![Sample our Social Sciences journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/110822/TOC-Subject-Sample-2021-Social-Sciences.jpg"})
Abstract
This article examines the legislative effectiveness of gay and lesbian state legislators. Legislative effectiveness is measured by legislators' perceptions of their effectiveness, their ability to block legislation, and their ability to see the bills they introduce pass. Although the number of openly gay and lesbian legislators is small, making this study exploratory, the findings are illuminating. Without controlling for legislators' relationships with colleagues or their ideology, lesbians, but not gay men, were less effective than other legislators, but after controlling for these factors lesbians were not significantly less effective than others.
Notes
1. Descriptive or passive representation involves a legislature resembling the make-up of the constituency. This is in contrast to substantive or active representation which involves working on behalf of others (CitationPitkin 1967).
2. This relates to a difficulty in using only the passage of bills introduced as a measure of effectiveness. Some minority legislators let others introduce and take the lead on legislation to increase the odds it passes.
3. The gay and lesbian legislators were identified from the Gay and Lesbian Victory Fund's Leadership Institute in April 2005. To this list I added Senator Carole Migden (CA), and removed Mitchell McKim who lost his last election. Just prior to mailing the survey, Republican Senator Paul Koering (MN) announced that he was gay. He was not included in the survey since he had not served as an openly gay legislator for more than a couple months as of the first mailing.
4. By 2008 that number had grown to 79 (calculated by author from the Gay and Lesbian Victory Fund's Leadership Institute collected in March 2008).
5. The national averages were 79% and 24% respectively.
6. The average gay or lesbian legislator was born in 1957 and had served 6.1 years, the average non-gay legislator was born in 1952 and served 5.8 years. For those responding, the average gay or lesbian legislator's district was 93.7% urban and 36.2% college educated compared to 89.0% and 33.0% for non-gay legislators.
7. Since there were three lesbian senators in California out of a total of 12 Democratic female legislators, all female Democratic state senators in California were sent the survey. This resulted in 162 not 159 legislators being included in the sample.
8. The average respondent was born in 1952, and had five years of experience, compared to 1955 and seven years for nonrespondents. Seniority was the only difference that was statistical significance at the .10 level. Of the respondents 33% were gay or lesbian, 40% female, 91% Anglo, 64% in majority party, and 26% in the Senate. This compares to 33%, 34%, 89%, 75%, and 37% respectively for nonrespondents. Since a key to the project was the pairing of gay and non-gay legislators, it is important to point out that there were only two cases where a gay or lesbian legislator from a state and chamber responded but a non-gay or lesbian did not respond. There were, however, 19 non-gay or lesbian legislators who responded from states that did not have a gay or lesbian respondent. I chose to include these cases in the analyses because they came from states that were similar to states that had a gay or lesbian respondent. For example, although there were no gay or lesbian respondents from New Hampshire but two non-gay or lesbian respondents, there were gay and lesbian respondents from neighboring Vermont and similarly small Delaware. To be sure the inclusion of these cases did not alter the conclusions I compared bivariate correlations between sex/sexual orientation and the dependent variables (included in and ) using a data set that included all the cases with a data set that did not include these 21 cases. The results were very similar. The largest difference was for the correlation between the non-lesbian women variable and ideology (.31; p = .02 compared to .21; .06). I also examined to make sure within the state that homosexual and heterosexual legislators were of the same sex. There were a couple cases that sex did not match up. Again, I examined the correlations and they were very similar to those of the other two data sets noted above.
9. The lead-in to these statements was the following instruction: “Using a 7-point scale, indicate the degree to which you agree or disagree with the following statements. Circle the number that corresponds to your answer. 1 = strong agreement and 7 = strong disagreement.”
10. Since these items were only moderately correlated (r = .33 and p = .00) I re-ran the equation using each indicator as the dependent variable. This produced similar conclusions.
11. Originally, I collected the number of bills passed. However, a dichotomous variable was used since the data were highly skewed (30% saw no bills pass). Additionally, the distribution of the data was such that using multinomial logistic regression produced unreliable results.
12. Examining whether gay and lesbian legislators were able to gain leadership positions would be another way to measure effectiveness. However, when gays and lesbians were matched to other legislators, nonleaders were matched with nonleaders: as such, this comparison would be ill-advised.
13. The item that measures whether the legislator enjoyed his or her colleagues differs from the other two. It involves the legislator's responses to colleagues' actions not their actions. Because of this I considered dropping it from the scale, but the alpha was larger when all three items were used. It is likely a combination of the treatment and the reaction to the treatment combined that influences effectiveness. The analyses were also conducted examining each item separately instead of the scale. This did not indicate that one item was consistently more strongly related to the dependent variables.
14. I use stata's prvalue command to calculate the probabilities.
15. The correlations for perceptions of general effectiveness with block bills was .18 (p = .10), .11 (p = .34) for bill passage. For bill passage and block bills it was .23 (p = .04).