An interesting aspect of predicting climate change opinion is that the effects of numerous factors are contingent upon an individual’s political orientation, whether these factors are educational attainment (Hamilton 2008; Hamilton and Keim 2009; McCright 2011; McCright and Dunlap 2011a), self-assessed scientific understanding (Hamilton 2011; McCright and Dunlap 2011a), objectively-assessed scientific understanding (Hamilton et al. 2012), or income (the contribution of the present study). Across all of these categories, greater political polarization is found among relative elites.

McCright (2011) suggested two theoretical perspectives—information processing theory and the elite cues hypothesis—that might explain the interaction between political orientation and education in predicting climate change concern and belief. Information processing theory explains how values shape the way individuals understand the presentation of new information on a given issue. Many scholars have questioned the efficacy of a purely information-based strategy when communicating about climate change, as political orientation appears to play such a dominant role in structuring how an individual reacts to scientific knowledge. Information processing theory explains this by noting that many individuals lack the cognitive skills necessary to process all types of information they come across, instead processing information through experiential, social, and ideological filters (Wood and Vedlitz 2007). In cases of ambiguity, people especially engage in information processing, relying on their predispositions to form opinion (Jones and Baumgartner 2005). Thus, in the case of processing information on climate change, we can expect an individual’s political values to play an important role in developing beliefs.

Consistent with information processing theory is the elite cues hypothesis. Rather than reacting directly to scientific information, people may also take cues from political elites they identify as trustworthy when forming opinions on controversial issues. In cases of controversial issues that undergo intense public debates, people are willing to change their attitudes in accordance with cues taken from political elites. When the Clinton administration campaigned to build support for the Kyoto treaty, public debates did little to change the overall balance of national public opinion. Beneath this, however, strong Democrats and strong Republicans each rallied around the respective positions of their partisan leaders (Krosnick et al. 2000).

Ideally, survey researchers can measure scientific knowledge by quizzing respondents on a range of questions assessing their objective knowledge on an issue in order to understand its influence in forming opinions about scientific controversies (Malka et al. 2009). However, many surveys instead ask respondents to report their understanding of a given issue (which may be measuring how tuned in an individual feels to a particular information source as opposed to their objective knowledge). Both types of research find that increased levels of scientific knowledge or self-reported understanding predicts a smaller likelihood of climate skepticism for liberals and Democrats, but has either no effect on or increases the likelihood of climate skepticism for conservatives and Republicans (Malka et al. 2009; McCright and Dunlap 2011a).

Other research approaches the topic of scientific literacy indirectly through educational attainment. This research finds similar patterns regarding climate skepticism, as educational attainment works differently for liberals and Democrats than it does for conservatives and Republicans. Hamilton (2008) initiated this line of research through an analysis of national public opinion regarding the polar regions. He found that increased education predicted greater levels of concern over global warming among liberals and moderates, but predicted lower levels of concern among conservatives. These findings were replicated at the regional scale by Hamilton and Keim (2009), Hamilton (2011), and Hamilton and Lemcke-Stampone (2013). McCright and Dunlap (2011a) tested this relationship through 10 years of national survey data in the U.S., finding both ideology and partisan identification to have significant interactions with education in predicting attitudes about climate change. They expanded previous research to find that, among conservatives and Republicans, increased levels of both education and self-reported scientific understanding predicted lower levels of concern and less accurate beliefs about global warming. In related research, Gauchat (2012) found that increased education among conservatives predicted greater levels of distrust in scientific authority.

McCright (2011) connected the moderator effect of political orientation on education to information processing theory and the elite cues hypothesis noting that in an age of self-selecting media constellations, increasingly educated individuals may view themselves as very informed on a scientific issue even if they do not expose themselves to informed perspectives. Even if conservative individuals are very in tune with the information communicated through talk radio, FOX News, or conservative think tanks, they may not have accurate understandings of climate change despite how informed they feel. Thus, “greater educational attainment and more focused attention to their respectively favored news outlets and political elites may increase how much citizens think they understand about the issue” (2011:249). But because liberals and conservatives diverge in selecting their information sources, the effects of increased education work differently across these groups in predicting global warming beliefs and concern.

What has not been previously considered is the possibility of a similar interaction effect between political orientation and income. Much research within environmental sociology indicates that income serves as a poor predictor of environmental concern in the U.S. (Van Liere and Dunlap 1980; Jones and Dunlap 1992; Mohai and Bryant 1998). Some argue that a relationship exists between increased income and pro-environmental concern, such as the affluence hypothesis (Franzen 2003; Franzen and Meyer 2010) and the postmaterialism perspective (Inglehart 1990, 1995). However, these perspectives have been criticized for overlooking how environmental concern can be expressed through nonfinancial means, how materialist values can express concern over environmental degradation, or that shifts in pro-environmental values are not limited to affluent societies (Brechin and Kempton 1994; Dunlap and Mertig 1997; Dunlap and York 2008).