6
If the institutional structure appears to have an impact on satisfaction with democracy, the
results of figure 1 are not surprising. Figure 1 reports levels of satisfaction with
democracy in the various countries of the CSES dataset. The results vary from a high of
30.9% ‘very satisfied’ respondents in Australia to 0.7% similar respondents in Russia.
For the most part the advanced industrial countries report higher levels of satisfaction,
although it is worth noting that in no country is more than one third of respondents very
satisfied with the way democracy works. Exceptions to this include Belgium, which
reported a lower than average proportion of very satisfied respondents. To determine
whether there are institutional predictors of satisfaction with democracy, two tables prove
useful. Tables 1a and 1b examine the proportion of satisfied respondents with respect to
various institutional features. Table 2 reports the results of multivariate regression
analysis.
Table 1a and 1b about here.
Table 1a reports the proportion of satisfied respondents (very satisfied and fairly
satisfied) in each of the countries with sample sizes exceeding 1,000. It demonstrates the
prevalence of parliamentary systems among those with higher proportions of satisfied
respondents. Single-member district electoral systems and countries with monarchs also
tend to have higher proportions of satisfied respondents. Together, these three features
typify the Westminster democracies of Canada, the UK and Australia. To better illustrate
the variations contained within table 1a, table 1b demonstrates the average proportion of
respondents who were satisfied with democracy by institutional feature. The table
confirms that parliamentary systems have higher proportions of satisfied respondents.
They also have smaller standard deviations, suggesting greater uniformity of responses.
This is not surprising as presidential and semi-presidential systems have been employed
in newer democracies, who we could expect to have differing levels of satisfaction from
established first wave presidential regimes such as the United States. The table also
notes that single-member districts have higher proportions of satisfied respondents. This
confirms research that suggests single-member districts provide better links between
voters and politicians, and provide clear choices to voters. This high result is based on
Convention