Measuring and Analyzing Protest Potential From a Survey Data Recycling Framework

Abstract

As active involvement in protest has been legitimized as an acceptable form of political activity, citizens’ protest potential has become an important measure to understand contemporary democratic politics. However, the arbitrary use of a forced-choice question, which prevents those who have previously participated in protests from expressing willingness to engage in future protest, and the limited coverage of international surveys across countries and years have impeded comparative research on protest potential. This research develops a new systematic weighting method for the measurement of protest potential for comparative research. Using the 1996 International Social Survey Program survey, which asks two separate questions about “have done” and “would do” demonstrations, I create a weighting scale for the forced-choice question by estimating the predicted probabilities of protest potential for those who have already participated in demonstrations. Capitalizing on the survey data recycling framework, this study also controls for harmonization procedures and the quality of surveys, thereby expanding the cross-national and temporal coverage beyond the affluent Western democracies. The results show that this weighting scale provides a valid measure of protest potential, and the survey data recycling framework improves comparability between surveys.

How do political opportunities impact protest potential? A multilevel cross-national assessment

“…The type of political system or regime has factored centrally in such research, considering how democracy (Jo and Choi, 2019), "free" or "not-free" status (Slavina, 2021), political regime (Lavrinenko, 2021(Lavrinenko, , 2022, and "established democracy" (Oser and Hooghe, 2018) generally predict protest. In addition, a regime's "durability" has also been considered, finding that "long standing democracy" (Kwak, 2022) and "years of continuous democracy" (Jenkins et al, 2008) all significantly predict greater protest. Also, some studies have used operationalizations of "rights" to predict protest, such as Freedom House's political rights and civil liberties (Weiner and Federico, 2017) and Democracy Barometer's "liberties" (Muliavka, 2021)-although the former found a nonsignificant effect.…”

“…Some studies have modeled the effects of POs, but typically just one PO per study (e.g. Dodson, 2015;Giugni and Grasso, 2018;Kwak, 2022;Lavrinenko, 2021Lavrinenko, , 2022Oser and Hooghe, 2018;Weiner and Federico, 2017); the most popular PO considered is often regime democracy or some variant. For example, the early research by Jenkins et al (2008) indicated it is important to consider the contextual environment (including time and place) in which protest occurs, but only included one key PO variable (years of continuous democracy) in their otherwise important study.…”

“…While some past research has considered whether someone has or has not previously protested using one specific tactic (Benson and Rochon, 2004; Dauphinais et al, 1992; Dodson, 2015; Gattinara et al, 2021; Jo and Choi, 2019; Kurer et al, 2019; Lorenzini, 2015; Muliavka, 2021; Paulsen, 1994; Schussman and Soule, 2005), another common approach is to study individuals’ “protest potential”—the disposition one has toward protesting (e.g. Anderson and Mendes, 2005; Jenkins and Wallace, 1996; Kalaycioğlu, 1994; Kriesi et al, 1992; Kwak, 2022; Lavrinenko, 2021, 2022; Lee, 2003; Lee and Norris, 2000; Leysens, 2006; Marsh and Kaase, 1979; Morrison, 1998; Spehr and Dutt, 2004; Weiner and Federico, 2017). Protest potential combines both past behavior and future intent, and often includes multiple forms of protest—often combinations of forms like demonstration attendance, boycotting, petition signing, labor striking, and so forth.…”

How to Capture Global Protest Trends: Using Survey Data Recycling Data to Construct Cross-National Trends in Protest

“…We test for both the linear and quadratic forms, which have both received support in the literature. At the national survey level, we control for harmonization adjustments and survey quality, which are provided by the SDR data set and control for intersurvey variability reflecting artifacts of harmonization decisions and/or differences in survey quality (see also, Kwak, this issue; Slomczynski et al, this issue). Table 1 provides descriptive statistics for these variables.…”

“…Sixty-four percent of our surveys had one or more issues on these seven variables. Such inconsistencies between survey documentation and the underlying data have been found to affect the substantive results produced by analyses of surveys (Biemer, 2016: Groves & Lyberg, 2010; Kwak, this issue; Oleksiyenko et al, 2019). For each data quality measure, higher values represent poorer data quality.…”