. “Continuity and Change in the Social Stratification of Aging and Health over the Life Course: Evidence from a Nationally Representative Longitudinal Study from 1986 to 2001/2002 (Americans’ Changing Lives Study).” Journals of Gerontology: Series B 60B(1):15–26.

26.

Hunter

Nan D.

2008. “Risk Governance and Deliberative Democracy in Health Care.” The Georgetown Law Journal 97:1.

27.

Johnson

Erik W.

Schwadel

Philip

. 2019a. “It Is Not a Cohort Thing: Interrogating the Relationship between Age, Cohort, and Support for the Environment.” Environment and Behavior 51(7):879–901.

28.

Johnson

Erik W.

Schwadel

Philip

. 2019b. “Political Polarization and Long-Term Change in Public Support for Environmental Spending.” Social Forces 98(2):915–41.

29.

Kitschelt

Herbert

Rehm

Philipp

. 2006. “New Social Risk and Political Preferences.” pp. 52-82 in The Politics of Post-Industrial Welfare States, edited by Armingeon

Bonoli

Routledge.

30.

Kluegel

James R.

Miyano

Masaru

. 1995. “Justice Beliefs and Support for the Welfare State in Advanced Capitalism.” Pp. 81–105 in Social Justice and Political Change: Public Opinion in Capitalist and Post-communist States, edited by Kluegel

J. R.

Mason

D. S.

Wegener

New York: Aldine de Gruyter.

31.

Kriner

Douglas L.

Reeves

Andrew

. 2014. “Responsive Partisanship: Public Support for the Clinton and Obama Health Care Plans.” Journal of Health Politics, Policy and Law 39(4):717–49.

32.

Larsen

Christian Albrekt

. 2008. “The Institutional Logic of Welfare Attitudes: How Welfare Regimes Influence Public Support.” Comparative Political Studies 41(2):145–68.

33.

LeCount

Ryan Jerome

Abrahamson

Kathleen

. 2017. “Self-Rated Health and Attitudes about US Health Care Policy.” Sociological Spectrum 37(4):237–49.

34.

Lenz

Gabriel S.

2009. “Learning and Opinion Change, Not Priming: Reconsidering the Priming Hypothesis.” American Journal of Political Science 53(4):821–37.

35.

Yunmei

Amory

Kacy

Shi

Luzi

. 2021. “Social Change and Gendered Reaction to the Threat of Victimization, 1973–2016: An Age-Period-Cohort Analysis.” The Sociological Quarterly: 1–27. https://doi.org/10.1080/00380253.2021.1945978

36.

Luo

Liying

Hodges

James S.

2020. “The Age-Period-Cohort-Interaction Model for Describing and Investigating Inter-cohort Deviations and Intra-cohort Life-Course Dynamics.” Sociological Methods & Research. Online First.

37.

Margalit

Yotam

. 2013. “Explaining Social Policy Preferences: Evidence from the Great Recession.” American Political Science Review 107(1):80–103.

38.

Marsden

Peter V

Smith

Tom W

Hout

Michael

. 2020. “Tracking Us Social Change over a Half-Century: The General Social Survey at Fifty.” Annual Review of Sociology 46:109–34.

39.

McCright

Aaron M

Xiao

Chenyang

Dunlap

Riley E.

2014. “Political Polarization on Support for Government Spending on Environmental Protection in the USA, 1974–2012.” Social Science Research 48:251–60.

40.

McIntyre

Adrianna

Blendon

Robert J.

Benson

John M.

Findling

Mary G.

Schneider

Eric C.

2020. “The Affordable Care Act’s Missing Consensus: Values, Attitudes, and Experiences Associated with Competing Health Reform Preferences.” Journal of Health Politics, Policy and Law 45(5):729–55.

41.

Meuleman

Bart

Chung

Heejung

. 2012. “Who Should Care for the Children? Support for Government Intervention in Childcare.” Pp. 107–33 in The Future of the Welfare State: Social Policy Attitudes and Social Capital in Europe, edited by Ervasti

Andersen

J. G.

Fridberg

Ringdal

London: Elgar.

42.

Mize

Trenton D.

2019. “Best Practices for Estimating, Interpreting, and Presenting Nonlinear Interaction Effects.” Sociological Science 6(4):81–117.

43.

Morone

James A.

2016. “Partisanship, Dysfunction, and Racial Fears: The New Normal in Health Care Policy?” Journal of Health Politics, Policy and Law 41(4):827–46.

44.

Mustillo

Sarah A.

Lizardo

Omar A.

McVeigh

Rory M.

2018. “Editors’ Comment: A Few Guidelines for Quantitative Submissions.” American Sociological Review 83(6):1281–83.

45.

Olafsdottir

Sigrun

Pescosolido

Bernice

. 2010. “Is Health Care Reform Contested in the US? Public Attitudes toward Government Involvement in Health Care, 1973–2010.” Unpublished manuscript, Department of Sociology, Boston University.

46.

Pacheco

Julianna

Haselswerdt

Jake

Michener

Jamila

. 2020. “The Affordable Care Act and Polarization in the United States.” Russell Sage Journal of the Social Sciences 6(2):114–30.

47.

Pew Research Center. 2012. “Partisan Polarization Surges in Bush, Obama Years: Trends in American Values, 1987–2012.” Retrieved February 1, 2020. https://www.peoplepress.org/2012/06/04/partisanpolarization-surges-in-bush-obama-years.

48.

Pew Research Center. 2020. “Millennial and Gen Z Republicans Stand Out from Their Elders on Climate and Energy Issues.” Retrieved June 9, 2021. https://www.pewresearch.org/fact-tank/2020/06/24/millennial-and-gen-z-republicans-stand-out-from-their-elders-on-climate-and-energy-issues.

49.

Quadagno

Jill

. 2006. One Nation, Uninsured: Why the U.S. Has No National Health Insurance. New York: Oxford University Press.

50.

Quadagno

Jill

. 2010. “Institutions, Interest Groups, and Ideology: An Agenda for the Sociology of Health Care Reform.” Journal of Health and Social Behavior 51(2):125–36.

51.

Quadagno

Jill

. 2014. “Right-Wing Conspiracy? Socialist Plot? The Origins of the Patient Protection and Affordable Care Act.” Journal of Health Politics, Policy and Law 39(1):35–56.

52.

Quadagno

Jill

Pederson

JoEllen

. 2012. “Has Support for Social Security Declined? Attitudes toward the Public Pension Scheme in the USA, 2000 and 2010.” International Journal of Social Welfare 21(Suppl. 1):S88–S100.

53.

Rehm

Philipp

. 2009. “Risks and Redistribution: An Individual-Level Analysis.” Comparative Political Studies 42(7):855–81.

54.

Riley

Matilda White

. 1987. “On the Significance of Age in Sociology.” American Sociological Review 52(1):1–14.

55.

Roberts

Brent W.

Walton

Kate E.

Viechtbauer

Wolfgang

. 2006. “Patterns of Mean-Level Change in Personality Traits across the Life Course: A Meta-analysis of Longitudinal Studies.” Psychological Bulletin 132(1):1–25.

56.

Ryder

Norman B.

1985. “The Cohort as a Concept in the Study of Social Change.” Pp. 9–44 in Cohort Analysis in Social Research, edited by Mason

W. M.

Fienberg

S. E.

New York: Springer.

57.

Schwadel

Philip

Garneau

Christopher R. H.

2014. “An Age–Period–Cohort Analysis of Political Tolerance in the United States.” Sociological Quarterly 55(2):421–52.

58.

Shapiro

Robert Y.

Young

John T.

1989. “Public Opinion and the Welfare State: The United States in Comparative Perspective.” Political Science Quarterly 104(1):59–89.

59.

Shi

Luzi

Yunmei

Pickett

Justin T.

2020. “The Public Salience of Crime, 1960–2014: Age–Period–Cohort and Time–Series Analyses.” Criminology 58(3):568–93.

60.

Silva

Jennifer M.

2012. “Constructing Adulthood in an Age of Uncertainty.” American Sociological Review 77(4):505–22.

61.

Skocpol

Theda

. 1995. “The Rise and Resounding Demise of the Clinton Plan.” Health Affairs 14(1):66–85.

62.

Skocpol

Theda

. 1996. Boomerang: Health Care Reform and the Turn against Government. New York: W. W. Norton.

63.

Smith

Tom W.

Davern

Michael

Freese

Jeremy

Morgan

Stephen

. 2018. “General Social Surveys, 1972–2018” [machine-readable data file]. Chicago: NORC.

64.

Street

Debra

Cossman

Jeralynn Sittig

. 2006a. “Altruism or Self-Interest—Social Spending and the Life Course.” Journal of Sociology and Social Welfare 33(3):73–99.

65.

Street

Debra

Cossman

Jeralynn Sittig

. 2006b. “Greatest Generation or Greedy Geezers? Social Spending Preferences and the Elderly.” Social Problems 53(1):75–96.

66.

Tilley

James

Evans

Geoffrey

. 2014. “Ageing and Generational Effects on Vote Choice: Combining Cross-Sectional and Panel Data to Estimate APC Effects.” Electoral Studies 33(1):19–27.

67.

Twenge

Jean M.

Carter

Nathan T.

Campbell

W. Keith

. 2015. “Time Period, Generational, and Age Differences in Tolerance for Controversial Beliefs and Lifestyles in the United States, 1972–2012.” Social Forces 94(1):379–99.

68.

U.S. Census Bureau. 2018. “Another Look at Health Insurance Coverage Rates for Young Adults.” Retrieved June 9, 2021. https://www.census.gov/newsroom/blogs/random-samplings/2018/09/another_look_at_heal.html/.

69.

Yang

. 2008. “Social Inequalities in Happiness in the United States, 1972 to 2004: An Age-Period-Cohort Analysis.” American Sociological Review 73(2):204–26.

70.

Yang

Land

Kenneth C.

2008. “Age–Period–Cohort Analysis of Repeated Cross-Section Surveys: Fixed or Random Effects?” Sociological Methods & Research 36(3):297–326.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.28 MB