(CENTER ON PAGE)

The Effects of Race, Sex, and Individual Ideology on

People’s Opinions of Abortion

 

Justin Starling

 

(NEW PAGE)

Dr. Shaffer

Political Analysis

Spring Semester 2006

 

Introduction

            Throughout the history of American political debate, few issues have been as divisive as abortion. Abortion is the termination of a pregnancy before birth, resulting in the death of the fetus. In 1973 the Supreme Court decided in the case of Roe v. Wade that a woman has the constitutional right to an abortion during the first six months of pregnancy. Before Roe v. Wade, abortions were performed illegally and were considered high risk procedures. According to the Planned Parenthood website, illegal abortions caused no less than 193 deaths of women in 1965 and represented seventeen percent of all pregnancy deaths (www.plannedparenthood.org). According to the National Right to Life website, over 47 million abortions have been performed since 1973, the year abortions became legal (www.nrlc.org). The constitutionality and morality of abortion continue to be debated within American society and government. The issue of abortion is even helping to shape the membership of all three branches of American government. Abortion has a major impact on how many groups within our society vote. The nomination and confirmation process for judicial appointments has often become a debate over abortion. It has become a defining issue for conservative, who commonly oppose abortion, and liberals, who commonly support a woman’s right to choose whether to have an abortion. Public policies and political culture are greatly influenced by this debate, so it is imperative to understand the factors that mold the debate.

            The purpose of this research paper is to show that the issue of abortion is indeed a divisive social, moral, and philosophical issue that questions both life and humanity. In this paper, I plan to analyze and reveal how variables such as race (white/black), sex (male/female), and ideology (conservative/liberal) affect people’s opinions on abortion. I plan to expose which variables help shape people’s attitudes concerning abortion and which variables lead people to support or oppose abortion.    

            My initial expectations are that blacks are more likely to support abortion compared to whites, females are more likely to support abortion compared to males, blacks are more likely to be liberal compared to whites, females are more likely to be liberal compared to males, and people who are liberal are more likely to support abortion compared to conservatives.

 

Model and Hypotheses:

Assume that race and sex are the earliest independent variables, ideology is the

 

intervening variable, and abortion is the dependent variable.

 

 

                                    (H1)

Race

(Black)       (H3)

 

                                                                 

                                                Ideology          (H5)                 Abortion                                 

                   (H4)                      (Liberal)                                   (Support)

 

 

Gender

(Female)                       (H2)

 

 

The hypotheses are:

 

            Hypothesis 1: Blacks are more likely to support abortion compared to whites.

 

            Hypothesis 2: Females are more likely to support abortion compared to males.

 

            Hypothesis 3: Blacks are more likely to be Liberal in self- identification compared to whites.

 

            Hypothesis 4: Females are more likely to be Liberal in self- identification compared to males.

 

            Hypothesis 5: Liberal self- identifiers are more likely to support abortion compared to Conservative self- identifiers.

 

           

Literature Review

 

Blacks are more likely to support abortion compared to whites.

            The article “Blacks, Whites, and Attitudes Toward Abortion” examines the significance of race, religiosity, and other demographic characteristics on abortion attitudes (Combs and Welch, 1982). The study also attempted to track any changes in the salience of race on abortion attitudes during a nine-year period. The study used data from the National Opinion Research Center’s General Social Survey during the period of 1972-1980.  

            Racial differences were computed after adding controls for demographic and religious factors. Based on the study, blacks were less likely to support abortion compared to whites, but the racial gap had narrowed since 1976. Demographic characteristics and to a greater degree religiosity accounted for much of the racial gap. Factors such as education level, income, urban or rural residence, Southern or Northern residence, and religiosity that were hypothesized to affect blacks’ attitudes on abortion appeared to do so (Combs and Welch, 1982).

            The article “Race Differences in Abortion Attitudes: Some Additional Evidence” examines the differences in abortion attitudes between whites and blacks and studies the effects of social and cultural variables on those attitudes (Wilcox, 1990). The study was performed to test the theory that racial differences in abortion attitudes are declining. For the study, data from the National Opinion Research Center’s General Social Survey was examined from the period of 1972-1988. In the study, frequently used demographic variables such as income, education level, gender, age, and Southern and urban residence were used. Measures of religiosity were introduced in the 1984 survey. These measures included frequency of prayer, frequency of church attendance, strength of attachment to denomination, and interpretation of the Bible.

            Based on the study, blacks continue to be less likely to support abortion compared to whites, but racial differences in abortion attitudes are declining. According to the study, the decline in racial differences is not a product of blacks becoming more supportive of abortion as their socioeconomic statuses improve, but instead is a product of whites becoming less supportive of abortion. The study also reveals that blacks are not significantly different from whites in their support of abortion once the new religious variables in the multivariate analysis are introduced (Wilcox, 1990). 

            These studies suggest that my first hypothesis will be rejected by the data, but it is important to test the data since it has never been done in Mississippi alone.

 

Females are more likely to support abortion compared to males.

            The article “College Student’s Attitudes Toward Abortion and Commitment to the Issue” examines abortion attitudes, commitment, and abortion experience among males and females (Carlton et al., 2000). The study was conducted with 1118 students (350 males and 663 females) from a midsized southeastern university. The students participated in the survey to fulfill a requirement for a general psychology course.

            The study indicated that females were slightly less likely to support abortion compared to males. The study also found that females are more committed to the abortion issue which indicates that females are more knowledgeable on the issue, more likely to have an opinion, and more likely to act on their views. However, men are not disconnected from the issue because they are affected as partners, friends, fathers, relatives, and health providers (Carlton et al., 2000).

            The article “Conflicting Beliefs About Abortion: Legal Approval and Moral Doubts” examines the relationship between legal approval and moral concern in the abortion issue (Scott, 1989). For the study, data were used from a random digit dialed national cross-section telephone survey conducted during the summer of 1986 at the Survey Research Center of the University of Michigan.

            Based on the study, there is little gender difference in attitudes towards the legality of abortion. Despite little gender difference in attitudes towards legality, there is a stark difference in attitudes concerning morality. Sixty- eight percent of women regard abortion as wrong or wrong in some cases, but only 52% of men classify abortion as wrong or wrong in some cases. Pro-choice respondents account for most of this gender difference due to the fact that 46% of women favor legal abortion despite their feeling that abortion is morally questionable compared to 29% of men. This gender difference remains even after education level, religious preference, and church attendance are controlled (Scott, 1989).     

 

Blacks are more likely to be Liberal in self- identification compared to whites.

            The article “Exploring the Racial Divide: Blacks, Whites, and Opinion on National Policy” examines racial differences in public opinion and studies factors that contribute to the racial divide in public opinion (Kinder, 2001). For the study, data were primarily gathered from the 1992 American National Election Study, which contained a sample of 1,883 whites and 289 blacks. The questions were designed to gather information on national policy attitudes, demographics, and principles.

            Based on the study, there are two racial divides: over racial policies and the welfare state. Concerning the welfare state, blacks are more likely to be liberal compared to whites. Blacks wanted more government services and federal spending than did whites. The divide over the welfare state was not as large as the one over racial policies. The divide over racial policies is huge. Blacks are concerned with equality and feel connected to their racial group while whites are worried about big government and commonly feel racial resentment. The analysis does suggest that the racial divide would be reduced dramatically if differences in principles and identification could be eliminated (Kinder, 2001).

            The article “Race and Ideology: A Research Note Measuring Liberalism and Conservatism in Black America” examines black support for conservative positions on social and economic issues compared to whites (Seltzer and Smith, 1985). For the study, data were collected from the National Opinion Research Center’s 1982 General Social Survey. This survey included a special over sampling of blacks, resulting in 1,323 whites and 510 blacks. Issues that construct the current liberal-conservative spectrum in America were analyzed. Differences in education and region were then analyzed for their possible role in the racial differences.

            Based on the study, blacks are much more likely to be liberal compared to whites on issues of government spending on social programs, military spending, and most socio-culture issues. Blacks with a higher education and larger income tend to be more liberal than poor, less educated blacks. Northern blacks also tend to be more liberal than Southern blacks. Despite blacks being more liberal in general, they hold conservative attitudes on several socio-cultural issues such as abortion on demand, stricter law enforcement, prayer in school, and homosexual rights (Seltzer and Smith, 1985).           

 

Females are more likely to be Liberal in self- identification compared to males.

            The article “Identity Politics and Local Political Culture” examines identity politics and political culture in 30 United States communities (DeLeon and Naff, 2004). For the study, data were collected from the Social Capital Benchmark Survey. During the period of July to November 2000 a national sample of 3,003 adults and independent samples of adults in 40 communities (the study only used 30 of the 40 cities) were collected.

            The national survey reveals a gender gap in political ideology. Nationally, women are slightly more likely to be liberal compared to men. The study showed that location also affects political ideology. In 3 of the 30 communities surveyed, women were less likely to be liberal compared to men (DeLeon and Naff, 2004).

            The article “The Changing Politics of American Men: Understanding the Sources of the Gender Gap” examines gender differences in partisanship and presidential voting to determine the sources of the gender gap (Kaufman and Petrocik, 2004). For the study, data were gathered from the National Election Study during the period of 1952-1996. The gender gap had become a well documented electoral fact by the 1980’s. Women were more likely to be liberal compared to men.

            The study confirmed the fact that a gender gap does exist and brought new evidence regarding the causes of the gap between men and women. The gender gap was thought to be the product of women moving to the left, but the study showed it was a product of changing politics in men. Men have become increasingly conservative since the mid 1960’s while women have remained fairly constant. The study sheds new light on this “female centric” issue. The study also noted that only race, social class, religion, and religiosity rival gender in the ability to predict party preference and voting (Kaufman and Petrocik, 2004).   

 

Liberal self- identifiers are more likely to support abortion compared to Conservative self- identifiers.

            The article “Abortion Politics in the United States, 1972-1994: From Single Issue to Ideology” examined the politicization of the abortion issue and the cultural, ideological, social, and demographical factors that shape the polarization of this issue (Hout, 1999). For the study, data were gathered from the National Opinion Research Center’s General Social Survey during the period of 1972-1994.

            Based on the study, at the time Roe v Wade was decided by the Supreme Court, both conservatives and liberals were divided over the issue. This would not be the case by the late 1990’s. Conservatives had moved toward a more pro-life stance while liberals had moved toward more support for abortion. Some people had even begun to identify themselves as liberal or conservative based solely on the abortion issue. There is a direct correlation between abortion attitudes and party identification. The more likely a person is to support abortion, the more likely he or she is to identify with the Democratic Party. The more likely a person is to oppose abortion, the more likely he or she is to identify with the Republican Party (Hout, 1999).

            The article “Attitudes Toward Abortion, Religion, and Party Affiliation among College Students” examines potential factors affecting abortion attitude formation (Hess, 2005). The study was conducted by interviewing 392 undergraduate students (278 female and 114 male) at a medium sized Midwestern university. The survey contained forty questions concerning abortion attitudes, religious attitudes, and other factors potentially contributing to stances on the issue of abortion.     

            Based on the study, Democrats expressed more support for abortion than did Republicans. As expected, liberals expressed more support for abortion than did conservatives and at a rate nearly twice that associated with party identification. Three major factors that influenced abortion attitudes were presented: a person’s conceptualization of the “beginning of life,” religious practices, and knowledge and support of an individual who has undergone an abortion (Hess, 2005).

 

 Methods

            To test my model, I used information drawn from The Mississippi Poll project, a series of statewide public opinion polls conducted by the Survey Research Unit of the Social Science Research Center (SSRC) at Mississippi State University and directed by political science professor Stephen D. Shaffer. In order to maximize my sample size and therefore minimize my sample error, I combined or pooled telephone surveys conducted in two years-- 2000 and 2002. The 2000 Mississippi Poll surveyed 613 adult Mississippi residents from April 3 to April 16, 2000 and had a response rate of 49%, while the 2002 Mississippi Poll surveyed 608 adult Mississippi residents from April 1 to April 14, 2002 for a response rate of 50%. The two years combined contained 1221 respondents.  With 1221 respondents interviewed, the sample error is 2.9%, which means that if every adult Mississippian had been interviewed, the results could differ from those reported here by as much as 2.9%. The pooled sample was adjusted or weighted by demographic characteristics to ensure that social groups less likely to answer the surveys or to own telephones were represented in the sample in rough proportion to their presence in the state population. In both years, a random sampling technique was used to select the households and each individual within the household to be interviewed, and no substitutions were permitted. The SSRC's Computer Assisted Telephone Interviewing System (CATI) was used to collect the data.

            I relied on four variables included in both years of the Mississippi Poll. Sex was measured in a very straightforward way as a dichotomous variable. Race was also measured in a straightforward way, asking respondents to label themselves as white, black, or other. Ideology was a self-identification question, asking respondents the following question: "What about your political beliefs? Do you consider yourself very liberal, somewhat liberal, moderate or middle of the road, somewhat conservative, or very conservative?"  Support for abortion was measured by asking respondents if they agreed with the statement, "By law, a woman should be able to have an abortion as a matter of personal choice." Respondents were asked to choose between strongly disagree, disagree, undecided or don’t know, agree, or strongly agree categories. 

            In order to have enough people to analyze in the multivariate analyses, I recoded or combined categories of two of the variables. Five ideological self-identification categories were combined into three groups-- liberals included those considering themselves as "very" or "somewhat" liberal, conservatives were those identifying themselves as "somewhat" or "very" conservative, and the middle category of "moderate/middle of the road" constituted an intermediate "moderate" grouping. Five abortion support categories were combined into three groups—abortion supporters included those who “strongly agree” or “agree” with the before mentioned statement, abortion opponents included those who “strongly disagree” or “disagree” with the statement, and the middle category of “undecided or don’t know” constituted an intermediate group. Sex and race did not have to be recoded          

 

 FINDINGS- BIVARIATE

 

 

Table 1

 

Race Difference in Support for Abortion

 

 

 

Abortion

 

Race

 

White

 

 

Black

Pro-Life

 

54.7%

52.2%

Undecided

 

4.8%

2.3%

Pro-Choice

 

40.5%

45.4%

N Size

768

383

 

Gamma= .07

Chi-Square > .05

Note: Percentages total 100% down each column

Source: 2000 and 2002 Mississippi Polls, conducted by Mississippi State University

 

The first hypothesis of my model states that blacks are more likely to support abortion compared to whites.  In the 2000 and 2002 Mississippi Poll, 52.2% of blacks indicated they were Pro-Life or against abortion, compared to 54.7% of whites who indicated they were Pro-Life. The magnitude of the relationship is only 2.5%, which is the percentage difference between whites and blacks in the Pro-Life position.  The magnitude, measured by the gamma value, that reflects the relationship between race and support for abortion is a mere .07, which is only a slight indication that blacks are more likely to support abortion compared to whites.  Furthermore, the chi-squared statistic is not significant at the .05 level, indicating that the relationship between race and support for abortion, found in the 2000 and 2002 statewide polls, cannot be generalized to the entire population. Thus, my hypothesis that blacks are more likely to support abortion compared to whites is rejected. 

 

Table 2

 

Gender Difference in Support for Abortion

 

 

 

Abortion

Gender

 

Male

 

 

 

Female

Pro-Life

 

57.2%

51.2%

Undecided

 

3.7%

4.2%

Pro-Choice

 

39.2%

44.6%

N Size

544

644

 

Gamma= .112

Chi-Square > .05

Note: Percentages total 100% down each column

Source: 2000 and 2002 Mississippi Polls, conducted by Mississippi State University

 

 

Hypothesis 2 of my model states that females are more likely to support abortion compared to males.  In the 2000 and 2002 Mississippi Poll, 51.2% of females indicated that they were Pro-Life, compared to 57.2% of males who indicated they were Pro-Life. The magnitude of the relationship is only 6%, which is the percentage difference between males and females in the Pro-Life position. The gamma value reflecting the relationship between sex and support for abortion is .112, which is only a slight indication that females are more likely to support abortion compared to males.  Furthermore, the chi-squared statistic is not significant at the .05 level, indicating that the relationship between gender and support for abortion, found in the 2000 and 2002 statewide polls, cannot be generalized to the entire population. Thus, my hypothesis that females are more likely to support abortion compared to males is rejected. 

 

Table 3

 

Race and Ideology Differences

 

 

 

Ideology

Race

 

White

 

 

 

Black

Liberal

 

17.3%

34.5%

Moderate

 

31.6%

33.3%

Conservative

 

51.1%

32.2%

N Size

 

746

351

Gamma= .309

Chi-Square < .001

Note: Percentages total 100% down each column

Source: 2000 and 2002 Mississippi Polls, conducted by Mississippi State University

 

 

Hypothesis 3 of my model states that blacks are more likely to be liberal in self- identification compared to whites.  In the 2000 and 2002 Mississippi Poll, 34.5% of blacks identified themselves as liberal, compared to 17.3% of whites who identified themselves as liberal. The magnitude of the relationship is 17.2%, which is the percentage difference between the races in liberal ideology. The gamma value reflecting the relationship between race and ideology is .309.  Furthermore, the chi-squared statistic is significant at the .001 level, indicating that the relationship between race and ideology, found in the 2000 and 2002 statewide polls, can be generalized to the entire population. Thus, my hypothesis that blacks are more likely to be liberal in self- identification compared to whites is accepted.

 

Table 4

 

Gender Differences in Ideology

 

 

 

Ideology

Gender

 

Male

 

 

 

Female

Liberal

 

21.0%

25.7%

Moderate

 

31.8%

31.7%

Conservative

 

47.2%

42.6%

N Size

 

525

606

Gamma= -.096

Chi-Square > .05

Note: Percentages total 100% down each column

Source: 2000 and 2002 Mississippi Polls, conducted by Mississippi State University

 

 

            Hypothesis 4 of my model states that females are more likely to be liberal in self-identification compared to males.  In the 2000 and 2002 Mississippi Poll, 25.7% of females identified themselves as liberal, compared to 21% of males who identified themselves as liberal. The magnitude of the relationship is only 4.7%, which is the percentage difference between the sexes in liberal ideology. The gamma value reflecting the relationship between gender and ideology is a mere -.096. The chi-squared statistic is not significant at the .05 level, indicating that the relationship between gender and ideology, found in the 2000 and 2002 statewide polls, cannot be generalized to the entire population. Thus, my hypothesis that females are more likely to be liberal in self-identification compared to males is rejected.

 

Table 5

Ideology Difference in Support for Abortion

 

 

 

Abortion

Ideology

 

Liberal

 

 

Moderate

 

 

 

Conservative

Pro-Life

 

38.5%

47.7%

65.7%

Undecided

 

5.7%

3.2%

3.4%

Pro-Choice

 

55.8%

49.1%

30.9%

N Size

 

265

342

495

 Gamma= -.349

Chi-Square < .001

Note: Percentages total 100% down each column

Source: 2000 and 2002 Mississippi Polls, conducted by Mississippi State University

 

 

            Hypothesis 5 of my model states that liberal self identifiers are more likely to support abortion compared to conservative self identifiers.  In the 2000 and 2002 Mississippi Poll, 65.7% of conservatives were Pro-Life, compared to 38.5% of liberals. The magnitude of the relationship is 27.5%, which is the percentage difference between conservatives and liberals in the Pro-Life position. The gamma value reflecting the relationship between ideology and support for abortion is -.349. Furthermore, the chi-squared statistic is significant at the .001 level, indicating that the relationship between ideology and support for abortion, found in the 2000 and 2002 statewide polls, can be generalized to the entire population. Thus, my hypothesis that liberal self identifiers are more likely to support abortion compared to conservative self identifiers is accepted.

 

FINDINGS- MUTIVARIATE

 

Table 6

 

Race and Ideology in Support for Abortion

 

(Liberals Only)

 

 

 

Abortion

Race

 

White

 

 

 

Black

Pro-Life

 

33.9%

43.8%

Undecided

 

7.9%

3.3%

Pro-Choice

 

58.3%

52.9%

N Size

127

 

121

 Gamma= -.142

Chi-Square > .05

Note: Percentages total 100% down each column

Source: 2000 and 2002 Mississippi Polls, conducted by Mississippi State University

 

Table 7

 

Race and Ideology in Support for Abortion

 

(Moderates Only)

 

 

 

Abortion

Race

 

White

 

 

 

Black

Pro-Life

 

49.3%

44.8%

Undecided

 

4.5%

0.9%

Pro-Choice

 

46.2%

54.3%

N Size

 

221

116

Gamma= .124

Chi-Square > .05

Note: Percentages total 100% down each column

Source: 2000 and 2002 Mississippi Polls, conducted by Mississippi State University

 

 

Table 8

 

Race and Ideology in Support for Abortion

 

(Conservatives Only)

 

 

 

Abortion

Race

 

White

 

 

 

Black

Pro-Life

 

65.0%

67.0%

Undecided

 

4.0%

1.8%

Pro-Choice

 

31.0%

31.3%

N Size

 

371

112

Gamma= -.028

Chi-Square > .05

Note: Percentages total 100% down each column

Source: 2000 and 2002 Mississippi Polls, conducted by Mississippi State University

 

            Tables 6-8 are multivariate tables that control for the intervening variable of ideology. The findings show that among the individuals who were liberal, moderate, and conservative, race is not an important factor in determining an individual’s support for abortion. Neither race is consistently or significantly more or less supportive of abortion compared to the other race. Among self-identified liberals, 33.9% of whites were Pro-Life compared to 43.8% of blacks (Table 6). The magnitude of the relationship between race and abortion attitudes among self- identified liberals is only 9.9%. The gamma value reflecting the relationship between race and support for abortion among liberals is only    -.142.  So, the chi-squared statistic is not statistically significant at the .05 level.  

            Among self-identified moderates, 49.3% of whites are Pro-Life, compared to 44.8% of blacks (Table 7). The magnitude of the relationship is only 4.5%, the percentage difference between moderate, Pro-Life whites and blacks. The gamma value reflecting the relationship between race and support for abortion among moderates is only .142.  So, the chi-squared statistic is not statistically significant at the .05 level.

            Among self-identified conservatives, 65% of whites are Pro-Life, compared to 67% of blacks (Table 8).  The magnitude of the relationship is only 2%, the percentage difference between conservative, Pro-Life whites and blacks. The gamma value reflecting the relationship between race and support for abortion among conservatives is a mere -.028.  Furthermore, the chi-squared statistic is not statistically significant at the .05 level.

            Although the multivariate tables 6-8 do not show a strong relationship between race and abortion attitudes, they do reveal ideology as an important factor in determining an individual’s support for abortion. Conservative whites are more likely to be Pro-Life compared to liberal whites.  For example, 65% of conservative whites are Pro-Life compared to only 33.9% of liberal whites. Furthermore, 31% of conservative whites are Pro-Choice compared to 58.3% of liberal whites. 

            As the multivariate tables 6-8 show, conservative blacks are more likely to be Pro-Life compared to liberal blacks.  For example, 67% of conservative blacks are Pro-Life compared to 43.8% of liberal blacks, and 31.3% of conservative blacks are Pro-Choice compared to 52.9% of liberal blacks. As individuals of either race become more conservative, they become more likely to be Pro-Life.   

 

Table 9

 

Gender and Ideology in Support for Abortion

 

(Liberals Only)

 

 

 

Abortion

Gender

 

Male

 

 

 

Female

Pro-Life

 

34.2%

41.6%

Undecided

 

9.0%

3.2%

Pro-Choice

 

56.8%

55.2%

N Size

 

111

154

Gamma= -.077

Chi-Square > .05

Note: Percentages total 100% down each column

Source: 2000 and 2002 Mississippi Polls, conducted by Mississippi State University

 

 

Table 10

 

Gender and Ideology in Support for Abortion

 

(Moderates Only)

 

 

 

Abortion

Gender

 

Male

 

 

 

Female

Pro-Life

 

52.5%

43.5%

Undecided

 

1.9%

4.3%

Pro-Choice

 

45.6%

52.2%

N Size

158

184

 

Gamma= .151 

Chi-Square > .05

Note: Percentages total 100% down each column

Source: 2000 and 2002 Mississippi Polls, conducted by Mississippi State University

 

 

Table 11

 

Gender and Ideology in Support for Abortion

 

(Conservatives Only)

 

 

 

Abortion

Gender

 

Male

 

 

Female

 

Pro-Life

 

67.5%

63.9%

Undecided

 

2.9%

4.0%

Pro-Choice

 

29.6%

32.1%

N Size

 

243

252

Gamma= .071 

Chi-Square > .05        

Note: Percentages total 100% down each column

Source: 2000 and 2002 Mississippi Polls, conducted by Mississippi State University

 

 

            Tables 9-11 are also multivariate tables that control for the intervening variable of ideology. The findings show that among the individuals who were liberal, moderate, and conservative, gender is not an important factor in determining an individual’s support for abortion. Neither gender is consistently or significantly more or less supportive of abortion compared to the other gender. Among self-identified liberals, 34.2% of males were Pro-Life compared to 41.6% of females (Table 9). The magnitude of the relationship is only 7.4%, the percentage difference between liberal, Pro-Life males and females. The gamma value reflecting the relationship between gender and support for abortion among liberals is a mere -.077.  Furthermore, the chi-squared statistic is not statistically significant at the .05 level. 

            Among self-identified moderates, 52.5% of males are Pro-Life, compared to 43.5% of females (Table 10). The magnitude of the relationship is only 9%, the percentage difference between moderate, Pro-Life males and females. The gamma value reflecting the relationship between race and support for abortion among moderates is only .151. Therefore, the chi-squared statistic is not statistically significant at the .05 level.

            Among self-identified conservatives, 67.5% of males are Pro-Life, compared to 63.9% of females (Table 11).  The magnitude of the relationship is only 3.6%, the percentage difference between conservative, Pro-Life males and females. The gamma value reflecting the relationship between race and support for abortion among conservatives is a mere .071.  Furthermore, the chi-squared statistic is not statistically significant at the .05 level.

            Tables 9-11 do not show a significant relationship between gender and abortion attitudes. However, these tables do reveal ideology as an important factor in determining an individual’s support for abortion, just as tables 6-8, examining race differences, did. Conservative males are more likely to be Pro-Life compared to liberal males.  For example, 67.5% of conservative males are Pro-Life compared to 34.2% of liberal males, and 29.6% of conservative males are Pro-Choice compared to 56.8% of liberal males. 

            The multivariate tables 9-11 also show that conservative females are more likely to be Pro-Life compared to liberal females.  For example, 63.9% of conservative females are Pro-Life compared to 41.6% of liberal females, and 32.1% of conservative females are Pro-Choice compared to 55.2% of liberal females. As individuals of either gender become more conservative, they become more likely to be Pro-Life.     

 

CONCLUSION

 

(H3)                                                         (H5)

            Race                                              Ideology                                                 Abortion

            (Black)                                            (Liberal)                                                 (Support)

 

           

            The critical analysis, performed in this paper, of the effects of race (white/black), gender (male/female), and ideology (conservative/liberal) on people’s opinions of abortion has shown that not all of the variables play a significant role in determining abortion attitudes. The effect that gender has on ideology and abortion attitudes was found to be statistically insignificant. The direct effect that race has on abortion attitudes was also found to be statistically insignificant.

            Only two of the relationships examined in the bivariate and multivariate tables were significant. The relationships between race and ideology and between ideology and abortion attitudes were significant. Only two of five hypotheses were accepted, Hypothesis 3 (blacks are more likely to be liberal in self-identification compared to whites) and Hypothesis 5 (liberal self-identifiers are more likely to support abortion compared to conservative self-identifiers).

            The articles “Exploring the Racial Divide: Blacks, Whites, and Opinions on National Policy” and “Race and Ideology: A Research Note Measuring Liberalism and Conservatism in Black America” support Hypothesis 3 with similar findings on a national scale. The latter article also offers a possible explanation of the lack of a significant race difference in abortion attitudes. The article contends that blacks hold relatively conservative attitudes on several socio-cultural issues. The analysis in this paper does show that blacks are more likely to be liberal in self- identification compared to whites. A portion of analysis in this paper is contradictory to the analysis found in the literature review. Though not statistically significant, my analysis shows that blacks are more likely to be pro- choice compared to whites.

            The article “Abortion Politics in the United States, 1972-1994: From Single Issue to Ideology” supports Hypothesis 5 with similar findings on a national scale, and the article “Attitudes Toward Abortion, Religion, and Party Affiliation among College Students” supports Hypothesis 5 with similar findings from a medium sized Midwestern university. The first article emphasizes the importance of abortion as an issue and concern in society. The analysis in this paper does show that liberal self- identifiers are more likely to support abortion compared to conservative self- identifiers.

            The analysis in this paper does not prove that three of the hypotheses were wrong but simply fails to be expansive enough to prove or disprove three of the five hypotheses. Because of the captivating qualities of this issue and its close connection with people’s perception of human life, further research is recommended. A barbaric act and the freedom of choice are at the heart of this issue. More research and education are needed to assist society in addressing the issue of abortion. 

             

 

References

 

Carlton, Casey L., Eileen S. Nelson, and Priscilla K. Coleman. 2002. “College Student’s Attitudes Toward Abortion and Commitment to the Issue.” Social Science         Journal, 37:4.

Combs, Casey L., and Susan Welch. 1982. “Blacks, Whites, and Attitudes Toward Abortion.” Public Opinion Quarterly, 46:4.

DeLeon, Richard E., and Katherine C. Naff. 2004. “Identity Politics and Local Political Culture.” Urban Affairs Review, 39:6.

Hess, Jennifer A. 2005. “Attitudes Toward Abortion, Religion, and Party Affiliation Among College Students.” Current Psychology, 24:1.

Hout, Michael. 1999. “Abortion Politics in the United States, 1972-1994: From Single Issue to Ideology.” Gender Issues, 17:2.

Kaufman, Karen M., and John R. Petrocik. 1999. “The Changing Politics of American Men: Understanding the Source of the Gender Gap.”  American Journal of             Political Science, 43:3.

Kinder, Donald R. 2001. “Exploring the Racial Divide: Blacks, Whites, and Opinions on National Policy.” American Journal of Political Science, 45:2.

Scott, Jacqueline. 1989. “Conflicting Beliefs About Abortion: Legal Approval and Moral Doubts.” Social Psychology Quarterly, 52:4.

Seltzer, Richard and Robert C. Smith. 1985. “Race and Ideology: A Research Note Measuring Liberalism and Conservatism in Black America.” Phylon Quarterly,       46:2.

Wilcox, Clyde. 1990. “Race Differences in Abortion Attitudes: Some Additional Evidence.” Public Opinion Quarterly, 54:2.