Developing Party Identification: A Further Examination of Life-Cycle, Generational, and Period Effects

Abstract

Recent analyses demonstrate that the partisan loyalties of birth cohorts have not increased over the full course of more than two decades, and these findings have been used to support a generational explanation for the weak party loyalties of young Americans. Philip E. Converse (1976) argues that partisan strength did increase with age between 1952 and 1964, which he views as a steady-state period in the development of party identification. Converse argues that there is considerable evidence for a life-cycle explanation during this period. Converse's support for the life-cycle explanation is a good deal weaker than he claims. By including blacks in his analysis, Converse mixed period effects with life-cycle effects. An analysis of the white electorate replicating Converse's techniques shows little gain in partisan strength among young whites during the 1952-1964 period, while an analysis including blacks suggests that Converse's inability to control for race led him to overestimate the presence of life-cycle effects. In addition, an analysis including results from the 1976 election survey conducted by the Center for Political Studies of the University of Michigan demonstrates that there is little support for the life-cycle thesis if one examines the entire 24-year period for which appropriate data are now available.