The states of public opinion on the environment

ABSTRACT

State-level public opinion on the environment within the US is examined, using data from the General Social Survey (GSS) from 2000 to 2010. A Multilevel Regression and Post-Stratification (MRP) approach is taken to estimate three different survey items on environmental issues at the US state level. This allows for a comparable understanding of state-level public opinion on environmental issues, which is then connected to economic, environmental, and political conditions, and representation in policy making. The findings indicate the MRP technique is a sophisticated way to make accurate estimations of state-level public opinion, and those estimates provide further insight into the shaping of public opinion and public policy.

KEYWORDS:

• US states
• public opinion
• multilevel regression

Notes

1. Spending: 1 = too little, 0 = about the right amount and too much; efficacy, 1 = strongly disagree and disagree, 0 = neither agree nor disagree, agree, and strongly agree; sacrifice, 1 = very willing and fairly willing, 0 = neither willing nor unwilling, not very willing, and not at all willing.

2. Other demographic variables were not included in the final estimations for one of two reasons. First, insufficient data were available from either the GSS or US Census. Second, the variable was inconsequential to the analysis, mostly due to a lack of statistically significant results. Sex is included in the interest of consistency, and the contribution the variable makes in the spending and efficacy models.

3. Insufficient data were available for Nebraska, Nevada, New Hampshire, Rhode Island, and Utah.

4. District of Columbia is included in the estimation models.

5. Two categories for sex × four categories for age × four categories for degree × 46 states.

6. This is the adjusted real dollar value when controlling for inflation over time, with 2009 as the base year.

7. This measure of political ideology presents a limitation, as there are many other alternatives available. However, it was identified based on its temporal sensitivity, availability, and substantive value in the predictive models. See Erikson et al. (1993), Berry et al. (1998), and Berry et al. (2010).

8. ACU ratings are left out of these models, since both measures rely on some of the same roll-call votes.

9. 45 states × 6 years. District of Columbia is not included in the OLS models. Years included are 2000, 2002, 2004, 2006, 2008, and 2010.

10. It would be inappropriate to use any of these demographic factors in the regression models presented later, as these demographic factors estimate the state-level public attitude scores.

11. Stata produces an individual coefficient for each group for each variable at Level 2.

12. District of Columbia was not considered a part of any region.