Between 1992 and 1996, nonskilled workers' level of support for social welfare declined slightly. While the magnitude of this change is itself quite small (.10 on the seven-point NES welfare state item), it nevertheless reveals that nonskilled workers' experiences under a Democratic administration did nothing to stem their small but significant long-term decline in support for the welfare state. The complementary political effects of these changes suggest that nonskilled workers' shift away from an historically strong alignment with the Democratic Party is more than a temporary fluctuation.
How have changes in the race, religion, class, and gender cleavages affected the total social cleavage? We examine this question in Figure 8.3, which displays the lambda index scores for each of the ten elections. We also recalculate the index by ignoring race, allowing us to examine whether we obtain a different picture of trends in the absence of ongoing changes affecting the powerful racial cleavage.
The darker line shows the over-time development of the total social cleavage since 1960. The magnitude of the total social cleavage increased slightly between 1992 and 1996 (from .13 to .14). Given the 1960 index score of .11, the total social cleavage has grown in magnitude during this 36-year period. When we ignore the race cleavage, our estimates (plotted by the lighter line) reveal far less change between 1964 and 1996, but no evidence of a decline in magnitude. Because these estimates ignore race, the disproportionately large religious cleavage in 1960 has a considerably greater impact on the magnitude of the overall social cleavage, thereby leading to an inflated index score for that year. However, as discussed in Chapter 4, these features were unique to this particular election. Taken together, these results provide clear evidence for the ongoing--and even increasing significance--of social cleavages.
The analyses developed in this chapter provide a useful vantage point from which to understand long-term patterns of change in social cleavages as well as some recent developments. With regard to group- specific patterns of change, African-Americans and Jewish voters remain disproportionately Democratic in their political alignments, whereas conservative Protestants and non-professional employers