close× Call Us
close×
Title Effects of Party Cues on the Perceived Legitimacy for International Organizations
Post date 05/07/2019
C1 Background and Explanation of Rationale

This study explores when and to what extent party cues impact the perceived legitimacy of international organizations (IOs). It builds on and extends a stream of research interested in effects of elite communication on public opinion toward international issues and organizations (Maier et al. 2012; Guisinger and Saunders 2017; Torcal et al. 2018; Dellmuth and Tallberg 2019). This stream in turn relates to a broader literature on the effects of elite cueing and framing on public opinion (for overviews, see Druckman and Lupia 2000; Chong and Druckman 2007; Busby et al. 2018).

In our earlier research, we have assessed the effects of communication by three types of elites on the popular legitimacy of IOs: national governments, civil society organizations, and IOs themselves (Dellmuth and Tallberg 2019). We found that both national governments and civil society organizations are capable of swaying people’s opinions of IOs, particularly through negative messages, while communication by IOs themselves tends to be less effective. In this study, we take our research one step further by focusing specifically on the effects of party cues on the perceived legitimacy of IOs. This is based on the intuition that party cues may function as a heuristic for citizens, not only in domestic politics (Levendusky 2010; Druckman et al. 2013; Leeper and Slothuus 2010; Brookman and Butler 2017), but also in opinion formation regarding IOs. For this purpose, we have designed two vignette experiments that tests the effects of party cues, party polarization, and partisan identity on citizen perceptions of IO legitimacy.

C2 What are the hypotheses to be tested?

We operationalize legitimacy beliefs using the measure of confidence in IOs. The experiments test three hypotheses about effects of party cues, party polarization, and partisan identity on citizens’ legitimacy beliefs vis-à-vis IOs, building on the work of Druckman et al. (2013).

H1 Party cues: When partisans receive a frame sponsored by their party and a conflicting frame sponsored by another party, they will be more likely to move in the direction of their party’s frame than in the direction of the other party’s frame.

H2 Party polarization: In a polarized environment, when partisans receive a frame sponsored by their own party and a conflicting frame sponsored by another party, they will be more likely to move in the direction of their party’s frame than in the direction of the other party’s frame to a greater extent than they do in a non-polarized environment.

H3 Partisan identity: The strength of the effect of party cues (H1) and party polarization (H2) will be stronger among citizens with a stronger partisan identity.

C3 How will these hypotheses be tested? * We devised two identically designed vignette experiments, which appear in the same survey. The order of the experiments is block randomized for each respondent to reduce the likelihood of spillover effects from one experiment to another. One experiment focuses on party cues regarding military spending on NATO, and the other experiment on party cues regarding reductions in the number of refugees accepted under the UN’s refugee convention. These two issues share several features that make them well suited to test our hypotheses (Druckman et al. 2013). First, both issues received attention in public debates in Germany and the US prior to our study. Second, while both issues are topically relevant, we can suspect that the public’s attention on them is not crystallized and instead somewhat conflicted, due to multiple considerations on each issue. Third, both issues are such that the main parties in Germany and the US tended to hold different positions, but not persistently dramatically opposed positions, which allowed our treatments to vary the level of party polarization. In each experiment, 2000 respondents are randomized into four treatment groups of 400 respondents each and a control group of 400 respondents. Respondents are assigned the same condition in both experiments, since we worry that the degree of party polarization will otherwise be confusing and since this design makes it easier to assess any potential spillover effects (Transue et al. 2009; Druckman et al. 2013). The five groups were assigned the following types of information, followed by the question about confidence in an IO: - Control group: only factual information on the issue - Treatment group 1: factual information on the issue + balanced frames about the main arguments on each side - Treatment group 2: factual information on the issue + balanced frames about the main arguments on each side + party endorsements - Treatment group 3: factual information on the issue + balanced frames about the main arguments on each side + party endorsements + low polarization environment - Treatment group 4: factual information on the issue + balanced frames about the main arguments on each side + party endorsements + high polarization environment It should be observed that treatment group 1 offers an alternative baseline for comparison next to the control group, since it allows us to test how respondents react to frames in the absence of party endorsements. In the US, we focused on the Democrats and the Republicans as the two main opposing parties. In Germany, we focused on the CDU/CSU and the SPD as the two main opposing parties.
C4 Country Germany, United States
C5 Scale (# of Units) 2000 in each country
C6 Was a power analysis conducted prior to data collection? Yes
C7 Has this research received Insitutional Review Board (IRB) or ethics committee approval? not provided by authors
C8 IRB Number not applied, since not considered necessary for previous, similar experiments
C9 Date of IRB Approval not provided by authors
C10 Will the intervention be implemented by the researcher or a third party? YouGov
C11 Did any of the research team receive remuneration from the implementing agency for taking part in this research? No
C12 If relevant, is there an advance agreement with the implementation group that all results can be published? not provided by authors
C13 JEL Classification(s) not provided by authors