Changing Public Views of Gender Roles in seven nations: 1988-2002

The paper analyzes the change of gender-role attitudes from 1988 to 2002 along two dimensions: the consequences of female labor-force participation for the children and a general gender ideology. Both dimensions are measured by two core items which can be regarded as equivalent across countries. We focus on those countries for which we have data for all three points in time (1988, 1994 and 2002), i.e. Austria, West Germany, Great Britain, the United States, Ireland, the Netherlands and Hungary. While in countries that have established long-time series for gender-role attitudes a liberal trend could be demonstrated to operate at least from the 1970s to the 1990s, scholars are discussing a possible trend reversal in more recent times. Actually, what the most recent ISSP data show is less a trend reversal than a leveling off the liberal trend. This is particularly true of Great Britain and the United States, which were already characterized by less traditional attitudes; whereas in most of the countries with more traditional attitudes the liberal trend is continuing (in particular: Austria and West Germany). The leveling off in Great Britain and the United States cannot be explained as a methodological artifact, i.e. a ceiling effect, as Scandinavian countries have far lower levels of traditionality. Hungary appears to be a special case, with traditional attitudes coming to the fore after the collapse of socialism but now becoming more liberal. A wider look at the development of other post-socialist societies suggests that there is a marked liberal trend, particularly in countries where attitudes have been most traditional. Unfortunately, ISSP data do not allow us to examine whether most former socialist countires experienced a temporary traditional blip, in the wake of the breakdown of socialism. A main focus of our paper is to understand the importance of cohorts for attitudinal change. We examine how much change is due to aging of period effects that influence all cohorts in a similar way, and how much is due to cohort replacement