itional nuclear family arrangements, the more likely they will oppose legalized abortion.
The flaw with these studies is that they focus on women's attitudes, without reference to how the same web of social variables effect men's attitudes. This study will attempt to rectify this flaw by examining the effects of one particular social variable on abortion attitudes, while controlling for gender. The dependent variable in question will be marital status, and two research questions will be addressed: First, are there differences in abortion attitudes by marital status? Second, do these differences hold across genders?
Hypotheses and variables. From the previous research literature discussed above, two general research questions were derived. From these research questions, two testable hypotheses can be derived.
H1: Married people will be more likely to support abortion restrictions than unmarried people.
H2: Married women will be more likely to support abortion restrictions than married men.
The dependent variable in each of these hypotheses is "abortion restrictions." This will be measured by responses to the following question from the 1988 General Social Survey (GSS): "Please tell me whether or not you think it should be possible for a pregnant woman to obtain a legal abortion if the woman wants it for any reason?" Respondents were instructed to answer either yes, no, or don't know/no answer. The don't know/no answer responses were treated as non-responses, resulting in a dichotomous variable.
The independent variable in both hypotheses is marital status. Marital status was measured as the response to following question on the 1988 GSS: "Are you married, widowed, divorced, separated, or have you never been married (choose only one)?" Responses were coded as 1 through 5, resulting in a nominal variable.
The second hypothesis contains a control variable, gender. Unlike the independent and dependent va...