Elite Cueing, Gender, and the Partisan Gap in Environmental Support

ABSTRACT

This study extends research on how political and social contexts mediate the relationship between men’s and women’s political identities and their support for the environment. Our test consists of a 3 × 2 factorial survey experiment that interacts an individual’s partisan identity with three conditions modeling elite support for federal spending to protect the environment. Democrats demonstrate consistently high rates of support regardless of legislative sponsorship. Among Republicans, when Democratic legislators sponsor environmental legislation, support drops about 18% compared to conditions where a bipartisan group sponsors legislation. Support among Republican women, however, remains relatively stable across experimental conditions. Republican variations in support are almost entirely attributable to declines in support among Republican men, whose support drops an average of 24% when Democratic legislators sponsor legislation. In conclusion, intraparty gender differences in opinions and support should be considered more broadly, especially when examining polarized public issues.

KEYWORDS:

• Elite cueing
• political polarization
• gender
• environmental opinion
• experimental methods

Acknowledgments

The authors would like to thank Christine Horne, Julie Kmec, and Dylan Bugden for feedback on earlier versions of this paper, as well as participants at an ASA Section on the Environment in August 2021.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1. The effects size tells us about the relationship between variables, and the proportion of explained variance.

2. Participants were paid $1.20 per survey, which equates to an average pay rate of $15.37 per hour (based on actual recorded completion times).

3. Of the 400 recruited Democrats, 6 self-identified as Republicans in this study. Of 400 recruited Republicans, 17 self-identified as Democrats.

4. Participants were asked to recall the political party of legislators sponsoring environmental spending in the vignette which they were assigned to read. Those answering incorrectly were dropped from the experiment.

5. Analysis of studentized residuals is more effective for detecting outlying observations than standardized residuals. Studentized residuals take into consideration the potential that an outlier is so influential to the regression model that the estimated function is influenced to the degree that the outlier is not flagged using the standardized residual criterion. By removing each observation and recalculating the function, observations that are more than three standard deviations from the mean of a function that is not influenced by the given observation are dropped from the final regression analysis.

6. A traditional experimental control group might avoid describing the political party of Senators supporting legislation, instead referring only to e.g. “U.S. Senators proposing … .” Given that the environment is now a Democratic owned issue, however, we were concerned that study participants might reasonably assume that the sponsoring Senators in this condition were Democrats. Moreover, a fictitious news clipping that described U.S. Senators proposing legislation but absent the party identification of those Senators would detract from the plausibility and realism of the scenario. Consequently, we proceeded with bi-partisan sponsorship as the control group. This approach still allows for assessing the within-group and between-group differences that are the focus of our study.

7. We do not include a control assessing survey respondents’ support for governmental spending generally (i.e. absolute support, see Johnson and Schwadel 2019a), as our focus is understanding how individual partisans’ support (differentially) varies across conditions of elite partisan sponsorship. Research focused on estimating group differences in absolute support for environmental spending in the broader population would necessitate an alternative methodological approach (i.e. a nationally representative survey). Group estimates of absolute support for environmental spending are thus not readily interpretable or comparable as differences along partisan or gender lines may be driven by support for governmental spending more generally.

8. The “other” category includes the following values: “Black/African American” N = 41, “Asian/Pacific Islander” N = 104, “Native American or American Indian” N = 5, “Hispanic” N = 45, “Multi racial” N = 17, “Other” N = 3, “Prefer not to answer” N = 4.

9. The first question asked: “Some people seem to follow what’s going on in government and public affairs most of the time, whether or not there’s an election going on. Others aren’t that interested. Would you say you follow what’s going on in government and public affairs most of the time, some of the time, only now and then, or hardly at all?” Second, respondents were asked: “Do you happen to know which party elected the most members to the House of Representatives in this last election?” Choices include “Republicans,” “Democrats,” or “I don’t know/I cannot recall.” An accurate response potentially indicates greater political involvement. Third, respondents were asked if they are registered to vote, allowing for a yes or no response. The Cronbach’s alpha value is 0.866.

Additional information

Notes on contributors

Azdren Coma

Azdren Coma is a doctoral candidate and sociology instructor at Washington State University. He is currently working on research aiming to understand public support for climate change legislation throughout the United States. Azdren also works for the Social and Economic Sciences Research Center (SESRC) at WSU in various capacities related to survey research and methodologies.

Erik W. Johnson

Erik W. Johnson is an Associate Professor of Sociology at Washington State University. His research focuses on the emergence, development and institutional outcomes of environmental movements. Johnson’s research has appeared in journals such as Social Forces, Environmental Sociology, Mobilization, Environment and Behavior, Nonprofit and Voluntary Sector Quarterly, and Social Problems.

Philip Schwadel

Philip Schwadel is a Professor of Sociology at the University of Nebraska-Lincoln. He received his PhD from Penn State University and was previously a Postdoctoral Researcher with the National Study of Youth and Religion and a Senior Researcher with the Pew Research Center. His research focuses on Americans’ religious and political behaviors—and the intersections between the two—with emphases on the associations between religion and social class, social contexts, social change, and youth. Professor Schwadel’s work appears in generalist journals such as Social Forces, Social Science Research, and Contexts; and in disciplinary journals including Sociology of Religion, Journal for the Scientific Study of Religion, Criminology, and Environment & Behavior.