What did Quebeckers Want? Impact of Question Wording, Constitutional Proposal and Context on Support for Sovereignty, 1976–2008

Abstract

Abstract The idea of sovereignty has been part of Québec's political landscape for more than 45 years, giving rise to two referendums. Throughout this period, pollsters surveyed the population. Using close to 700 polls published since the first election of Parti Québécois in 1976, this article examines how support for sovereignty evolved outside the referendum campaigns and since the last one in the Fall of 1995. Using longitudinal multilevel analysis, it takes into account two levels of explanation: first, the wording of the question, which includes the type of constitutional proposal polled; and second, the passage of time and the events that marked it. This meta-analysis confirms that sovereignty combined with an association with the rest of Canada has always stood apart from more "extreme" constitutional options as having the highest support, and it therefore automatically resurfaced at the start of any referendum campaign. It also confirms that the different events that marked the period only had a temporary effect on support for sovereignty. Keywords: sovereigntypollsquestion wordinglongitudinal multilevel analysisQuébec Acknowledgments Authors wish to thank Claude Gauthier at CROP, Grégoire Gollin at Createc and Jean-François Lisée, who gave us access to the data and information that they had collected. We owe special thanks to Maurice Pinard from McGill University, who had compiled polls on support for sovereignty since 1962 and gave us access to his compilation. This helped us greatly at the start of the project. This research was financed by the Social Sciences and Humanities Research Council (SSHRC) of Canada, grant number: 410-2009-1582. Notes 1. The data come from a search of the archives of newspapers and magazines (La Presse, Le Journal de Montreal, Le Devoir and L'Actualité) along with the Canadian Opinion Research Archive (CORA) at Queens University. Data from pollsters themselves were also collected, either through their Internet site (Léger Marketing) or by contacting them directly. Data were also collected by consulting the work of researchers. Finally, some researchers and pollsters provided access to their own data compilations. We thank Claude Gauthier at CROP, Grégoire Gollin at Createc, Jean-François Lisée, and especially Maurice Pinard from McGill University, who had compiled polls since 1962 and gave us access to his compilation. The database is available (in French) on Claire Durand's website http://www.mapageweb.umontreal.ca/durandc/souverainete/recherche_souverainete.html 2. The polls that could not be used were those whose question wording was unknown or whose results for the question on support for sovereignty were incomplete so that it was impossible to estimate the proportion of support after proportional distribution of non-disclosers. Polls whose sample size was unknown were retained. It was usually possible to infer an approximate value when data were missing, based on the usual methodological practices of the pollster. The main cause of missing data stemmed from lack of information in the publication of poll results in the media. Missing data are therefore unrelated to other survey methodology characteristics. Any removal of data did not lead to a systematic bias. 3. The term "non-disclosers" refers to all the respondents who either say that they do not know for whom they will vote or who refuse to reveal their intentions. Most polling firms do not keep or publish sufficiently detailed information to differentiate between the two. Moreover, it is estimated that part of the respondents who say that they do not know for whom they will vote are in fact people who do not want to say. 4. The data did not allow for testing effects beyond the cubic effect; it would have required more data over a longer period. 5. On this topic, see Hox (Citation2002); Snijders and Bosker (Citation1999), Singer and Willett (2003), who use the term "multilevel analysis" and Bryk and Raudenbush (Citation1992) who use the term "hierarchical models." The software used is HLM 6.0. 6. Since the amount of data for the separation and independence proposals was too small and since the two proposals generated similar support, they were merged into a single variable: extreme. 7. This is the average support for sovereignty-partnership in June 1995.