A study by two political scientists released today found that African-American voters in CA did not support Prop 8 to nearly the degree indicated by a much publicized (and here and here and many others) exit poll. According to a survey conducted last month of a random sample of voters, 58% of African-Americans voted for Prop 8, a higher rate that the 52% support it received overall, but far below the 70% figure found by the CNN exit poll. That exit poll result, the authors say, was an outlier – not consistent with their follow-up poll, pre-election polls, or precinct-level analysis, which found 57 to 59% support for Prop 8 in heavily African-American precincts.
The study, by Ken Sherrill (CUNY-Hunter College) and Patrick Egan (NYU), provides the clearest documentation yet of how various segments of the electorate contributed to the outcome, with important ramifications for future campaigns.
What characteristics were most predictive of how an individual voted on Prop 8?
–> Party identification, political views, religiosity and age, in that order.
Religiosity explained much of the difference between African-American or Latino and white voters. Weekly attendance at religious services is far more likely for African-Americans (57%) and a bit more likely for Latinos (47%) than for whites (43%). Among persons who reported attending services at least once a week, support for Prop 8 was close to the same across all racial and ethnic groups.
However, among persons attending services less than weekly, there was a racial difference: in that group, the yes on 8 vote was 48% among African-Americans, 46% among Latinos, 36% among whites, and 33% among Asian-Americans. Therefore it would be wrong to use this new survey as the basis for leaping to the conclusion that there is no serious problem with racial disparity in support for same-sex marriage — there is. The lgbt movement needs to rethink its race analysis and politics – that lesson from Prop 8 is still quite valid. The level of support for Prop 8 among less religious black and Latino voters is sobering.
Far more significant than race and even more significant than religion was political ideology: 81% of Republicans voted yes, compared to 30% of Democrats; and 82% of self-characterized conservatives were yes voters, compared to 22% of liberals.
Using multi-variate analysis, Egan and Sherrill found that the percentage of votes affected by each characteristic was as follows:
- Party identification 15.2%
- Ideology 14.6%
- Religiosity 11.8%
- Age 8.7%
- Race/ethnicity 5.5%
- Gender 4.9%
- Knowing someone gay 0.6%
As is obvious from these figures, by far the least significant factor of those studied was whether the individual knew someone gay or lesbian. Overall, almost 75% of all voters said they had gay family members or friends, which means that most conservative and Republican voters who know gay people voted for Prop 8.
To me, this signals a need to adjust some of the old slogans — coming out per se is necessary but far from sufficient for achieving change. Milk was a good movie, but we're no longer living in that moment. The society has moved way past the time when only a small number of straight people realized that they knew a gay person. At least for conservative voters, on this particular issue, the impact of having someone they know come out or of seeing appealing lgbt families on tv is probably insignificant. It would be interesting to know whether having gay friends or family members has a bigger impact on voters for issues other than marriage, or whether it's just lost some of its political oomph.
The effect that all those years of coming out has had up to now is quite real, though, and evident from the study. By comparing data from 2008 to data from the 2000 election in California, when voters adopted a statute limiting marriage to heterosexuals by a 61% majority, the authors were able to map the change in voting patterns within birth cohorts. For example, among "boomers" – those born between 1940 and 1960 – support for legalizing same-sex marriage grew by 13 points between 2000 and 2008. For those born between 1960 and 1982, it grew by 4%. A majority in both of those groups voted against Prop 8. For voters born before 1940, however, there has been no shift in opinion.