Investigating Differences in Public Support for Gay Rights Issues

Abstract

Public opinion surrounding two of the most well-known gay rights issues—same-sex marriage and adoption by gays and lesbians—varies widely: About 30% of the public favors same-sex marriage, while about 50% favors adoption. The purpose of this study was to examine and compare the demographic variables that predict attitudes toward same-sex marriage and adoption by gays and lesbians. Political ideology and attendance at religious services emerged as the strongest predictors. However, a close examination of the two issues shows that there were important differences between them. These differences may aid gay rights advocates in crafting media campaigns designed to increase support for same-sex marriage.

KEYWORDS:

• marriage
• adoption
• public opinion
• gays
• lesbians
• demographics
• religion

Gay rights issues tend to be controversial topics that often make headlines and inspire public debate. Although taken as a whole, the American public has more favorable attitudes toward homosexuality than at any time in history (Avery et al., 2007), a great deal is yet to be learned about how the public thinks about homosexuality. Public support for specific gay rights issues is not uniform; in fact, it can vary widely, for example, public opinion surrounding two of the most well-known issues—same-sex marriage and adoption by gays and lesbians: About 30% of the public favors same-sex marriage, while about 50% favors adoption (“Less opposition,” 2006).

The purpose of this study is to examine and compare the demographic variables that predict attitudes toward same-sex marriage and adoption by gays and lesbians. This is important because investigating differences in support for these two issues will provide a more nuanced picture of how the public views homosexuality and gay rights in general. Additionally, this study is important because understanding these differences could aid gay rights advocates in securing civil rights for families headed by gays and lesbians. Research demonstrates that a large number of gay and lesbian couples have children (Gates & Ost, 2004) and that their families face significant economic, legal, and social disadvantages when they cannot marry (Herek, 2006; Perrin, 2002). Understanding the demographic differences between supporters or opponents of the two issues may help gay rights advocates target people who support adoption but not same-sex marriage with a media campaign explaining the value of marriage for gays and lesbians with adopted children. This could be especially valuable in states with ballot initiatives proposing to ban same-sex marriage.

Previous studies explored demographic variables that influenced attitudes toward homosexuality in general (Grapes, 2006; Herek, 1987, 2000; LaMar & Kite, 1998; Ohlander, Batalova, & Treas, 2005; Ratcliff, Lassiter, Markman, & Celeste, 2006; Wilkinson & Roys, 2005) same-sex marriage (Herek, 2002; Olson, Cadge, & Harrison, 2006) and adoption (Herek, 2002). However, they have either not (a) considered a wide range of demographic variables, or (b) compared same-sex marriage and adoption side-by-side. This study endeavors to fill this gap in the literature. Below, I review literature on public opinion and gay rights issues, focusing on studies that discovered variables that strongly predicted attitudes toward homosexuality, same-sex marriage, and adoption by gays and lesbians.

Previous research shows that gender, religiosity, education, political orientation, and age are important to consider in examining public opinion of homosexuality and gay rights issues. In the section below, I present literature that explores each of these variables.

GENDER AND HOMOSEXUALITY

Studies demonstrate that gender can predict attitudes toward gays and lesbians. Herek (2000), in a nationally representative telephone survey, found differences between male and female respondents when looking at individuals' feelings toward homosexuality. The results showed that male respondents viewed both gay men and lesbians significantly more negatively than female respondents. Ratcliff et al. (2006) found this same pattern in two experiments involving undergraduate students; LaMar and Kite (1998) had similar observations in an experiment with undergraduates.

In addition to research examining gender gaps in views toward gays and lesbians in general, one study also investigated the effect of gender on attitudes toward specific gay rights issues. Herek (2002) conducted a national survey asking respondents for their perspectives on same-sex marriage and adoption by gays and lesbians. He discovered that both male and female respondents opposed same-sex marriage about equally. However, female respondents were significantly more likely to support adoption for both gays and lesbians than male respondents.

One of the primary reasons for gender differences toward homosexuality is the concept of sexual prejudice, which involves individual's attitudes toward their own gender and sexuality (Herek, 1986, 2000, 2002). Herek (2000) explained that many heterosexuals experience same-sex attractions and become anxious because of homosexuality's stigmatized status in society. Additionally, because homosexuality is often associated with gender inversion (i.e., gay men are more like women than heterosexual men), heterosexuals are fearful of being labeled homosexual because such a label might suggest that they are not a “real” man or a “real” woman. For these reasons, some heterosexuals adopt homophobic attitudes to assure themselves and others that they are not (stigmatized) homosexuals. Herek (1986, 2000, 2002) pointed out that heterosexual men feel these pressures more than heterosexual women, thus, explaining why their attitudes toward homosexuality (particularly gay men) are more negative than women's.

RELIGIOSITY AND HOMOSEXUALITY

Religiosity is another variable that predicts attitudes toward gays and lesbians. Herek (1987) performed an experiment to assess the relationship between attitudes toward homosexuality and religiosity. Subjects completed the Religious Orientation Scale (which included measures of attendance at religious services) and Religious Ideology scales, as well as the Attitudes Toward Gays and Attitudes Toward Lesbians scales. The results demonstrated that subjects who were more religious held more negative attitudes toward both gays and lesbians. Wilkinson and Roys (2005) also found similar results in their two experiments involving undergraduates.

Olson, Cadge, & Harrison (2006) conducted a nationally representative survey in 2004 to examine the relationship between religiosity and attitudes toward same-sex marriage. Religiosity was measured primarily in terms of attendance at religious services. The results demonstrated that people who scored higher on measures of religiosity were more likely to hold negative attitudes toward same-sex marriage than those who scored lower on the religiosity scale. Olson et al. noted religiosity was more powerful than any other variable included in their study in predicting attitudes.

Religiosity is an important variable in predicting attitudes toward homosexuality and gay rights issues because religious institutions exert a significant influence on individuals' attitudes (Olson et al., 2006). If individuals are frequently exposed to messages in their religious institution condemning homosexuality, they are likely to have negative views of gays and lesbians. As Olson et al. pointed out, “congregations, regardless of their specific religious affiliation, can be highly cohesive political communities in which opinion tends to converge on certain issues, particularly those that are morally charged” (p. 356).

EDUCATION AND HOMOSEXUALITY

Research shows that education can predict attitudes toward homosexuality or gay rights issues. Ohlander et al. (2005) analyzed attitudes toward “homosexual relations” in a study involving the 1988–1995 General Social Surveys. They found that level of education impacted attitudes: People with a higher level of education held more positive attitudes toward gays and lesbians than those with a lower level of education. This effect was present even with controls for sex, age, income, marital status, religiosity, and political ideology. Grapes (2006) discovered similar results in her analysis of the 1998 General Social Survey.

Ohlander et al. (2005) determined the positive relationship between education and favorable attitudes toward gays and lesbians was a result of education's association with support for civil liberties and its connection to cognitive sophistication. They noted that education encourages people to tolerate differences and to view basic civil liberties as essential for everyone. Additionally, they explained that education increases individuals' cognitive abilities, which augments their reasoning skills: People are better able to contend with new ideas and are more comfortable evaluating information (i.e., arguments in favor of gay rights) in an unbiased manner.

OTHER VARIABLES AND HOMOSEXUALITY

Research demonstrates that identifying as a political liberal predicts support for gay rights issues. In an analysis of the 1992 and 1996 American National Election Studies, Brewer (2003) showed that liberals were more likely to have favorable attitudes toward allowing gays to serve in the military and laws protecting gays and lesbians from job discrimination. According to Haberberle (1999), this is not surprising because support for gay rights has traditionally been associated with liberals, while opposition has typically been associated with conservatives.

Age is another demographic variable that predicts attitudes toward homosexuality and gay rights issues. Avery et al. (2007) noted that older Americans tend to be less supportive of gay rights than younger Americans. They suggest that this is related to the fact that older people tend to have more conservative political orientations than younger people.

RESEARCH QUESTIONS

Given the literature reviewed above, it is not clear which demographic variables best predict attitudes toward same-sex marriage and adoption by gays and lesbians. Only one study examined attitudes toward both issues, but was primarily concerned with gender and did not provide information about other demographic variables (Herek, 2002). Other studies produced conflicting results about which variables are most important in predicting attitudes toward homosexuality or gay rights issues (Grapes, 2006; Herek, 1987, 2000; LaMar & Kite, 1998; Ohlander et al., 2005; Olson et al., 2006; Ratcliff et al., 2006; Wilkinson & Roys, 2005) Therefore, I posed the following research questions:

• RQ 1: Which demographic variables best predict attitudes toward same-sex marriage?

• RQ 2: Which demographic variables best predict attitudes toward adoption by gays and lesbians?

• RQ 3: What are the demographic differences between same-sex marriage and adoption?

METHOD

To answer the study's research questions, I performed a secondary analysis of the News Interest Survey conducted by Princeton Survey Research Associates International for the Pew Research Center for People and the Press. This nationally representative, weighted telephone survey was administered between March 8 and March 12, 2006. Its sample consisted of 1,405 adults living in the continental United States selected through random digit dialing. The response rate was 24%. Because the two questions relevant to this study's dependent variables were only included in form 1 of the survey, the total sample for this study included 710 respondents. Table 1 shows detailed information about this study's sample.

TABLE 1 Descriptive Statistics of the Sample

Categories	Levels	n
Sex	Male	274
	Female	308
Age (years)	18–29	73
	30–49	214
	50–64	165
	65 and older	130
Level of education	High school or less	245
	Some college or trade school	168
	College degree or higher	169
Political ideology	Conservative	239
	Moderate	239
	Liberal	104
Frequency of attendance at religious services	Never to a few times a year	244
	Once a month or more	338

Measurement of Demographic Variables

Demographic characteristics of the survey respondents served as this study's predictor variables. The demographics that I chose to include predicted attitudes toward homosexuality and/or gay rights issues in previous research, but their effect has not been considered in comparative analyses of same-sex marriage and adoption. The demographic variables I selected were: sex; age (18–29, 30–49, 50–64, 65 and older); level of education (high school or less, some college or trade school, college graduate or higher); how often they attended religious services (never to a few times a year, one a month or more); and their political ideology (conservative, moderate, or liberal). Because “don't know” and “refused to answer” were grouped together on the survey there was not a way to distinguish between the two answers. Therefore, I recoded any response of “don't know” and “refused to answer” as missing. These responses comprised less than 10% of these survey items.

Measurement of Gay Rights Variables

Attitudes toward same-sex marriage and adoption by gays and lesbians served as this study's dependent variables. The survey asked participants if they strongly opposed, opposed, favored, or strongly favored the following statements: (a) “allowing gays and lesbians to adopt children” and (b) “allowing gays and lesbians to marry legally.” Again, I recoded any response of “don't know” and “refused to answer” as missing. These responses comprised less than 10% of responses for both survey items.

RESULTS

This study examined demographic differences in attitudes toward same-sex marriage and adoption by gays and lesbians. A total of 642 people completed the survey question about same-sex marriage and 655 people completed the adoption item. The Pearson's correlation coefficient for the two issues was .74, p < .01. The mean for attitude toward same-sex marriage was 2.17 (SD = 1.0) and the mean for adoption was 2.31 (SD = 1.0). To determine if there was a significant difference in attitudes toward the two issues, I performed a paired samples t test. The results indicated that the mean for attitudes toward adoption was significantly greater than the mean for attitudes toward same-sex marriage, t(618) = 4.66, p < .001. The 95% confidence interval for the mean difference was .08–.20. Having established that respondents held significantly more positive attitudes toward adoption by gays and lesbians than same-sex marriage, I address the study's research questionsbelow.

Research Question 1 asked which demographic variables would predict attitudes toward same-sex marriage. To address this question, I conducted a multiple regression analysis with five demographic measures (sex, age, level of education, political ideology, and attendance at religious services) as predictors and attitude toward same-sex marriage as the dependent variable. The linear combination of demographic measures was significantly related to attitudes toward same-sex marriage, R ² = .30, F(5, 597) = 51.03, p < .001.

As the first panel of Table 2 shows, all predictor variables except sex were significant. Political ideology was the strongest predictor, followed by attendance at religious services. These results indicate that people who are liberals and people who never or rarely attend religious services were likely to hold more favorable attitudes toward same-sex marriage.

TABLE 2 Attitudes toward Same-Sex Marriage and Adoption by Gays and Lesbians Regressed on Predictors

	Attitudes toward same-sex marriage (N = 642) R ^2 = .30	Attitudes toward adoption by gays and lesbians (N = 655) R^2 = .23
Sex (0 = Male)	.07	.12***
Age (years) (1 = 18–29; 2 = 30–49; 3 = 50–64; 4 = 65+)	−.10**	−.07**
Level of education (1 = high school or less; 2 = some college or trade school; 3 = college degree or higher)	.10**	.13***
Political ideology (1 = conservative; 2 = moderate; 3 = liberal)	.33***	.34***
Frequency of attendance at religious services (1 = never to a few times a year; 2 = once a month or more)	−.31***	−.18***
** Significant at the .01 level.
*** Significant at the .001 level.
Coefficients are betas.

Research Question 2 addressed which demographic variables would predict attitudes toward adoption by gays and lesbians. I used multiple regression analysis to answer this question, employing the same five demographic measures as above as predictor variables and attitudes toward adoption by gays and lesbians as the dependent variable. The results were significant, R ² = .23, F(5, 1230) = 71.59, p < .001.

In this model, all of the predictor variables were significant, as depicted in the second panel of Table 2. Again, political ideology was the most powerful predictor. Attendance at religious services was the second most powerful predictor, but it was followed relatively closely by level of education. These data demonstrate that people who are liberal, who never or rarely attend religious services, and who have a relatively high level of education were most likely to hold positive views of adoption by gays and lesbians.

Research Question 3 concerned the difference in demographic patterns between same-sex marriage and adoption by gays and lesbians. To address this question, I conducted a repeated measures analysis of variation (ANOVA). The combination of the two variables “attitudes toward same-sex marriage” and “attitudes toward adoption by gays and lesbians” served as the dependent variable, which I called “gay rights.” The five demographic measures from the study's previous analysis served as the between subjects factors. The results indicated a significant difference between the two issues, F(1, 572) = 7.93, p < .01, partial η² = .02. There was also a significant interaction effect between “gay rights” and “attendance at religious services,” F(1, 572) = 9.23, p < .01, partial η² = .02.

The interaction between the two variables indicated that frequency of attendance at religious services predicted different attitudes toward same-sex marriage and adoption by gays and lesbians. The data show that people who never or rarely attend religious services viewed both same-sex marriage and adoption by gays and lesbians similarly (M = 2.60 and M = 2.62, respectively) but people who attend religious services more regularly felt significantly more positive about adoption (M = 2.11) than same-sex marriage (M = 1.83). The interaction between the two variables is shown in Figure 1. The means for all predictor variables for same-sex marriage and adoption are displayed in Table 3.

FIGURE 1 Interaction between gay rights and attendance at religious services.

TABLE 3 Means for Predictor Variables for Attitudes toward Same-Sex Marriage and Adoption by Gays and Lesbians

Predictors	Levels	Attitudes toward same-sex marriage	Attitudes toward adoption by gays and lesbians
Sex	Male	2.13	2.21
	Female	2.42	2.19
Age (years)	18–29	2.37	2.56
	30–49	2.30	2.38
	50–64	2.12	2.33
	65 and older	1.88	2.11
Level of education	High school or less	2.01	2.16
	Some college or trade school	2.15	2.29
	College degree or higher	2.39	2.59
Political ideology	Conservative	1.73	1.84
	Moderate	2.30	2.52
	Liberal	2.90	2.91
Frequency of attendance at religious services	Never to a few times a year	2.60	2.62
	Once a month or more	1.83**	2.11**
** Significant at .01 level.
(1 = strongly oppose; 2 = oppose; 3 = favor; 4 = strongly favor)

DISCUSSION

Previous research demonstrates that there are many variables to consider in predicting attitudes toward homosexuality and gay rights issues. The purpose of this study was to determine which variables would most strongly predict attitudes toward same-sex marriage and adoption by gays and lesbians and to compare the demographic patterns of support between the two issues. For both same-sex marriage and adoption by gays and lesbians, political ideology and attendance at religious services emerged as the strongest predictors. However, a close examination of the two issues shows that there were important differences between them. These differences may aid gay rights advocates in crafting media campaigns designed to increase support for same-sex marriage.

For both same-sex marriage and adoption by gays and lesbians, political ideology was the strongest predictor variable: Liberals were more likely to have favorable attitudes toward both issues than moderates and conservatives. Attendance at religious services was the second most powerful predictor for both issues. People who never or rarely attend religious services held more positive attitudes than those who regularly attend religious services. However, for same-sex marriage, attendance at religious services was almost as powerful in predicting attitudes as political orientation, but for adoption, it was about half as powerful. The other demographic variables included in this study—age and level of education—had a relatively small influence on attitudes for both issues. Sex had an effect on attitudes toward adoption, but not same-sex marriage, a finding consistent with Herek (2002).

The results of the repeated measures ANOVA confirmed the demographic differences between same-sex marriage and adoption by gays and lesbians. The significant interaction between the dependent variable (a combination of the variables “attitudes toward same-sex marriage” and “attitudes toward adoption”) and “attendance at religious services” indicated that the frequency with which people attend religious services is important to consider in predicting attitudes toward the two issues, a finding suggested, but not affirmed by, previous research (Herek, 1987; Olson et al., 2006; Wilkinson & Roys, 2005). The interaction demonstrated not only that people who regularly attend religious services hold more negative views of same-sex marriage and adoption than people who do not, but also that they feel significantly more positive about adoption by gays and lesbians than same-sex marriage.

Olson et al. (2006) explained that religious congregations are “highly cohesive political communities” that exert a substantial influence on their members (p. 356). It is important to point out that that the majority of this nationally representative sample (58%, n = 338) reported attending religious services more than once a month and that these individuals were likely to hold negative views of gay rights issues. For gay rights advocates to succeed in winning civil rights (i.e., the right to marry) for gays and lesbians, it is essential for them to concentrate their efforts on persuading people who frequently attend religious services that these rights are necessary. Targeting opinion leaders in religious congregations with appropriately framed persuasive messages may be one way of accomplishing this goal.

The finding that people who regularly attend religious services have significantly more positive feelings about adoption by gays and lesbians than same-sex marriage points to a general strategy for gay rights advocates: As a starting place, it may be efficacious to frame appeals for marriage rights to religious opinion leaders in terms of how legalized same-sex marriage would benefit the children of gay and lesbian couples. Messages should emphasize that many gays and lesbians are raising children (Gates & Ost, 2004) and, that across a broad array of social and psychological measures, their children are indistinguishable from children raised by heterosexuals (Perrin, 2002). Further, messages should stress that the children of gays and lesbians suffer because their parents cannot marry (Herek, 2006). As Herek pointed out, “children benefit when their parents (regardless of the latter's sexual orientation) are financially secure, physically and psychologically healthy, and not subjected to high levels of stress” (p. 616). Framing pro same-sex marriage messages in terms of children may be an effective way of influencing churchgoers to consider altering their position on same-sex marriage, as it builds on adoption, an issue about which they already feel relatively positive.

This study was not without its limitations. Conducting a secondary analysis constrained the variables that were available for analysis. For example, measures of media consumption would have added richness to my set of predictor variables, but such variables were not included in my data set. Additionally, the effect sizes for several of the results associated with research question three were quite small. While this is somewhat unsurprising given that same-sex marriage and adoption were so strongly correlated, their results should be interpreted with caution.

In spite of its limitations, this study provided potentially valuable information about the variables that best predict attitudes toward same-sex marriage and adoption by gays and lesbians. Moreover, it compared the two issues and was able to reveal subtle, yet important, differences between them. Future research should continue to investigate the climate of public opinion that surrounds same-sex marriage, adoption, and other gay rights issues to better understand how gays and lesbians are regarded in society. These insights could assist gay rights advocates in developing new and innovative strategies designed to advance the social, political and legal interests of gays and lesbians.