Strengthening Canadian Democracy

Civic Literacy in Comparative Context

Why Canadians Should Be Concerned

Henry Milner July 1, 2001


In last November’s federal election, just over 60 percent of potential voters made it to the polls. There are many reasons for this, but one, too often neglected, is declining levels of political knowledge. Can we expect people who do not keep up with political events to vote?

This is the question addressed in this paper. Using data from a series of indicators, it develops the concept of “civic literacy” as a framework for comparing societies’ capacity for informed political participation. The main part of the paper sets the Canadian situation into a wider, international context; it then goes on to identify a number of policy initiatives needed to raise civic literacy in Canada. These include efforts to enhance newspaper reading and to improve adult and civics education. Our political institutions should also be reformed in ways that will encourage political actors to engage in activities aimed at educating citizens about political matters. Given the nature of Canada and its constitution, much of the initiative in these areas must be taken at the provincial level.

The civic literacy approach developed in the paper is contrasted with the prevailing current in the contemporary literature on declining political participation, which revolves around the concept of social capital, as developed notably by Robert Putnam. It is argued that social capital does a poor job of explaining differences in levels of voter turnout across countries or its recent decline. Canada, for example, places seventh in the World Values Survey in social trust and eighth in membership in various kinds of voluntary groups – two key measures of social capital – but only ranks above the US and Switzerland in voter turnout. Civic literacy, on the other hand, shows a high level of correlation with voter turnout.

Finally, the effects of low civic literacy are not neutral. In low civic literacy societies like the United States, the interests of the economically disadvantaged carry less weight since they are often excluded from informed political participation through lack of civic competence. In contrast, high civic-literacy societies such as the Scandinavian countries tend to adopt policies that take into account the full range of interests in society because they encourage political participation across all sectors of society. The end result is more egalitarian economic and social policies. Given current levels of civic literacy in Canada, there is reason to be concerned that we are more likely to follow the American than the Scandinavian path in the coming years.

There is nothing I want more than to be a teacher and maybe create more people like me who recognize the importance of taking responsibility for the world…. I want to see more people voting…. I don’t read the newspapers, I don’t watch the news. I figure, if something important happens, someone will tell me.

Justin Trudeau in The Globe and Mail, February 3, 2001


Is it possible to be a good citizen without being politically informed? Can we expect people who do not keep up with political events to vote? Turnout in Canada’s most recent elections is certainly not reassuring. We need to know more about the relationship between what Canadians know about politics and their levels of political participation.

This is the question addressed in this paper. To begin to answer it,1 I bring to bear data from a series of indicators, developing the concept of “civic literacy” as a framework for comparing societies’ capacity for informed political participation. The main part of the paper sets the Canadian situation into a wider, international context; it then goes on to identify a number of policy initiatives needed to raise civic literacy in Canada.

This approach differs in important respects from the main current in the contemporary literature on declining political participation, which revolves around the concept of social capital, as developed notably by Robert Putnam.2 I argue that Putnam’s conceptualization is not all that helpful in explaining differences in levels of voter turnout across countries or its recent decline in Canada. For Putnam, voting is one of a wider set of activities and attitudes that have been negatively affected by a general decline in the participatory proclivities of American society. His Social Capital Index used to compare US states is composed of fourteen indicators that include participation and membership in various kinds of associations, level of social trust, and two “measures of informal sociability”: “spend a lot of time visiting friends”; and “average number of times entertained at home in the last year” – as well as voter turnout in presidential elections.3 When applied internationally, the indicator of social capital typically used combines associational participation and levels of social trust, the data for which are derived from the World Values Survey (WVS).

In the 1990 WVS, Canadians placed a respectable seventh in the proportion agreeing that “most people can be trusted” (after the Scandinavians, Dutch and Americans), and eighth in membership in various kinds of voluntary groups (after the Scandinavians, Dutch, Americans and Germans). Yet the story is very different when we look at political participation. In a comparative analysis of the turnout of potential voters in the 1990s,4 Canada placed above only France, Japan, the US and Switzerland. And the situation has deteriorated since. Canada’s turnout among eligible voters declined by 2.5 percent in 1997 and by another 6 percent in 2000, leaving Canada above only the US and Switzerland, and prompting Canada’s Chief Electoral Officer to think out loud about adopting compulsory voting (as in Australia, Belgium, Greece and Italy.)

Hence, though Putnam decries the decline of social capital in the United States in recent decades, the reality for both Canada and the US is that, comparatively speaking, social capital is markedly higher than voter turnout. This anomaly is due, I contend, to the fact that social capital, as conceived by Putnam, is not directly related to political participation. The more crucial variable is what I term civic literacy – simply put, the knowledge to be effective citizens. Many of the kinds of activities associated with social capital do not contribute to civic literacy and, hence, do not enhance political participation. Indeed, as argued below, enhancing social capital is, by Putnam’s reasoning, primarily a societal concern. The promotion of civic literacy, on the other hand, is partly a societal responsibility, but is also, to an important degree, influenced by a country’s political institutions and the opportunities and incentives these provide for political actors to engage in the educational activities needed to nurture and sustain an informed polity.

The conclusion in short: democracy is, to transpose Putnam’s contention about social capital, stronger in a community blessed with a substantial stock of civic literacy. Stronger in the sense that levels of political participation are higher; stronger, too, in the sense that policy decisions are more likely to take into account the full gamut of interests and perspectives present in society. As discussed in the concluding section below, since informed individuals can better identify the effects policy options have upon their own interests and those of others in their community, high civic-literacy societies are more likely to attain long-term egalitarian economic outcomes. In low civic literacy societies like the United States, the interests of the economically disadvantaged carry less weight since they are more often excluded from informed political participation through lack of civic competence. Given current levels of civic literacy in Canada, there is reason to be concerned that a similar pattern may soon emerge in this country.

What is Civic Literacy and Why it Matters

The expression “civic literacy” encapsulates the closely linked concepts of “civic engagement” (a key component of social capital for Putnam) and “literacy” or political knowledge. Unfortunately, we do not have the kind of systematic data needed to directly compare countries as to their level of political knowledge. The ideal indicator would be a direct one: that is, one based on the ratings of a large number of countries on a battery of similar or identical questions that, together, allow for a comparison of average levels of political knowledge. Unfortunately, such a scale is not to be found.5

Nevertheless, data have been accumulating about what individuals, especially Americans, know about politics, and that is a good place to start. The result of systematic gauging of public knowledge of American political institutions, processes and leaders has been, as one observer summarized the situation, that “the political ignorance of the American voter is one of the best documented data [findings] in political science.”6 Of the many surveys that illustrate this, the most authoritative is the National Election Survey (NES), which for many years has been asking, among other things, which party is more likely to favour a strong government.7 Only in 1964 did more than half the respondents choose the Democrats (as opposed to the Republicans or “no difference”).

There is far less Canadian data, but what does exist is not especially comforting. There appears to have been a significant decline in the level of political knowledge in the last 10 to 15 years. In the 1984 Canadian Election Study (CES), close to 90 percent of Canadians could identify their provincial premier.8 This went down to 77 percent in the 1997 Canadian Election Study. The 1997 CES posed three additional knowledge questions, asking respondents to identify the finance minister (37 percent correct), the US president (84 percent correct), and Canada’s only woman prime minister (41 percent). Only 21 percent could answer all 4 (27 percent answered 3 correctly, 28 answered 2, and 16 answered 1).9 This decline, according to Howe, is likely to continue, especially given the comparative ignorance of young Canadians.

In a 1990 survey carried out for the Royal Commission on Electoral Reform and Party Financing, 5 percent of Canadians could not name the Prime Minister. In a similar survey commissioned by IRPP in March 2000, this had grown to 11 percent…. In 1990, 56 percent of 18 to 29 year olds were able to answer at most one of three political knowledge questions correctly (Who is the PM? Who is the Liberal leader? Who is the NDP leader?). For the survey sample as a whole, the figure was 16 points lower at 40 percent. By 2000, the younger group was lagging further still: when asked to identify the PM, finance minister and official opposition, fully 67 percent of 18 to 29 year olds scored no more than one out of three compared to 46 percent for the sample as a whole.10

Other data sources point to similar disquieting conclusions. A 1997 citizenship survey reported in The Globe and Mail found that 45 percent of respondents failed to answer at least 12 of the 20 basic questions similar to those on the citizenship test given to immigrants (only 62 percent could name the three levels of government). The results led a prominent journalist to conclude that “A large segment of Canadian society lacks much of the civic knowledge required to understand and effectively participate in the country’s political life…[They are] cut off from media stories, political debates and community discussions…[and] lacking the shared … context to interpret the aspirations of their fellow citizens,”11

If participation in political life requires some minimal knowledge of one’s political environment, it is telling that 28 percent of those who sent in the 1997 CES post-electoral questionnaire could not identify a single candidate in their constituency. While this may seem to be fairly low, methodological factors may have biased the result somewhat. First, one expects higher rates of correct responses from mailed-in ballots as it gives respondents time and opportunity to find out the correct answer. In addition, we can assume that less than 72 percent of the answers were correct. We do not know the precise proportion since, unlike other countries in the Comparative Study of Electoral Systems (CSES), the CES did not determine whether the names entered were correct. Because of this, Canada is absent from a recent report which placed New Zealand first (83 percent), followed by Germany (76), Norway (69), Britain (60), Australia (58) and the US (52) in the proportion of respondents able to identify correctly a candidate in the last election.12

Of course, a single question of this kind invariably places some countries at a relative advantage (e.g. unitary New Zealand) or disadvantage (e.g. Norway, which has only multi-member districts). Still, the rankings are corroborated by various comparative studies, though these are typically incomplete in terms of the number of countries involved and/or the type of questions posed. A good example is a study conducted in January 1994 by the Times-Mirror Center for the People and the Press using representative samples from five countries. The differences in responses to the five questions about international affairs13 are striking: the Canadian respondents (with an average of 1.90 correct answers) placed above the US (with an average of 1.67 right answers), but well below the others. The British averaged 2.07, and the French 2.13. The German respondents (with an average of 3.58 right answers) were far and away the most knowledgeable.14

The relative performance of the three English-speaking countries in the Times-Mirror test corresponds to those found in a study of domestic political knowledge based on compilations of questionnaires asking Canadian, American and British samples a number of questions testing knowledge related to their countries’ legislatures.15 In another study limited to first-year university students, the British again outclassed their American and Canadian counterparts. On questions about past and present political leaders, the average percentage correct was 25 for both the US and Canada; for Britain it was 42. In ranking foreign countries on the basis of population and GDP, the three countries’ scores were 43, 45 and 72, respectively.16

Canada was also a participant in the international study with a relatively large number of comparable countries that first drew public attention to important differences in levels of knowledge relevant to civic literacy, namely that conducted for the National Geographic magazine in the late 1980s.17 Of the respondents from the 10 countries, Canadians placed in the middle in their ability to locate correctly 16 places on the world map. Sweden led with 11.6 average right answers, followed by Germany with 11.2. Then came Japan, France and Canada with 9.6, 9.3, and 9.2 respectively, followed by the US, UK and Italy with 8.6, 8.5 and 7.6. Mexico and Russia came last at 7.4.

A survey using a larger number of countries and skewed toward international politics showed similar results – except that Canada fared worse. The survey tested respondents’ ability to identify the United Nations Secretary General (from a list of 5) and to name a UN agency.18 Figure 1, which compares the combined proportions of respondents correctly answering both questions, illustrates that the results correspond reasonably well to the impressions gained from the above studies. The small countries of Northern Europe (along with Austria) had the highest awareness, followed by Germany, France, Italy, Australia and the UK. Canada, Ireland and, finally, the United States brought up the rear. While knowledge of international affairs is not an unbiased reflection of civic literacy, it is not an entirely distorted one either.

Since the capacity to comprehend written information is a sine qua non of political knowledge, levels of civic literacy are almost certainly influenced by levels of literacy more generally. There is, then, much to be learned from the results of a highly sophisticated cognitive proficiency test developed by Statistics Canada and administered by the OECD during the 1990s in twenty countries. The International Adult Literacy Survey (IALS) tests the comprehension of a large sample of the population (aged 16 and over) on three types of written materials: narrative prose; documents, such as train schedules and medication instructions; and problems requiring application of basic arithmetic skills.19

In what follows, the results for the fourteen mature democracies among the 20 IALS countries are included.20 Overall, Canada places in the middle. On the prose literacy scale, Canada ranks 5th behind Sweden, Finland, Norway and the Netherlands (with the United States and the United Kingdom 10th and 13th). On the document literacy scale, however, Canada ranks 8th, and it places only 9th on the quantitative literacy scale.

Figure 2 displays the average percentage of each country’s respondents who score in the lowest of five categories in the three tests. This group might best be described as functionally illiterate, corresponding, in our terms, to people clearly lacking the literacy skills to comprehend the written material needed to be competent citizens. As we can see, the ranking corresponds fairly well to differences observed in levels of factual political knowledge. Canada is at the bottom of a wide middle category, along with Belgium, Switzerland and Australia. While the fit with knowledge of international affairs displayed in Figure 1 is not perfect, it is close enough to suggest that general literacy skills and levels of political knowledge are associated.

Civic Literacy and Educational Attainment

For four of the last five years, the authoritative United Nations Development Programme has rated Canada highest on its Human Development Index (HDI). These rankings are determined in part by GDP per capita and life expectancy. But the aspect of the HDI which, more than any other, has enabled Canada to consistently place at the top is the “educational attainment index.” How does this jibe with the IALS findings on comparative literacy levels?

In fact, the IALS scores of the participating countries bear little resemblance to their ranking on the “educational attainment index.” Indeed, the rankings are virtually reversed: of the countries in the IALS, Canada places first in educational attainment, the United States second, and New Zealand third, with Sweden, Germany and Switzerland at the bottom.21

To understand this discrepancy, we first note that, unlike the IALS, the HDI was established to compare all countries rich and poor. One component of the educational attainment score is the percent of the population able to read; but this component simply places all the rich democracies in the same category (99 percent). This means that it is the second component, the average number of years spent in school, which differentiates countries comparable to Canada on the educational attainment index. And Canadians, on average, do indeed spend more years in school than people in all other countries, followed by Americans. So why are Canadians’ long years in school not reflected in their functional literacy scores?

In part, this reflects the fact that while Canadians on average spend more years in school, this does not translate into days, given Canada’s short school year, especially at the university level.22 But beyond this, quantitative ratings such as the educational attainment index tell us nothing about the quality of the education; that is, what is taught and retained. This is the dimension of education most relevant to civic literacy.

One way of assessing educational quality is to consult comparative tests of the knowledge of older students. The only recent such study is a test of skills in mathematics and science – the Third International Mathematics and Science Study (TIMMS) – given to students in the final year of secondary school.23 Among those from the comparable 13 countries, Canadian 16 year-olds placed 7th in mathematics and 8th in physics.

In weighing such results, it is important to consider not only mean scores, but also the distribution of scores. Of particular concern, from a civic literacy perspective, are those at the bottom end of the scale, whose lack of literacy skills and political knowledge may be serious barriers to the most basic forms of democratic participation, such as voting. Those countries identified as having high levels of political participation and adult literacy, the Scandinavian countries and the Netherlands (though not Germany), not only achieve high average scores on math and science literacy, they do especially well when it comes to the average scores of their lower-scoring students. In Figure 3, we set out the combined average score in the TIMSS for those at the five-percent level. Comparing Figures 2 and 3 we can see that countries with lower levels of functional illiteracy among their adult populations (in the IALS) tend to be those with the highest levels of scientific literacy among the lower stratum of late secondary-school students.

Canada’s slightly above average score here is probably misleading since Canada’s relatively high rate of high-school dropouts (discussed further below) means that a disproportionately large number of potential low scorers were likely not included. Support for such a conclusion is found in the results of the IEA (International Association for the Evaluation of Educational Achievement) Reading Literacy Study of 1991, which tested reading skills (narrative, expository and document) of 14 year-old students in 32 countries. “In all three domains, New Zealand, the United States and Canada (British Columbia) showed the widest range of achievement between the lowest five percent and the highest five percent”.24

The data in Figure 4, based on the 1997 IALS report, are also not encouraging for Canada. The high civic-literacy countries, such as Sweden and the Netherlands, do especially well in promoting literacy among those with the lowest level of education. Only 27 percent of Canadians over 16 without high school completed are functionally literate, placing Canada 6th out of 10, compared to 62 percent for the first-place Swedes, and 17 for the last-place Americans.25

Such shortfalls in the achievement of universal literacy in Canada are worrying. The attainment of a high level of civic literacy depends, in good measure, on the capacity of schools to bring students near the bottom to the basic level of cognitive proficiency. Students must attain the functional literacy required to comprehend the basic texts and documents associated with competent citizenship. The link between formal schooling and civic literacy is largely determined by the degree to which the schools succeed in this endeavor. But also of importance are the educational activities, formal and informal, that nurture and sustain civic literacy after formal schooling is completed. One obvious source of later learning is adult education. The only close-to-systematic data comparing countries on the provision of adult education is from the IALS and combines adult education with training programs. The Nordic countries lead with annual participation rates over or at 50 percent, followed by New Zealand at 48 percent. Except for Ireland and Belgium, which trail with levels in the 20s, the remaining comparable countries cluster between 44 and 37 percent. Canada is at 39 percent. Unfortunately, we have no breakdown between adult education and training programs, but we do know that a distinct majority in Canada is in job training.26

Another study also suggests that Canada’s record in adult education leaves a great deal to be desired. Table 1 reveals that not only do Canadians– as expected – participate less in adult education than Swedes, but that the difference is especially acute among those with the least formal education. The ratio of university graduates participating in adult education compared to those with less than nine years of education is 9 to 1 in Canada, but only 3 to 1 in Sweden.

In the IALS survey, Canada places fourth in average number of hours of adult education and training (after New Zealand, Denmark and Finland – Sweden is not included), but falls behind Australia, the Netherlands, Norway and Switzerland if job-related education and training is excluded.27 This pattern is confirmed by a study by the Department of Human Resources and Regional Development, which identified 32 percent of Canadians as having participated in 1993 in some form of adult education and training, among whom over 60 percent were involved in job-related training.28

Civic Literacy and Media Consumption

Apart from education, the primary contributor to functional literacy and political knowledge is media consumption. Indeed, it is possible to regard using the media as a regular source of information as a form of adult education. Individuals who read newspapers, in Canada as elsewhere, are politically more knowledgeable. Baker et al. report that “the predictor that accounted for the most variation in knowledge about parliament among Canadians was one that tapped whether they read a newspaper daily.”29

Figure 5 sets out the comparative data on annual daily newspaper circulation.30 We can see that Canada fares very poorly, ranking above only Italy and Ireland.

But despite these low circulation numbers, a larger proportion of Canadians than Americans report getting their news from newspapers.31 The reason for this is the high consumption in the United States of alternative media forms less likely to promote civic literacy, in particular television. Table 2 presents comparative data on two aspects of television consumption: hours of television viewing per week and per capita spending on television advertising. The latter measure is relevant from a civic literacy perspective because other things being equal, the more commercial considerations enter into decisions on “news and public affairs” programming, the lower their actual public-affairs information content.32 The United States leads in both categories. Canada, however, is not far down the list, ranking fifth in average hours of television viewing per week and seventh in per capita expenditures on television advertising.

The data in Figure 5 and Table 2 are used to create a combined indicator termed “television dependency.” The television-dependency scale is based equally on countries’ relative rating on newspaper reading and commercial television consumption. The latter is a compound measure composed equally of the two elements in Table 2, average weekly television watching and per capita spending on television advertising. Countries, the data reveal, tend to be high in commercial television consumption, or in newspaper circulation, but not both (or neither). Indeed, they cluster into two polar types: in one, there are roughly four to five people for each daily newspaper and commercial television watching is widespread and frequent; in the other, newspaper circulation is around one for every two people and there is far less recourse to commercial television. The overall television dependency scores for Canada and other comparable countries can be found on the y-axis in Figure 6.33

As expected, there is a close relationship between television dependency and the measures of literacy and political knowledge described above. Figure 6 plots the close relationship between TV dependency and functional illiteracy (from Figure 2); the correlation coefficient of 0.59 indicates a strong positive relationship between the two variables. Figure 7 is even more striking, revealing a remarkably strong inverse relationship between knowledge of the United Nations and TV dependency.

The Impact of Civic Literacy on Political Participation

The data presented in the previous sections strongly suggest that Canadians, despite Canada’s top ranking on the United Nations Human Development Index, should not be complacent about the state of civic literacy in this country. Low levels of political knowledge, relatively high rates of functional illiteracy and high consumption of commercial television all point to a polity comparatively weak on the civic literacy scale.

In this section, we consider the implications of these findings for political participation. What is the relationship between civic literacy and political participation?

Certainly, at the individual level, the connection between political knowledge and voting is well established. The Post-Kaiser-Harvard survey study of American voters cited above found that “those in the highest third of the survey in terms of political knowledge were twice as likely to have voted in the 1994 presidential election as those in the lowest third.”34 The relationship between knowledge and voting is even stronger for off-year legislative elections.35 There are similar findings for Canada. We know, from an analysis of the 1997 CES data, that the more politically knowledgeable were nine percent more likely to have voted. The only variable that had a greater effect among the other nineteen tested – which included education and political interest – was age.36 Likewise, a recent IRPP study reported that 69 percent of those who had answered no more than one political knowledge question correctly said they had voted in the 1997 federal election, compared to 90 percent of those with three right.37

Aggregate data also support the contention that voter turnout is closely related to levels of political information. While aggregate turnout comparisons are usually limited to legislative elections, the analysis below focuses on turnout in local elections,38 which is arguably a more faithful expression of civic engagement.39 Such data are far more difficult to assemble, however, with the result that, in a few cases, especially the federal states, they are somewhat incomplete.40 This includes Canada, for which mean turnout in contested municipal elections in Ontario and Alberta is used, along with data from Quebec.41

Figure 8 plots average voter turnout in local elections (excluding countries with compulsory voting in such elections) against TV dependency. We observe that the Scandinavian countries, along with the Netherlands and the German-speaking countries, are high in turnout and low in TV dependency. The opposite is the case for the English-speaking countries. The correlation coefficient of 0.64 corresponds to that derived when average voter turnout in local elections is plotted against functional literacy (0.63) and knowledge of the United Nations (0.69). The strength of these correlations suggests a connection between levels of political knowledge in a country and the degree of political participation by its citizens. Canada’s position on Figure 8 makes it dramatically clear that we are lagging behind many comparable countries on some important barometers of the vitality of our democratic system.

These findings concerning media consumption and political participation are consistent with Putnam’s assertion that newspaper reading is associated with high levels of social capital and TV viewing with low levels.42 Indeed, Putnam indicts increased TV viewing as the main culprit for the decline in social capital in the United States over the past several decades. Television, he argues, destroys social capital by taking time away from civic engagement and by making viewers less trustful of the world outside their homes. Lower levels of political participation and a diminished democracy are the result.

An alternative account of the impact of television on the quality of democratic life is suggested by the foregoing analysis. Putnam, I would contend, is right about television, but – when it comes to voting at least – for the wrong reason. Missing from his analysis is the fact that television viewing, and commercial television viewing in particular, tends to replace newspaper reading, and thereby reduces the quantity and quality of relevant knowledge that helps determine whether the citizen decides to vote and otherwise participate politically. Television matters not because it erodes social capital but because it inhibits the development of civic literacy.

How to Raise the Level of Civic Literacy in Canada

This section signals possible ways of enhancing civic literacy in Canada under three headings: education, media use and political institutions. It draws largely on programs and practices already in place in the high civic-literacy Scandinavian countries. A number of possible courses of action are put forward, but the various proposals can be effective, it must be stressed, only as part of an integrated strategy to raise civic literacy in Canada.


Civics education

At first blush, the most likely method of improving levels of civic literacy is civics education. The province of Ontario, for one, is putting in place a new civics curriculum that places emphasis on participatory learning. There are many programs of this kind being implemented in the US43 and in other countries typically aimed at high-school students.44 However, the existing, by no means systematic, evidence suggests that adolescence is a stage in life not especially conducive to the kind of learning stressed in civics courses.45 The overwhelming evidence of low political knowledge among young people in the US leaves one skeptical about any long term effects of civics courses, given that such courses– typically about US government – are already part of the curriculum for the majori ty of American high-school students.46 While there is some American and Swedish evidence that civics courses taken in very late adolescence do enhance the political knowledge of students,47 even there we do not know to what extent the information acquired in such courses is retained into adulthood. Nor does there appear to be any systematic information enabling us to compare Canada with other countries on the provision of civics education courses. Given limited resources, the best course of action, it appears, would be to aim civics (i.e. introductory political science) courses at 16 to 18 year-olds still in school.

As far as younger students are concerned, priority should be placed on fostering habits of general literacy during the period of compulsory schooling. Part of this effort must consist of identifying and counteracting societal developments that lead young people, and young men especially, to drop out of school in their mid-teens and face a life of functional illiteracy.48 The challenge is to ensure that students are inculcated with literacy skills, and encouraged to develop what the IALS calls “habits of literacy”: reading newspapers and books, using libraries, making use of different sorts of maps, writing letters. The first priority should be to inculcate habits of literacy in the overall student population, including those likely to leave school at an earlier stage as well as those destined for post-secondary education.

Adult Education

Literacy habits and political knowledge acquired at school can be reinforced to promote civic literacy among adult citizens in three ways: the media, the activities of parties and other political actors and through educational – as distinguished from training – programs aimed at adults. To judge by the Speech from the Throne early in 2001, adult education is a major Canadian priority. But rhetoric is always easy. Even if more money goes to adult-education researchers and professionals, overall numbers of participants are not likely to be meaningfully affected. I am persuaded that the study circles run by political parties, trade unions and similar groups, so much a fixture of democratic life in Sweden, very much contribute to civic literacy.49 But words and money from Ottawa will not do the job here. The Canadian provinces generally lack the infrastructure around which such activities are organized in Swedish towns and cities. The first priority is to identify just what kinds of programs are being offered and who is using them,50 and then invite the provinces to develop a coordinated strategy. In the meantime, significant resources should be targeted at programs (described in the next two sections) that could more immediately promote reading and knowledge-enhancing forms of media consumption.

The Media


There are a number of avenues to foster reading. One, all but ignored in North America, is the use of subtitling, an unconscious but very effective method of reinforcing reading ability. Typically in the small language communities of Europe, including Scandinavia, all foreign language television programs, films and videos are subtitled, an idea most directly applicable to Quebec (as further discussed below). In Canada, subtitling in the other official language might be undertaken, an approach that would help foster bilingualism and, as an added bonus, contribute to enhanced reading skills. But while subtitling is technically easy and relatively inexpensive, there are political costs to such policies that politicians are likely to prove loath to pay.

There are, however, other specific measures that could encourage the reading of books. Though some provinces are more effective in this than others, all can learn from a number of Scandinavian programs. Libraries in Sweden, for example, provide free home delivery for shut-ins and services to hospitals and homes for the elderly.51 An exemplary measure is the Swedish program that places in the hand of the nurse at the time of the first post-natal home visit “The Child’s First Book,” a compilation of rhymes and stories for children. The program is designed to underscore that reading, like proper nutrition and hygiene, is vital to the development of a healthy child.

Still, the most important single aspect of maintaining reading literacy – and promoting civic literacy specifically– is regular newspaper reading. We have seen how civic literacy varies inversely with television dependency. With one daily paper in circulation for every two people, the Nordic countries stand out as that corner of the world where reading daily newspapers remains an unquestioned part of everyday life.52 And not accidentally. Efforts are made to encourage newspaper readership at all levels. For example, the daily newspaper 8 Sidor (8 Pages) is published in simple Swedish and distributed to those with low reading skills. Cassette recordings of the daily newspapers are sent out free of charge to people with dyslexia or vision impairments. And there are newspapers targeted at immigrants, the form and content of which is designed to help them understand current issues.

There is an unambiguous positive relationship between government efforts at encouraging newspaper reading and national circulation figures. Norway, Finland and Sweden subsidize daily newspapers that are not leaders in their markets to an amount averaging around 11 percent of their revenues, and 3 to 4 percent of all newspaper revenues.53 Countries like the Netherlands, Belgium, France and Denmark offer some form of indirect press subsidies beyond fairly standard VAT (GST) exemption or reduction.54 The overall result is that in much of Europe, unlike in North America where such exemptions are rare, reading newspapers remains a part of daily life for most people.

Given ingrained habits, cost-reducing subsidies are not likely to bring us up to Northern European readership levels, but they could stem the decline. And the arrival of Metro, a free, quality daily paper (originated and half-owned by a Swedish company) distributed in the subway in Toronto and Montreal might conceivably have some impact on the literacy habits of commuters.

The Electronic Media

So far, the Internet, like television and radio, is a complement and not a substitute for newspaper reading when it comes to civic literacy. The evidence is that those people who use the Internet for information are those who read newspapers for the same purpose.55 It is public television, despite its declining audience share, that still serves in most countries56 to complement the civic-literacy enhancement effect of newspaper reading.57 The United States, with its tiny public broadcasting sector, has witnessed the most dramatic replacement of newspapers by commercial television as a source of information.58

To avoid a similar outcome in Canada, efforts to reinforce public broadcasting must be redoubled. Recent moves have not been encouraging: a comparative study of financial support for public broadcasting between 1995 and 1998 found that Canada has been losing ground, with money for public television dropping 28.8 percent – the largest decline for all OECD countries.59 New funding has been promised the CBC in the 2001 Throne Speech, but not enough to keep its president from anticipating further cuts. Yet, as revealed by the latest MacLean’s annual poll,60 support for public broadcasting is stronger than one might expect, with only a third of respondents favouring the sale of the CBC to the private sector.

Of course, in the era of the Internet and 500 television stations, the cause of public television, like that of newspaper reading, may be a lost one.61 But the opposite may also prove to be the case. The MacLean’s poll found 61 percent to approve of restrictions on foreign content. There appears to be significant public support for the type of Canadian content most likely to be provided by a public broadcaster.

Political Institutions

Enhancing the Informational Activities of Political Actors

The high civic-literacy Scandinavian countries are leaders in providing a range of measures supporting the informational activities of political parties, the prime organizational actors disseminating political knowledge. Scandinavian newspaper subsidies were established not simply to foster diversity in the media, but also to facilitate the public presentation of party viewpoints. Adult education in these countries also serves this end. In Sweden, each of the parties is associated with one of eleven adult educational associations that organize state-supported study circles for their member organizations. Especially in the period before an election, the parties use study circles to familiarize party members and sympathizers with their programs and strategies.62

Another method of assisting parties in their educational activities, this pioneered in relatively high civic-literacy Germany, is the establishment of publicly funded think tanks associated with the parties. The 1992 report by the Royal Commission on Electoral Reform and Party Financing (the Lortie Commission) recommended that something similar be done in Canada.63

In addition to these measures, there is a generous system of direct subsidies to the political parties in the Nordic countries. The contrast with the United States, with its dependence on vast sums of private funds to pay for television advertising and other forms of political communication, is striking. The negative informational effect of unlimited campaign spending and unregulated access to television commercials in the United States (unlike all comparable countries) is indisputable. The effect is dampened in Canada by our provision of free television and radio time to the political parties.

A related set of policies from which we can learn concerns transparency in political information dissemination, namely the regulations that govern the availability and accessibility of information related to the public interest. I am thinking in particular of the Nordic countries’ requirement that public institutions open their books to interested citizens, with ombudspersons appointed to ensure that this is done.64

Proportional Representation

As argued in greater detail elsewhere, the well-established turnout boost associated with electoral systems based on proportional representation (PR) is, in large part, an expression of civic literacy: PR enhances civic literacy by simplifying the political map and reducing the incentives for politicians to distort information.65 In briefly summarizing that argument here, the discussion is limited to the effects of the electoral system, though it should be noted that PR systems typically go hand in hand with other regulations facilitating the representation of smaller parties.66

We know that PR boosts voter turnout – some estimate the effect to be as high as 9 to 12 percent – but we do not know just how. Most directly, the boost is due to every vote counting under PR. Each voter in a PR election is like a resident in a highly contested single-member district, which we know to experience higher turnout.67 But it is less a matter of the individual’s increased chance of affecting the outcome, which, in reality, is still minuscule, than of the different incentives placed on political parties.

These incentives affect, firstly, how parties invest their limited resources. In a first-past-the-post (FPTP) election, parties ignore many voters, investing much effort and money to get the attention of voters in close contests.68 Unlike PR, FPTP exaggerates a party’s geographic weak spots, creating a disincentive against involvement in regions where it has relatively little support. Under PR, by contrast, since every vote counts equally toward electing representatives of one’s preferred party, parties have an incentive to inform all voters of their programs.

Electoral systems also affect where parties position themselves politically and how consistently they abide by their principles. The representational logic of PRbased, multi-party systems is to inhibit precipitous changes to a party’s principles and identity, that is, the elements that constitute its place on the political map. Political actors, and the voters themselves, can thus count on a clearly drawn and stable political map on which to plot their own paths. For political actors, this includes reliable expectations about which other actors to cooperate with and over what issues. Under FPTP, on the other hand, electoral prospects are maximized by targeting the median voter, leading to the blurring of policy differences between major parties and an undue emphasis on personality at the expense of policy debate.

Finally, electoral systems affect how parties represent the platforms of their opponents. Under PR there is less incentive for political leaders– who may very well need their opponents’ support to form coalitions after the election – to inhibit the awareness of the electorate of alternative positions on the issues of the day. For FPTP-based political actors, on the other hand, coalition government is an unlikely outcome and they therefore have an interest in misrepresenting the policies of other parties in ways likely to cause them electoral harm.

The absence of PR thus means a public less politically informed than would otherwise be the case. Moreover, not all strata in society are equally affected. By making it easier to identify with a political party, PR fosters political participation especially at the lower end of the income and education ladders where information is at a premium. Unfortunately, it is extremely difficult to argue the case for PR in Canada. The great majority of Canadians, including most opinion leaders, have yet to notice that Canada is the only old Westminster democracy that has not partially or, in the case of New Zealand, fully abandoned FPTP.

Provincial Jurisdiction and a “Distinct” Strategy for – and by – Quebec

Given that many of these policies lie in provincial areas of jurisdiction, the specific application of the prescriptions outlined above will often differ to reflect the specific conditions in each province. This is clearly the case as far as Quebec is concerned. Without assuming all other provinces and regions to be the same, Quebec clearly differs from English-speaking Canada in the relative strength of its democratic underpinnings of civic literacy and political participation.

On the participation side, Quebecers turn out more in provincial, and, not infrequently, federal elections; they tend to be more interested in politics and more involved in political parties. Caldwell and Reed found that though Quebec’s overall participation rate in voluntary organizations is one fifth lower than Ontario’s, it is 50 percent higher for political organizations.69 And, using data from the 1990 World Values Survey, Johnston and Soroka found lower levels of organizational membership and interpersonal trust in Quebec but higher levels of political trust compared to the rest of Canada.70

As for literacy, civic and otherwise, Quebecers read fewer newspapers and watch more television than Canadians as a whole.71 This is especially the case among less-educated Quebecers, of whom 69 percent state television to be their primary source of information, compared to 53 percent in the rest of Canada.72 In the IALS prose comprehension test, a very high 28 percent of Quebecers were functionally illiterate (level 1) compared to 20 percent for the rest of the country,73 with the average score about 10 percent lower. The difference is due to the combined effect of lower levels of formal education among francophones (older ones especially), and a weaker “literacy habit.” On the (weekly) indicators of literacy practices, anglophones consistently outscored francophones: 66 percent to 54 percent on reading daily newspapers, 56 to 40 on reading books, 28 to 18 on visiting a library, 41 to 26 on writing letters.74 As far as younger Quebecers are concerned, the problem is not one of lower quality education: performance data collected for the provincial ministers of education in recent years show Quebec students, especially the francophones, near or at the top.75 The problem appears to be that Quebec high schools allow a larger proportion of students to drop out before completion.

Hence, while the federal government’s recent concerns with adult education and literacy are to be welcomed, Quebec will need to retain its autonomy in education, training and culture and communications in order to address its most pressing problems – high dropout rates and weak literacy habits. In the latter instance, more effort is needed to foster newspaper reading and library use, and to develop adult education programs reflecting the particular needs of its citizens. As noted, an effective instrument would be the gradual replacement of dubbing by subtitling in films, videos and even television. This will undoubtedly be a hard political sell, given that there is a strong pro-dubbing lobby among Quebec actors, and given the fact that some Quebec nationalists – though fewer these days than in the past – presume that greater direct access to English-language content weakens Quebec culture.

As far as high school dropouts are concerned, a good case can be made that some of the money awarded by Ottawa through the Millennium Scholarship program and the newly established university research chairs could have been better spent on programs addressing the excessive number of dropouts in Quebec, among boys particularly. As a general rule, the emphasis on “national” programs in these areas must give way to the principle that the provinces are best suited to identify and serve the civic-literacy needs of their citizens.

Conclusion: Civic Literacy and Social Equality

The central message of this paper is that high levels of civic literacy are necessary to sustain widespread political participation in a democracy. From this another conclusion follows: that high-civic literacy societies, because they encourage political participation across all sectors of society, will tend to adopt policies that take into account the full range of interests in society. If the desire of Canadians to sustain a welfare state qualitatively different from that of the United States76 is to be met, we must be aware of these important linkages between civic literacy and the maintenance of egalitarian social and economic policies.

Canada was unusual among the wealthy countries in the mid-1990s, in that its disposable income was as equally distributed at that point as it had been a decade earlier, with the result that Canada went from a position closer to the US to one closer to the Scandinavian countries. Yet, while Canada lies as close to Scandinavia as to the US in the generosity of its welfare state, this is not the case when it comes to civic literacy. Indeed, as we have seen, it is only the United States that Canada unambiguously surpasses on the indicators of civic literacy.

Figure 9, drawn from the data of the final report of the IALS,77 dramatically captures the close relationship between income inequality and civic literacy. Its message is undeniable: democratic societies that more equally distribute intellectual resources – i.e. high civic literacy societies – also more equally distribute material resources.78

Canada’s position on Figure 9 suggests that Canadians face an important choice at this juncture. We can move horizontally or vertically away from our outlying position: the former means attempting to raise literacy, civic literacy in particular, in the ways described above; the latter means allowing inequality to increase to the point where it is congruent with our poor showing on the civic literacy scale. A society with 40 percent of its citizens excluded by a lack of the basic resources needed for civic competence from active, informed citizenship will ultimately choose policies– and reinforce the institutions underlying these policies – significantly different from those of a society with only 15 percent excluded. As a recent study comparing American states concluded:

Even after adjusting for other factors that might predict state welfare policy – the degree of public liberalism in the state, the federal government’s welfare cost-matching rate for individual states, the state unemployment rate and median income, and state taxes – robust relationships were found between the extent of political participation by lower-class voters and the generosity of state welfare payments. In other words, who participates matters for political outcomes, and the resulting policies have an important impact on the opportunities for the poor to lead a healthy life.79

The choice we are facing should be placed in the context of accelerating economic globalization and the growing danger of a widening digital divide. The question is this: how can we avoid mirroring a globalized world economy, with its minority of “winners” and a majority of “losers,” losers not so much due to economic deprivation, but to their inability to take informed action to make their society better for themselves and others? Only high civic-literacy societies, institutionally arranged so that a substantial majority of their citizens can count on meaningful maps to guide them through the complex decisions that their community faces, can hope to do so. Canadians may not be agreed on whether to follow the well-paved, high-inequality highway the United States has laid out, or the rockier, more egalitarian path of Northern Europe, and Scandinavia in particular. But most would agree, I believe, that we should not simply rest on our laurels and find ourselves navigating with a political map from which the high civic literacy road is omitted.

  1. For a more detailed presentation of the arguments in this paper, see Henry Milner, Civic Literacy: How Informed Citizens Make Democracy Work (Hanover NH: University Press of New England, 2001). The research was conducted with the generous support of research grants from the Social Science and Humanities Research Council of Canada and the Quebec Department of Education (FCAR).
  2. Robert Putnam, Bowling Alone: The Collapse and Revival of American Community (New York: Simon and Schuster, 2000). A recent issue of the federal government’s new journal of policy research, Isuma (Gerald Baier and Herman Bakvis, “Think Tanks and Political Parties: Competitors or Collaborators,” Isuma, Vol. 2, no. 1, [Spring 2001]) was devoted to articles on the subject by Putnam and several Canadian collaborators.
  3. Putnam, Bowling Alone, p. 291.
  4. Data compiled by the Institute for Democracy and Electoral Assistance (IDEA). Unlike other compilations, IDEA takes into account differences in rates of registration by using age-eligible citizens as the base. IDEA, “Voter Turnout from 1945 to 1998 – A Global Report on Political Participation.” Accessed June 10, 2001. Available at https://www.idea. int/voter_turnout/index.html
  5. Standardized questions are common when it comes to efficacy and horizontal trust-related questions, for example: “do you agree that ”˜politicians don’t care about ordinary people like me?’” The most likely choice for devising and posing a country-neutral battery of political knowledge-oriented questions is the CSES (Comparative Study of Electoral Systems) group. CSES, however, abandoned efforts to include common political knowledge questions, in favour of leaving the content to the participating national surveys.
  6. Larry Bartels, “Uninformed Votes: Informational Effects in Presidential Elections,” American Journal of Political Science, Vol. 40, no. 1 (February 1996), pp. 194-230.
  7. National Election Studies, “The 1948- 1998 Time-Series Data Collections.” Accessed June 10, 2001. Available at ainfo/tsinfo.htm
  8. Robert Lambert, J.E. Curtis, B.J Kay and S.D. Brown, “The Social Sources of Political Knowledge,” Canadian Journal of Political Science, Vol. 21, no. 2 (June 1988) pp. 359-75.
  9. These calculations were provided by Agnieszka Dobrzynska, affiliated researcher at the Université de Montréal. Data from the 1997 Canadian Election Survey were provided by the Institute for Social Research, York University. The survey was funded by the Social Sciences and Humanities Research Council of Canada (SSHRC), grant number 412-96-0007 and was completed for the 1997 Canadian Election Team of André Blais (Université de Montréal), Elisabeth Gidengil (McGill University), Richard Nadeau (Université de Montréal) and Neil Nevitte (University of Toronto). Neither the Institute for Social Research, the SSHRC, nor the Canadian Election Survey Team are responsible for the analyses and interpretations presented here.
  10. Paul Howe, “The Sources of Campaign Intemperance,” Policy Options, Vol. 22, no. 1 (January 2001).
  11. Richard Gwyn, “Rediscovering Our Citizenship” (Toronto: Dominion Institute, 1997), pp. 21-2.
  12. Note that Australia and New Zealand used mailed-in ballots. See John Curtice and W. Phillips Shively, “Who Represents Us Best? One Member or Many?” Presented at the August 2000 International Political Science Association (IPSA) Conference, Québec, Canada.
  13. Respondents were asked questions current at the time: 1) the president of Russia (Boris Yeltsin); 2) the country threatening withdrawal from the nuclear non-proliferation treaty (North Korea); 3) Boutros Boutros-Ghali’s position (UN Secretary-General); 4) the ethnic group that had captured much of Bosnia and surrounded Sarajevo (the Serbs); and 5) the group with which Israel had just reached an accord (Palestinians, or PLO).
  14. Stephen Bennett et al., “The Impact of Mass Media Exposure on Citizens’ Knowledge of Foreign Affairs: A Five Nation Study.” Presented at the Annual Meeting of the American Political Science Association, Chicago, August 1995, p. 43.
  15. John Baker et al., “Citizens’ Knowledge and Perceptions of Legislatures in Canada, Britain and the United States,” Journal of Legislative Studies, Vol. 2, no. 2 (Spring 1996), p. 51.
  16. Steven Holloway, “Through a Glass Darkly: Documenting Asymmetries in Neighbourly Knowledge,” Bulletin of the Canadian Political Science Association, Vol. 24, no. 1 (Spring 1995), pp. 51-56.
  17. Gilbert Grosvenor, “Our Year in Review,” National Geographic (December 1989), pp. 816-18.
  18. William Millard, “International Public Opinion of the United Nations,” International Journal of Public Opinion, Vol. 5, no. 1 (Winter 1993).
  19. OECD, Literacy Skills for the Knowledge Society: Further Results from the International Adult Literacy Survey (Paris: OECD, 1997) and OECD, Literacy in the Information Age: Final Report of the International Adult Literacy Survey (Paris: OECD, 2000).
  20. I exclude Poland, Portugal, Hungary, Slovakia, Chile and the Czech Republic as inappropriate for comparison with Canada. I include Switzerland by combining the separate IALS results for German and French Switzerland, weighing the former at two-thirds and the latter at one-third. Note that the IALS figures for Belgium are in fact only for Flanders.
  21. UNDP, Human Development Report, (New York: Oxford, 1997), p. 146.
  22. A 1990 survey tied Canada with the United States in second to last place, with an average of about 180 teaching days in elementary and high school. Michael Barrett, “The Case for More School Days,” Atlantic Monthly (November 1990), pp. 78-106.
  23. TIMSS International Center for the Study of Testing, Evaluation, and Educational Policy, Mathematics and Science Literacy in the Final Year of Secondary School (Boston: TIMSS, 1998).
  24. W.B. Elley and A. Schleicher, “International Differences in Achievement Levels,” in W.B. Elley (ed.), The IEA Study of Reading Literacy: Achievement and Instruction in Thirty-two School Systems (Wheaton, UK: Pergamon, 1994), p. 59.
  25. OECD, Literacy Skills for the Knowledge Society, p. 29.
  26. OECD, Literacy Skills for the Knowledge Society, p. 93 and OECD, Literacy in the Information Age, p. 43.
  27. OECD, Literacy in the Information Age, p. 42.
  28. Department of Human Resources and Regional Development, Adult Education and Training in Canada (Ottawa: Ministry of Supply and Services, 1994).
  29. John Baker et al., “Citizens’ Knowledge and Perceptions of Legislatures,” p. 51.
  30. Of course, not all newspapers are the same. There is an important difference, with respect to the contribution to the reader’s knowledge, between the UK tabloids and broadsheets. Baker et al. (“Citizens’ Knowledge and Perceptions of Legislatures”) suggest that the inclusion of tabloid readers explained the lack of a significant positive effect on political knowledge found for newspaper reading – as opposed to radio listening – in Britain but not in the US and Canada. Yet, despite being unable to control for quality, I am confident that over the entire set of countries, gross per capita daily newspaper circulation reflects, as well as affects, civic literacy.
  31. Baker et al., “Citizens’ Knowledge and Perceptions of Legislatures,” p. 51.
  32. See Henry Milner, Civic Literacy: How Informed Citizens Make Democracy Work (Hanover, NH: University Press of New England, 2001), chap. 7.
  33. To standardize, in creating the composite, each indicator was recomputed to bring it to a score for which the maximum attained was close to 100. Note that Figure 6 only includes the fourteen mature democracies for which IALS data are available.
  34. Richard Morin, “Tuned Out, Turned Off: Millions of Americans know little about how their government works,” Washington Post National Weekly Edition, Feb 5-11, 1996, p. 7.
  35. Samuel Popkin and Michael Dimock, “Political Knowledge and Citizen Competence,” in S. Elkin and K. Soltan, (eds.), Citizen Competence and Democratic Institutions (University Park: University of Pennsylvania Press, 1999), p. 137.
  36. Tony Coulson, “Voter Turnout in Canada: Findings from the 19997 Canadian Election Study,” Electoral Insight, Vol. 1, no. 2 (November 1999).
  37. Howe, “The Sources of Campaign Intemperance.”
  38. There is a rough correspondence between countries’ rankings in national and local turnout averages. The exception is Switzerland, very low in national turnout, and close to the mean in turnout in local elections.
  39. Voting in local elections is understood to be a reflection of – and a spur to – more active forms of political involvement (See Benjamin Barber, A Place for Us: How to Make Society Civil and Democracy Strong [New York: Farrar, Strauss, 1998]). Moreover, local elections are less prone to being distorted by the “noise” of emotion-laden television campaigns designed to bring uninformed voters to the polls, since local elections seldom offer sufficient incentives for incurring the high costs of such campaigns. Hence it is at the local level that casting a ballot and being well informed about relevant issues can be expected to be most closely linked.
  40. It has required considerable effort to seek out these data from the relevant authorities and specialists. The federal states leave it to the regional units to collect municipal data and, in some cases, the states or provinces leave it up to the municipalities themselves – though this has become less common in recent years. When it comes to acquiring information from regional or national authorities, the most difficult are the United States and Switzerland. (I have raised the US numbers to base them on registered voters.) Among the unitary countries, the French data is the least comprehensive, with results only for the 1983 election and only for the 386 municipalities with a population over 20,000. In both cases, the data were acquired through a mailed survey of officials of municipalities. Harvey Schuckman, “What Motivates Civic Engagement in Local Politics: Deconstructing the Role of Local Political Institutions: An Examination of Mayoral Electoral Turnout,” presented at the American Political Science Associations annual meeting: Boston, September 1998 and Andreas Ladner and Henry Milner, “Politicization, Electoral Institutions, and Voting Turnout: the Evidence from Swiss Communal Elections in Comparative Context,” Electoral Studies Vol. 18, no. 2 (June 1999).
  41. Data for Ontario and Alberta have recently started to be collected by each province. Data for Quebec were collected by the author. In Quebec, municipal elections take place on a staggered basis on the same day every fall. In 1997, 35 percent of 782 elections saw the position of mayor contested, with an average turnout of 52.2 percent according to Louise Quesnel in “Le local dans la glace,” in R. Boily (ed.), L’année politique au Québec (Montréal: Les Presses de l’Université de Montréal, 1998). My own calculations for 1998 arrived at an average of 59 percent for the 94 mayoral elections (104 were acclaimed.) These numbers are similar to the 51 percent average for the eight large cities that elected mayors in 1997 and 1998, and the 54 percent for the 28 of the 53 Quebec municipalities with over 20,000 population that had a contested election for mayor in 1993 or 1994.
  42. Robert Putnam, “Tuning In, Tuning Out: The Strange Disappearance of Civic America,” PS: Political Science and Society, Vol. 28, no. 4 (December 1995).
  43. A good place for information about such programs is the Discussion List of the American Political Science Association Task Force on Civics Education (APSACIVED@nullH-NET.MSU.EDU).
  44. Carole Hahn (ed.), Becoming Political: Comparative Perspectives on Citizenship Education (Albany, NY: SUNY Press, 1998) and J. Torney-Purta et al. (eds.), Civic Education Across Countries: Twentyfour National Case Studies from the IEA Civic Education Project (Amsterdam: Eburon-IEA, 1999).
  45. The civics courses taken one hour per week from grade 7 or 8 by practically all German students seem to have little effect on the adolescents (Christa Händle et al., “Concepts of Civic Education in Germany Based on a Survey of Expert Opinion,” in Torney-Purta, Civic Education Across Countries, 1999). Australian researcher Hugh MacKay concludes, “typically, teenagers find little to interest or inspire them in the political process, and they often report that politics is the most boring subject discussed at home.” Quote from Civics Expert Group, Whereas the People: Report of the Australian Commission on Civics and Citizenship (Canberra: Australian Government Publishing Service, 1994), p. 182.
  46. Hahn, Becoming Political, p. 17
  47. This effect is most apparent when the courses are taken in the final years of high school. See Richard Niemi and Jane Junn, Civic Education (New Haven: Yale University Press, 1988) and Anders Westholm, Arne Lindquist and Richard Niemi, “Education and the Making of the Informed Citizen: Political Literacy and the Outside World,” in Ort Ichilov (ed.), Political Socialization, Education and Democracy (New York: Teachers’ College Press, 1989).
  48. A simple contrast of my province, Quebec, where 42 percent of boys and 29 percent of girls dropped out before completing grade 11, with Sweden, where 98 percent continue on to upper secondary school and 80 percent graduate within four years, is illuminating. See Skolverket, “Gymnasieskola för alla – andra. En studie om marginalisering och utslagning i gymnasieskolan,” (Stockholm, 1998).
  49. Consider the ABF, the workers’ educational association affiliated with the trade unions and Social Democratic Party, the largest of 11 adult education associations. The ABF annually organizes about 100,000 study circles for over a million participants. There are, in all, about 350,000 such study circles in Sweden, over and above the municipally-run Komvux courses for completing compulsory and upper secondary education given, in 1999, to more than 200,000 adults. Over 50,000 people participated in ABF study circles on EU membership in the period preceding the 1994 referendum. The fall 1999 program of the ABF in Umeå (where I did research) offered the usual range of courses in languages, computers, art, music, and nature appreciation, but also courses in organizing groups and co-operatives, in public speaking, writing and understanding media, as well as study circles on social and civil rights, the United Nations, war and peace, the future of democracy, feminism, various aspects of history and important contemporary books. For more information, see Milner, Civic Literacy, chap. 9.
  50. The federal minister responsible for such programs seems to have some misconceptions in this regard. Jane Stewart is quoted as saying that adult education has always been for “those who didn’t make it through high school,” quite the opposite of what is shown in Table 1. See The Globe and Mail, February 6, 2000, p. A4.
  51. Henry Milner, Sweden: Social Democracy in Practice (Oxford: Oxford University Press, 1989) and Henry Milner, Social Democracy and Rational Choice: The Scandinavian Experience and Beyond (London: Routledge, 1994).
  52. The most up-to-date figures are found in a new report by the World Association of Newspapers. In a press release on June 4th 2001, it stated: “The figures on newspaper reach, or readership, among adults find Sweden leading the world for the first time with 88 percent, overtaking Finland, which has dropped 5 points to 86 percent since 1999 and now ties second place with Norway.” World Association of Newspapers, “World Press Trends: Newspaper Growth Continues.” Accessed June 10, 2001. Available at forum/press. releases/WPT040601.html
  53. The overall level of subsidies declined somewhat during the economic slowdowns of the early-mid 1990s. The only other country known to this writer with a direct subsidy program comparable to that of Norway, Finland and Sweden is Austria. See A. Sanchez-Tabernero, Media Concentration in Europe (Manchester: European Institute for the Media, 1993), p. 229; Stig Hadenius and Lennart Weibull, “The Swedish Newspaper System in the Late 1990s. Tradition and Transition,” Nordicom Review, Vol. 4 (Winter-Spring 1999); and Raimo Salokangas, “From Political to National, Regional and Local: The Newspaper Structure in Finland,” Nordicom Review, Vol. 4, (Winter-Spring 1999).
  54. E. De Bens and H. Østbye, “The European Newspaper Market,” in McQuail Denis and Karen Siune (eds.), Media Policy: Convergence, Concentration and Commerce (London: Sage, 1998), p. 14.
  55. For example, a study of 529 American undergraduate students found that the Internet is used mainly as a source of entertainment, with use of the web as a news source confined essentially to those who get news from newspapers. See Scott Althaus and David Tewksbury, “Patterns of Internet and Traditional News Media Use in a Networked Community,” Presented at the Midwest Political Science Association, Chicago, (April 1998).
  56. In recent years, most countries have moved toward the American model, though stopping short of its almost entirely commercial character. One change has been to make public television more dependent on advertising revenue. New Zealand’s TVNZ, followed by Italy’s RAI and France’s F2 and F3, have gone furthest in this direction, with the CBC and Germany’s ARD and ZDF taking a middle road. The exceptions, with little or no advertising revenues, are the public-service television services in the Nordic countries, the BBC and the (Australian) ABC. De Bens and Østbye, “The European Newspaper Market,” 1998, pp. 14, 19.
  57. A Swedish study before and after the 1991 election found that daily TV news watching was significantly related to higher political knowledge, reflecting the fact that (at the time) all TV news was provided by high quality public-service channels. The most beneficial effect of media use was on those at the lower levels of education (Olle Findahl, Sverige elfte TV-val. [Umeå: Medier & Communication, University of Umeå, 1993]). Similarly, controlling for education, attitudes toward the EU and other relevant characteristics, Holtz-Bacha and Norris found Eurobarometer respondents who preferred public television to be significantly more informed about the EU than those who chose private TV. See Christina Holtz-Bacha and Pippa Norris, “To entertain, inform and educate.” Still the Role of Public Television in the 1990s? (Cambridge: Shorenstein Center, Harvard University, 1999).
  58. In the 1960s, less than 20 percent of Americans said they usually obtained news of the world from “TV only… This percentage rose roughly 10 percent per decade, reaching 50 percent in 1986.” Quoted in Terry Clark and Michael Rempel, Citizen Politics in Post-Industrial Society (Boulder Co: Westview, 1997).
  59. The US reduced its funding by 4.7 percent, while it went up slightly in France, Japan, Germany, and Britain. Canadian public television receives 16 percent of revenues from state funds compared to 32 for Germany, 34 for Italy, 25 for France, 23 for Japan, and 21 for the UK. (The information comes from a study for the OECD by the Centre d’études sur les médias at Laval University, reported in Le Devoir, August 10, 1999).
  60. MacLean’s, December 25, 2000, p. 29.
  61. If so, says American legal theorist Cass Sunstein in an important new book, democracy as we know it will be jeopardized. For edited newspapers, magazines, and television news are the general interest intermediaries providing common information experiences for citizens, the “social glue” of democracy. See Cass Sunstein, Republic.Com (Princeton: Princeton University Press, 2001).
  62. During election years, such partisan study circles can account for up to half of the study circles offered.
  63. See Baier and Bakvis, “Thank Tanks and Political Parties.”
  64. Axberger, H-G, “Public Access to Official Documents,” Current Sweden, Issue 414, (1996).
  65. Henry Milner, Making Every Vote Count: Reappraising Canada’s Electoral System (Peterborough: Broadview, 1999); Milner, Civic Literacy, chaps. 5 and 6).
  66. Shaun Bowler, Elisabeth Carter and David M. Farrell, “Studying Electoral Institutions and their Consequences: Electoral Systems and Electoral Laws,” presented at the August 2000 International Political Science Association (IPSA) Conference, Québec, Canada.
  67. For example, the Center for Voting and Democracy’s statistical analysis of the 1994 elections for the House of Representatives reports a clear correlation between margin of victory and voter participation, with a 13 percent difference in turnout between the 87 most contested and 54 most lopsided districts.
  68. Concerning last November’s US election, “The National Journal … reports that … four of the nation’s top eight media markets – Boston, Dallas, New York City, and Washington, DC – had a grand total of six presidential ads aired, while eight media markets in battleground states each aired more than 6,500 presidential ads.” Electronic report of the Center for Voting and Democracy, November 22, 2000.
  69. Gary Caldwell and Paul Reed, “Civic Participation in Canada: Is Quebec Different?”, Inroads, Vol. 8 (1999), pp. 215-22.
  70. R. Johnston and S. Soroka, “Social Capital in a Multicultural Society: the Case of Canada,” presented at the Canadian Political Science Association, Sherbrooke, Quebec, June 1999.
  71. Calculations based on 1999 circulation figures from the Canadian Newspaper Association website show roughly 13 daily newspapers circulated for every 100 Quebeckers, compared to 16.6 for Canada as a whole, while francophone Quebecers (in 1993) averaged 26 weekly hours of TV, compared to 22 for Canadians as a whole (International Television and Video Almanac, (New York: Quigley Publishing Company, 1996), p. 681.
  72. Canada Information Office, National Survey of Less Educated Canadians (Ottawa: Ministry of Supply and Services, 2000), p. 21.
  73. Statistics Canada, Education Indicators in Canada, Report of the Pan-Canadian Education Indicators Program to the Council of Ministers of Education. (Ottawa: Minister of Supply and Services, 1999).
  74. Jean-Pierre Corbeil, “Literacy: Does Language make a Difference?”, Canadian Social Trends, Issue 51 (Winter 1998), p. 2.
  75. This is especially the case with regard to scores in mathematics tests. See J. Douglas Willms, “Hypotheses about Community Effects of Social Outcomes,” Isuma, Vol. 2, no. 1 (Spring 2001) and Statistics Canada, Education Indicators in Canada, 1999.
  76. The 2000 MacLean’s poll found that only 21 percent of respondents agreed that Canada should move closer to the US in its policies and laws, while 72 percent disagreed.
  77. OECD, Literacy in the Information Age.
  78. It is because of the greater sophistication of high civic literacy societies, I contend, that redistribution need not be attained at the cost of economic productivity. An examination of the growth data for OECD countries over a recent ten year period shows that there need not be – and that, overall, there has not been – any trade-off per se between equality and economic growth, between wealth creation and distribution. See Milner, Civic Literacy, chap. 11.
  79. Norman Daniels, Bruce Kennedy, and Ichiro Kawachi, “Justice is Good for our Health,” Boston Review, Vol. 25, no. 1 (February/March 2000), pp. 4-10.

Henry Milner is editor of Making Every Vote Count: Reappraising Canada’s Electoral System (Broadview,  1999), and co-editor of Inroads,  a Canadian journal of opinion and policy. He is Professor of Political Science at Vanier College; Visiting Professor at Umeå University in Sweden, and Adjunct Professor at Université Laval. His book Civic Literacy: How Informed Citizens Make Democracy Work will be published in December by the University Press of New England (distributed in Canada by UBC Press).