good reason to extend our analysis to the last-named attitude as a high rate of political
participation among all groups in society is indispensible for an effective and responsive
democracy (APSA Task Force, 2004). I will use survey data on the civic attitudes of
students aged fourteen and construct a measure of ethno-racial diversity at the classroom
level to explore the relationships between diversity and the two aforementioned outcomes.
As explained below, the impact of diversity at such a micro level of analysis is likely to
be quite different from that of diversity at the neighbourhood, city or national level.
The next section reviews the theory and existing research on diversity and civic
attitudes. Subsequently, various contextual features of the three countries are discussed.
The third section explains the database, the indicators and the methods used. The fourth
section presents the results of the analyses and discusses these in relation to the theory
reviewed. The conclusion sums up.
Classroom diversity and civic attitudes
Putnam (2007) has observed that two contrasting theories apply when investigating the
impact of ethno-racial diversity on civic attitudes: the conflict and the contact perspective.
In the conflict perspective, the relative size of the minority group(s) is crucial. The larger
this size, the more members of the dominant group will feel threatened, the tighter will be
their in-group bonding and the more prejudiced they will become vis-à-vis the minority
group(s) (Blalock, 1967; Quinlan, 1995). By implication, hostility to out-groups should
be minimal in homogenous settings. According to Blalock (1967), this regularity applies
because a growing share of minority groups in the population increases the competition
over scarce resources between groups and gives minority groups more opportunities to
mobilize politically and challenge the privileges of the dominant group.
By contrast, the contact perspective postulates that isolation breeds stereotypes.
Prejudice can be overcome and intercultural understanding can be enhanced if groups
mingle and interact. However, inter-group interaction only yields such positive outcomes
if it occurs (1) on the basis of equality, (2) in settings of common experiences and
common objectives, and (3) on a frequent, lasting and intensive basis (Allport, 1954;