Federal Voting in Quebec_Part 2

Federal Vote Choice in Quebec 

It’s Sovereignty Stupid!


by M.Arnot, M.Campbell, E.Hjertaas, D.Moore & B.Nolan


Persuasive evidence for the impact of these three structural determinants can be found in the nature of Quebec’s cleavages in voting.  After these cleavages are introduced, the structural determinants will be employed to account for them, testing their explanatory power.

The most dramatic division in sovereignty support is language.  When examining voting behaviour in Quebec, this cleavage is strikingly large.  In the 2000 federal election, non-Francophones were 42 percentage points more likely than French speakers to cast their vote for a Liberal, while French speaking voters were an estimated 35.6 percentage points more likely to support the Bloc Quebecois (Blais et al. 2002: 95, 223)..  Study of the language cleavage has been hampered by the difficulty of attaining large enough samples for Anglophones and especially Allophones, but some trends can be seen from the limited data available.  A 2005 poll found that 62% of Francophones would vote for sovereignty, while 87% of Anglophones would vote against it.  Interestingly, 31% of Allophones are in favour of sovereignty (Léger Marketing: 4); this general trend is quite consistent over time.

Some researchers have stated that “the sense of cultural insecurity inherent in Quebec’s status as a small French-speaking island in an ocean of English leads many Francophones to consider sovereignty as a form of linguistic assurance” (Martin & Nadeau 2002: 151).   In fact, “among the most often voiced grievances [expressed by Francophones] are first those concerning their cultural life: first and foremost, worries with regard to their language, culture, and institutions” (Pinard & Hamilton 1986:235).   Unequal recognition of Francophones in Canada and the attitudes of superiority among English speakers have also been recognized as major grievances held by French speakers (Pinard & Hamilton 1986: 236-241).

Although there is not a proven causal link between opinion on sovereignty and fears for the future of the French language, there is a strong correlation. A significant number of Francophones believe that sovereignty would bring linguistic benefits; in a 1999 poll, 56% anticipated that the situation of French would improve in Quebec after separation, but only 9% thought it would worsen (Nadeau et al. 1999: 529).  Of those who believe that the situation of the French language and culture would improve with sovereignty, 74% would vote in favour, while 86% of those who believe it would worsen would be opposed (Martin & Nadeau 2002: 251).  It is known that the overwhelming majority of non-Francophones foresee negative consequences to independence, including linguistic ones (Nadeau et al. 1999: 529), and 47% indicated in 2005 that they would consider leaving the province if it became an independent country (Léger Marketing: 17)  An overwhelming majority of Anglophone Quebecers, 80%, foresee adverse ramifications for the English community in the province.  Strikingly, 70% believe the disappearance of their community to be somewhat or very likely after secession (529).  They are also in near-unanimous opposition to calls for sovereignty. In the 1995 referendum, only 5.6% of non-Francophones voted in favour, while 57.8% of the Francophone community supported sovereignty (Martin & Nadeau 2002:145).  The reason for this correlation without causation will be discussed in the next section.

Apart from language, age is the most critical socio-demographic factor in defining feelings towards sovereignty.  In fact, since the early 1980s language and age have been the only statistically strong and persistent cleavages on the question of sovereignty, far stronger than social or economic identity factors such as gender, social class, sector of employment, or educational level achieved (Martin, 1994: 347).  The slight gender cleavage can be largely explained away as a product of the life-cycle fact that women compose a larger proportion of the oldest generation (due to the greater longevity of women) which is the one least likely to be sovereignist (Martin & Nadeau, 2002: 144), the reasons for which will be explored in the paper.

In considering the dynamics that emerge from the demographic landscape of age, it is critical to understand the difference between generation and life-cycle.  Generation-based explanations turn on the concept of political socialization, which is the acquisition, at a young age, of political identities that persist through adulthood (Blais & Nadeau, 1992:146).  Environmental factors specific to a certain era can affect the political socialization of entire generations of young people.  Thus, a generational explanation of the age cleavage highlights factors unique to a generation that will remain relevant in influencing its opinions as it ages.  Life-cycle explanations, on the other hand, focus on the issues associated with being in certain stages in life, and thus should be constant in their influence on certain age-groups regardless of generation.  An important consequence that follows logically from this distinction is that a generational explanation, unlike a life-cycle explanation, has inevitable bearings on social change over time (Putnam, 2000: 248).

Whether or not life-cycle factors can cause major societal change depends entirely on the stability of the demographic makeup of the province.  Quebec experienced a precipitous drop in fertility rates – beginning in the mid 1950s and stabilizing somewhat in the 1970s (although with the end of the echo boom it has begun to drop again since the early 90s (Institut de la statistique Québec, 2001)) – constituting the most remarkable demographic development in Quebec in the 20th century (Caldwell & Fournier, 1987: 31).   From 1954 to 1967 alone, the birthrate dropped from over 4.0 to under 1.5 (Ibid.).  This has resulted in Quebec undergoing one of the most dramatic aging processes in the industrialized world (32).  Considering this, there is no reason to believe that life-cycle factors will be stable, and thus they can be expected to drive societal change in their own right.  Furthermore, these demographic facts are important to generational factors as they will serve to either amplify or mitigate the influence of given cohorts.  Ultimately, it is important to consider demographics in order to understand the constitution of the electorate, especially now as the pre-’54 generation (who are now over 50) move out of the majority, and as their children (the echo-boom) enter political society.

The third critical factor is region. Essentially, a region is an area where a significant degree of homogeneity on a given factor can be expected in the population because of the interplay of various shared characteristics. To facilitate studies of voting behaviour, region can be classified as the geographic area encompassed in the boundaries of an electoral riding, or groups of electoral ridings, though this classification will inevitably be flawed.  Sub-provincial regions are usually characterized as groups of ridings focused on sections of cities, suburbs, or areas with particular characteristics, like closeness to a border, or similar rural settlement patterns. The geographic location of a constituency determines distance from population centers; if a constituency is distant from a main center, a feeling of isolation or marginalization often surfaces. The distribution of natural resources affects the kind of economic activities that exist within the region. Urban or rural character is also important, because rural ridings have less population and slower economic growth than cities, and thus tend to be more conservative (Carty and Eagles, 2005: 8). Socio-economic factors can be included in a regional analysis.  These factors are heavily influenced by settlement patterns, as well as political factors due to the historical legacy of previous elections and political activities (9).

In terms of electoral support for each federal party, it is difficult to differentiate between regional and linguistic causality. In 1993, the strongest Liberal vote was confined to mostly Anglophone and Allophone ridings in Montreal, and strongly Anglophone ridings along the borders. The Bloc Québécois won 54 ridings, essentially, all of predominantly French-speaking Quebec (Drouilly, 1997: 47), which may be due in part to massive popularity of Lucien Bouchard as leader of the Bloc. The Conservative vote in 1993 collapsed across the country, so it is difficult to draw any conclusions. The one Conservative seat went to Jean Charest in Sherbrooke (a seat which later went Bloc), but this seems to be more as a result of the candidate than any regional effects. The NDP also fared terribly in Quebec that year, receiving only 1.5% of the vote, with its highest percentage (4.5%) in Outremont (48). In the 1993 election, Quebec seems to have been polarized between the Liberals and the Bloc.  These results reflect a “race vote,” as Pierre Drouilly calls it: an almost unanimous rejection of the idea of Quebec sovereignty by non-Francophones (138) (the notable divide within this group between Anglophones ans Allophones was discussed above). This usually translates into support for the Liberals as the only major alternative to the Bloc, but the Conservative party occasionally receives a strong federalist vote in Quebec. However, the idea of a “race vote” does not completely discount the idea of region as an influential factor. Francophone support for sovereignty is far from unanimous, even in the most strongly Francophone and BQ ridings.

In the 1995 referendum on sovereignty, the Francophone “yes” vote remained steady at about 60% unless the district was less than half Francophone, in which case it fell below 50% (293 Table 9).  In metropolitan Québec, 96.6% of residents are Francophones, the highest proportion in Quebec. If language were the only salient issue, metropolitan Quebec should also have voted “yes” in the highest proportion. This is not the case, with the overall “yes” vote at 55% and Francophone “yes” vote at 57%. This was the second lowest level of Francophone support in the province. The highest level of sovereignty support was in “la couronne de Montreal,” with 56.3% overall, and 65.2% among the 86.4% of its population that is Francophone. The district with the highest “yes” vote among Francophones is Montreal East, with 66.7%. Only 83% of its residents are Francophones and the overall “yes” vote was 55.4% (294 Table 10).  Language variations alone can not justify the wide disparities in Francophone support for sovereignty by region.

The contact hypothesis may help to explain these differences.  It holds that “Francophones in mixed ridings may have adjusted to the stresses of living between two worlds already, and may have economic or social ties to the Canadian majority,” which changes the nature of national identity and moderates perception of threat.  Following from this logic is that “those living in heavily Francophone areas, meanwhile, may feel besieged [by a separatist message] without the mitigating effect of cross-cultural contact” (Lublin and Voss, 2002: 92).  As will be discussed below, this phenomenon most likely has an important impact on national identity.

There is no significant denominational religious cleavage in Quebec voting behaviour. On the national scale, Catholics are inclined to support the Liberal Party, while Protestants, at least historically, have tended to divide their vote more or less evenly between the Liberals and the Conservatives (Kanji & Archer, 2002: 168). This does not occur in Quebec because of the Quiet Revolution and the rise of Quebec nationalism, which served to diminish the salience of the Roman Catholic Church in Quebec politics.  It is more probable that a religious cleavage exists based on religiosity, distinguishing religious voters from secular voters rather than voters of different denominations.  Beyond simply that Quebec “is distinguished” by a relatively low degree of religiosity (Blais, et al., 2002: 104) there is limited available research, forestalling its incorporation into our discussion, although it should certainly be explored in the future.

There is a strong body of evidence linking political knowledge to education (Fournier, 2002: 349; Lambert et al: 356).   Of all factors that contribute to differences in awareness, the most important is education (Gidengil et al., 2004: 50). Johnston et al. examined Quebecers regarding their support for the 1992 Charlottetown Accord. The group found that more educated voters were more likely to support the Accord, however “with the passage of time, differences … [in] levels of education … have decreased as determinants of sovereignty support” (Martin & Nadeau, 2002: 9).  In the 1995 Referendum, 44% of voters with a university degree supported sovereignty, as opposed to groups with secondary and some post-secondary, 52% of whom supported sovereignty.  Education is not a significant cleavage, though there is a clear pattern in sovereignty support. The median-educated Quebecer (secondary or some post-secondary) is slightly more likely to support sovereignty, however both lower and higher educated cohorts are 8-10 percentage points less likely to do so.  Blais and Nadeau have found that rather than increasing support for sovereignty, “increased levels of formal education in Quebec society might contribute to high levels of interest” (25).

In the 1995 Referendum, 44% of voters with a university degree supported sovereignty, as opposed to groups with secondary and some post-secondary, 52% of whom supported sovereignty.  Education is not a significant cleavage, though there is a clear pattern in sovereignty support. The median-educated Quebecer (secondary or some post-secondary) is slightly more likely to support sovereignty, however both lower and higher educated cohorts are 8-10 percentage points less likely to do so.

. There is a cleavage in sovereignty support based on ideology.  Sovereignists tend to be more “left” than non-sovereignists, and the movement itself found its origins in leftist ideology, eventually developing post-materialist justifications (this was discussed earlier).  Interestingly, the leftist composition of the movement is not based solely on the attraction it holds for those who share the same ideological leanings.  In the 1993 election, sovereignty was so important that sovereignist voters aligned themselves ideologically with the leftist sovereignty movement.  Their “positions on other issues came to cleave along one’s position on sovereignty” (Hinich et al., 1997).  The ideological cleavage in the movement appears to be mainly related to its ability to realign voters to match its leftist origins.

Significant cleavages in opinion on the sovereignty issue based on social class have faded nearly completely (Martin, 1994: 347).  During the industrialization of the post-war years the recognition of differences in material wealth and control over the means of production in Quebec between the English and the French populations shaped the proto- sovereignist movement (Martin & Nadeau, 2002: 148).  Economic inequality diminished from the 60s to the 80s, and with it the importance of social-class as a “sovereignist” identity component faded to the extent that now there is no statistically significant cleavage at all.

Applying the Structural Determinants to Cleavages

There are thus three main cleavages – language, age, and region – which can be explained using different combinations of the three structural determinants of sovereignty support.  The determinants’ explanatory power supports their validity.

Unequal respect and recognition

Recent studies have found that fears for the future of the French language are, in fact, not one of the major determinants of sovereignty support in Quebec. (Mendelsohn 2003: 525).  When all other variables are held equal, an individual who perceives a threat to the French language and believes that sovereignty could improve this problem does not have a much higher level of support for independence (525).  Mendelsohn’s explanation is that, in his study, only 33% of Francophones believed that sovereignty could improve their language’s situation, even though 56% of them thought it was threatened (528).  This large decrease from the levels of only a decade ago, according to the author, shows that now “for most Quebecers the threat to French comes not from Canada but elsewhere” (528).  An increased awareness of the results of globalization is a plausible recent change which seems to have decreased the impact of language concerns.

Instead, the linguistic differences in voting between Francophones, Anglophones, and Allophones occur because of the different feelings, identities, and interests that the groups possess.  These differences are manifested in all three of the determinants of the sovereignty support, including the first, which is the Francophone perception of unequal respect and recognition from English Canadians.  Low levels of Anglophone support for sovereignty are then partially explained by the mainly Francophone nature of this grievance (Pinard & Hamilton 1986:235).  For Allophones, Bill 101’s imposition of the French language may have resulted in an increased sympathy for this concern among this population as well.

There is a large amount of statistical support for the existence and influence of this feeling of unequal treatment.  In a 1980 survey, 54% of Francophones agreed with the statement that “English Canadians often tend to consider the French Canadians to be inferior to them” (238), and this has continued to be a prevalent opinion (Mendelsohn 2003: 527).  Francophones with such feelings were found to be more than three times more likely to vote for sovereignty than against it in the 1980 referendum, while individuals with very low levels showed a pattern that was almost the exact opposite (Pinard & Hamilton 1986: 245).  The findings pointed to the conclusion that the feelings of unequal respect and recognition are necessary for the Quebec independence movement and an important factor in recruitment (241). The perception of unequal respect and recognition is prevalent and influential in Quebec’s Francophone population, but not a factor for Anglophones.   This is a partial explanation for the linguistic cleavage.

National identity

 Language is also highly correlated with national identity, the second determinant.[1] Researchers have noted that in the post-war era the increasingly Quebec-oriented identity of Francophone Quebecers become rooted in language (Martin & Nadeau 2002: 146).  Identity has a clear correlation with separatist beliefs: in a survey of approximately 900 Francophone Quebecers, it was found that individuals with an exclusive Quebec identity (and therefore little to no attachment to Canada) had a 93% chance of being sovereignist, while those who favoured Canada had only a 25% chance of supporting sovereignty (94).  The authors concluded that “the great majority of Francophone Quebecers define themselves as Quebecers first and this pushes them toward the sovereignty option” (97).  When 68% of Francophone Quebecers identify more with their province, and only 6% have a stronger attachment to Canada, identity will play a “crucial” role in determining support for independence (91, 95).

In 1995, only 7% of Anglophones primarily identified with Quebec (and therefore had a reduced level of attachment to Canada) resulting in weaker forces pulling them towards independence.  The reason for this linguistic difference in identity is clear when one considers the linguistic basis of Quebec national identity that was mentioned above.   Once again, the lukewarm Allophone support for sovereignty seems to originate in feelings which are in between the two other linguistic groups.  In a population with many recent immigrants it is hardly surprising that they are more likely than English-speakers to be proud to be Canadian, but they are also more likely than Anglophones to be proud to be Quebecois (Léger Marketing, 2005 3).  This is an admittedly crude indicator of national identity, but it does at least show that their national identity is probably more favourable towards sovereignty than that of Anglophones.

National identify also provides a powerful account for the age cleavage.  Over 75% of those who identify primarily as Québécois voted “Yes” in the 1995 referendum, obviously a powerful indicator of sovereignist views (Martin & Nadeau, 2002: 145).  As has been shown, such identification is localized in the boomer and younger generations, but to consider only this is to leave a major question unanswered.  Many people, especially in the post-boom generation (Gen-X) do not support the sovereignty movement despite a Québécois identity.  What explains that?  Furthermore, statistical evidence suggests that identity only partially explains the apprehension of the oldest generations towards sovereignty, what other factors are relevant?

Finally, we can partially account for the regional cleavage through the determinant of national identity.  The seventeen ridings where francophone support is lower than 50% are found in the West Island, Outaouais and Beauce. Montreal’s West Island has many Anglophones and allophones, supporting the idea that contact between linguistic groups mitigates perceived identity conflict.  Since national identity is strengthened by such conflicts, contact between linguistic groups reduces support for sovereignty in such areas.  Outaouais and Beauce are excellent examples of the ability of economic ties to alter national identity; in the case of these two regions decreasing sovereignty support.  Economic considerations also serve as an independent determinant in voter decisions.

Rational calculations of the economic consequences of sovereignty

Differences in evaluations of the economic consequences of sovereignty between Francophones, Anglophones, and Allophones are the final element that accounts for the language cleavage.  Since this has been found to be the most important factor in determining support for sovereignty among those with a weak Quebec national identity, even small differences in opinion can have a major impact.  It has been found that although 49% of the population as a whole believes that the economic situation of a sovereign Quebec would worsen, this number is only 44% in the Francophone majority.  That the non-Francophone minority could raise the average by 5 percentage points indicates a sizeable difference in opinion.  Even more dramatically, while a near-unanimous 95% of Anglophones foresee an economic crisis after a vote for sovereignty, only 57% of Francophones share this opinion (Nadeau et al. 1999:529).  It is likely that this difference in perception between the two groups is evidence of egocentric economic thinking. Anglophones are inclined to believe that they would be economically disadvantaged in a sovereign Quebec: 70% believe that the disappearance of their linguistic community is likely if the movement should succeed (529).  It also seems logical that Allophones, as a result of Bill 101, would feel better-equipped to operate economically in a sovereign and presumably more French Quebec.

Turning now to the age cleavage, the generational explanation is that economic risk averseness (the motivating power of fear of economic loss) varies depending on the economic circumstances in which a generation is socialized.  A low risk-averseness is the product of – and synonymous with – the development of post-materialist values (Inglehart, 1971: 991), which, research suggests, are more prevalent among militant sovereignists than federalists (Martin & Nadeau, 2002:148).  The theory is basically that generations socialized in harsher economic circumstances will be relatively more materialist than generations socialized in easier economic circumstances (Martin, 1994: 346).

This leads us to consider the different economic circumstances in which successive generations were socialized in Quebec.  As was mentioned earlier, the majority French population in Quebec in the first half of the 20th century was primarily rural, and very poor.  Rapid economic development took place following the war, reaching a fever pitch between the mid 1950s and mid 1960s.  At that point it peaked and then slowed down through the 70s and 80s (though only modestly when considered relative to the economic deprivation of the period prior to the boom, remaining relatively stable to the present) (350-1).

We would thus expect, according to this argument, that the pre-boom generation would be the most materialistic, followed distantly by generation X, with the echo-boomers nearly as post-materialistic as the boomers themselves.  It is important to note that there is a huge disparity between the poverty of the pre-Silent Revolution era (which was marked by the depression and the war years), and that of the years in which generation X was socialized.

How does this correspond to the data?  Survey data from 1991 (before the larger part of the echo-boomers entered the electorate) found that, controlling for identity (the psycho-social explanation), the pre-boomers were far more affected by concern with possible economic consequences of separation than either baby-boomers or post-baby boomers (Martin, 1994: 355).  The difference between the boomers and post-boomers was far smaller, with the post-boomers somewhat more affected (Ibid.).  This is generally supportive of the argument, although it is important to note that with only one survey it is impossible to establish conclusively that the phenomenon in question can not be due to life-cycle.  There is a correlation with the respective generations’ level of support of sovereignty, but that does not prove causation, leaving the door open. As such, more research needs to be done to establish this hypothesis conclusively.

Turning to these other factors, it is interesting to note that for all three groups, the negative correlation between those people concerned with a serious economic downturn and who support sovereignty is statistically significant.  Conversely, the correlation that emerges from looking at the beliefs of those who foresee a positive economic impact from sovereignty is only statistically significant for the post-boomers (356).  This is the opposite effect of what the theory would have you believe.  It suggests a life-cycle explanation deriving from the fairly widely held belief among Quebecers that separation could be promising economically in the long term (in 1995 48% expected gains in the long term (Martin & Nadeau, 2002: 152).  Only post-boomers could really expect to benefit from a long-term improvement or even recovery in the economy (Martin, 1994: 351).

Additionally, it is very theoretically plausible that younger people would generally have more confidence in their ability to persist in difficult economic conditions, being more mobile (for example, due to a higher rate of people without dependents) and resilient.  Applying the same logic older people (who are dependent upon retirement savings or government pensions (Blais & Nadeau, 1992: 93)), would have far less confidence in their own ability to “pull themselves up by their bootstraps” in the event of a serious economic recession, drastically increasing their risk aversion.  This is logically an equally viable explanation of the 1991 survey results to what is put forward in the generational argument.  The test of it will be whether or not the boomers’ risk aversion will increase as they hit retirement (a process that is beginning now).

Differences in regional voting can also be partially attributed to economic considerations, as was touched on in the previous section.  Outaouais is located along the Ontario border near Ottawa, so the low level of sovereignty support is due in part to the nature of employment within the region. As 25% of the population works for the federal government (Monière & Guay, 1996: 241), they fear the loss of their jobs in a separate Quebec. This is due directly to the proximity of this area to Ottawa (229). Finally, Beauce, an overwhelmingly francophone area, provides another exception. It is mostly rural and borders on the American state of Maine. Beauce and the other border regions are less supportive of an independent Quebec because they fear the loss of benefits they derive from free trade agreements when these are renegotiated (242), as well as a change in their position relative to their neighbours.




Archer, Keith, “Voting Theory and Its Applicability in Canada” in Joanna Everitt and

Brenda O’Neill (eds.), Citizen Politics: Research and Theory in Canadian Political Behaviour. Agincourt: Methuen, 1985, 29-46.

Blais, André, Elisabeth Gidengil, Richard Nadeau, Neil Nevitte. Anatomy of a

Liberal Victory: Making Sense of the Vote in the 2000 Canadian Election. Ontario: Broadview Press, 2002.

—-. “It’s Unemployment, Stupid! Why Perceptions about the Job Situation Hurt the Liberals in the 1997 Election”, Canadian Public Policy – Analsyse de Politiques, VO. XXVI, no. 1, 2000.

Blais, André, Richard Nadeau. “Explaining Election Outcomes in Canada: Economy

and Politics.” Canadian Journal of Political Science, 1993, 26[4]: 775-90.

—-.“To Be or Not To Be a Sovereignist: Quebeckers’

Perennial Dilemma.” Canadian Public Policy, 18, 1992: 89-103.

Blais, André, Nadeau, Gidengil, Nevitte. “Measuring strategic voting in multiparty

plurality elections”. Electoral Studies, 2001.

Brown, Steven D; Ronald D. Lambert; Barry J. Kay; James E. Curtis. “In the eye of the

Beholder: Leader Images in Canada.” Canadian Journal of Political Science, Vol. 21, No. 4. Dec. 1988, 729-755.

Bryden, Joan. “Liberals Would Fare Best with Rae, Poll Finds.” The Globe and Mail. 18 December 2006: A11

Cairns, Alan C. “An Election to Be Remembered: Canada 1993”. Canadian Public

Policy – Analyse de Politiques, Vol. 20, No. 3. Sept, 1994, 219-234.

Caldwell, Gary, Daniel Fournier. “The Quebec Question: A Matter of Population.”

Canadian Journal of Sociology/Cahiers canadiens de sociologie, Vol. 12, No. 1/2 Spring 1987, 16-41.

Campbell, Angus, Philip E. Converse, Warren E. Miller & Donald E. Stokes. “The American Voter.” In Sylvia Bashevkin (ed.), Canadian Political Behaviour, (Agincourt: Methuen, 1985), 29-46.

Carty, R. Kenneth, Munroe Eagles. “Politics is Local: National Politics at the

Grassroots”. Toronto: Oxford University Press, 2005.

Clark, Campbell; Laghi, Brian. “Dion Liberals Jump in Poll.” The Globe and Mail. 4 December 2006: A1

Clarke, Harold D; Allan Kornberg. John MacLeod, and Thomas Scotto. “Too Close to

Call: Political Choice in Canada 2004.” PS: Political Science and Politics, Vol. 38, No. 2. Apr., 2005, 247-253.

Clarke, Harold D; Allan Kornberg. “Choosing Canada? The 1995 Québec Sovereignty

Referendum.” PS: Political Science and Politics, Vol. 29, No. 4. Dec., 1996, 676-682.

Clarke, Harold D, Jane Jenson, Lawrence LeDuc, and Jon H. Pammett. “Political

Choice in Canada”. Toronto: McGraw-Hill Ryerson, 1982: 524.

Cutler, Fred. “The Simplest Shortcut of All: Socio-demographic Characteristics and

Electoral  Choice.” The Journal of Politics, Vol. 64, No. 2. May, 2002, 466-490.

Drouilly, Pierre. Indépendence et Démocratie: sondages, éléctions et référendums au

Québec 1992-1997”. Montreal: Harmattan Inc., 1997.

Department of Finance Canada.  “Economic Union Impossible and Access to NAFTA

Arduous: The Minister of Finance Slashes at the Utopia of Separation.” Department of Finance News Release, 26 September 1995.  Accessed 16 December 2006 at: http://www.fin.gc.ca/news95/95-073e.html.

Fournier, Patrick, “The Uninformed Canadian Voter” in Joanna Everitt and Brenda O’Neill (eds.), Citizen Politics: Research and Theory in Canadian Political Behaviour. Toronto: Oxford University Press, 2002, 92-109

Gidengil, Elisabeth, André Blais, Neil Nevitte, Richard Nadeau, “Citizens” (Vancouver: UBC Press, 2004), Chapters 3, 41-71.

Gidengil, Elisabeth, André Blais, Neil Nevitte and Richard Nadeau. “Are Party Leaders

becoming more Important to Voting Choice in Canada?” Canadian Journal of Political Science, 2000, 1-24.

Hinich, Melvin J., Michael C. Munger & Scott de Marchi. “Ideology and the Construction of Nationality: The Canadian Elections of 1993”. Public Choice, Vol. 97, No. 3. Dec. 1998, 401-428.

Howe, Paul. “Rationality and Sovereignty Support in Quebec”. Canadian Journal of

Political Science/Revue canadienne de science politique, Vol. 31, No. 1. Mar, 1998, 31-59.

Inglehart, Ronald. “The Silent Revolution in Europe: Intergenerational Change in Post-

Industrial Societies.” American Political Science Review, 65 1971: 991-1017.

Jenkins, Richard W., “Media, Voters and Election Campaigns: The Reform party and the 1993 Election” in Joanna Everitt and Brenda O’Neill (eds.), Citizen Politics: Research and Theory in Canadian Political Behaviour (Toronto: Oxford University Press, 2002), pp. 215-229.

Jenkins, Richard W., “Media, Voters and Election Campaigns: The Reform party and the

1993 Election” in Joanna Everitt and Brenda O’Neill (eds.), Citizen Politics: Research and Theory in Canadian Political Behaviour. Toronto: Oxford University Press, 2002, 215-229.

Kanji, M., & K. Archer. “The Theories of Voting and Their Applicability in

Canada.” In Citizen Politics: Research and Theory in Canadian Political Behaviour, eds. J. Everitt & B. O’Neill. Toronto: Oxford, 2002.

Lanoue, David J. “Debates that mattered: Voters’ reaction to the 1984 Canadian

Leadership Debates.” Canadian Journal of Political Science. Vol. 24, No. 1. Mar. 1991, 51-65.

Léger Marketing, 2005. “Quebec Survey.” [Online edition] Published on May 14, 2005. Accessed on December 16, 2006.

—-. “Referendum Voting Intentions.” April 2006. Accessed on 18 December 2006. http://www.legermarketing.com/eng/intref.asp?prov=QC&l=1

Lublin, David and D. Stephen Voss. “Context and Francophone Support for the

Sovereignty of Quebec: An Ecological analysis.” Canadian Journal of Political Science 35:1. March, 2002, 75-101.

Martin, Pierre, Richard Nadeau. Choosing a Nation: The 1995 Referendum on

sovereignty in Quebec, Conference: Making Big Choices: Individual Opinion Formation and Societal Choice, Harvard University, 2000, 9.

Martin, Pierre. “Génerations politiques, rationalité économique et appui a la souveraineté

au Québec.” Canadian Journal of Political Science, 27, 1994, 345-59.

Martin, Pierre, Nadeau, Richard.  “Understanding Opinion Formation on Quebec

Sovereignty.” Citizen Politics: Research and Theory in Canadian Political Behaviour. Eds. Joanna Everitt, Brenda O’Neill. Ontario: Oxford University Press, 2002.

Meadwell, Hudson. “The Politics of Nationalism in Quebec.” World Politics, Vol. 45,

No. 2, Jan. 1993: 203-241.

Mendelsohn, Matthew. “Rational Choice and Socio-Psychological Explanation for

Opinion on Quebec Sovereignty” Canadian Journal of Political Science, 36:3, 2003, 511-537.

Mendelsohn, Matthew, Nadeau, Richard, “The Rise and Fall of Emerging Candidates”

Harvard International Journal of Press/Politics, 1999, 4: 63-76.

Merolla, Jennifer, Laura Stephenson. “Strategic Voting in Canada: A Cross Time

Analysis”. Claremont University and University of Western Ontario. 2001. California, Ontario.

Monière, Denis, Jean H. Guay. “La Bataille du Québec: Troisième épisode: 30 jours

qui ébranlèrent le Canada.” Quebec: Édition Fides, 1996.

Nadeau, André, Martin 1999 (Negative consequences of independence??)??

Pinard, Maurice, Richard Hamilton. “Motivational Dimensions in the Quebec

Independence Movement: A Test of a New Model.” Research in Social Movements, Conflicts and Change, 9, 1986: 225-80.

Putnam, Robert. Bowling Alone. New York: Simon Schuster, 2000.

Schneider, Barbara L.  Stevenson, David. “The Ambitious Generation.” New Haven:

Yale University Press, 1999.

Shaver, Andréa. “Teen Suicide.” Political and Social Affairs Division of the Government

of Canada.  August 1990.  Accessed on October 4th, 2006, at: http://dsp-psd.pwgsc.gc.ca/Collection-R/LoPBdP/BP/bp236-e.htm.

Stacey, Barrie. “Inter-Generation Mobility and Voting.” The Public Opinion Quarterly,

Vol. 30, No. 1, Spring 1966, 133-39.

Turcotte, André. “Fallen Heroes: Leaders and Voters in the 2000 Canadian Federal

Election.” in Jon H. Pammett and Christopher Dornan (eds.), The Canadian General Election of 2000. Toronto: Dundurn, 2001, 227-92.

“Younger Population By Age Group.” Quebec 2001 Population Census.  Quebec :

Institut de la statistique Québec, 2001. Accessed on October 3rd, 2006, at: http://www.stat.gouv.qc.ca/regions/lequebec/population_que/popjeune20_an.htm

[1] Remember that, in this instance, national identity refers to an individual’s level of attachment to Canada.  The variable represents the difference between those who identify with Canada, and therefore have a dual Canada-Quebec identity, and those with a mainly Quebec identity.