Chapter 2 : Asking and Answering Sociological Questions

In Chapter 2 you have learned about how sociologists expand the body of knowledge within the discipline by engaging in scientific research. In this exercise you will learn more about survey research, one of the principle methods used by sociologists, and will practice analyzing some survey data. The data you will be analyzing is taken from the General Social Survey (GSS). Ever since 1972 researchers at the National Opinion Research Center (NORC) of the University of Chicago have been surveying people from across the country in order to learn more about current trends; each year approximately 1,500 people are sampled and asked questions. As noted on the NORC website:

The basic purposes of the GSS are to gather data on contemporary American society in order to monitor and explain trends and constants in attitudes, behaviors, and attributes; to examine the structure and functioning of society in general as well as the role played by relevant subgroups; to compare the United States to other societies in order to place American society in comparative perspective and develop cross-national models of human society; and to make high-quality data easily accessible to scholars, students, policy makers, and others, with minimal cost and waiting.

Part I:

The first part of this exercise presents some basic descriptive information on attitudes about abortion at different points in time. One comparative question we can find an answer to by using this data is, "How have Americans' attitudes about abortion changed across time?" The first year in which GSS respondents were asked whether or not they thought it "should be possible for a pregnant woman to obtain a legal abortion if the woman wants it for any reason?" was in 1977; this question has been included in many of the surveys since that date. Table 1 provides information on the percentage of respondents who responded in the affirmative to this question in selected years.

Writing Assignment 1:

Consider the data in the above table. How have Americans' attitudes about abortion changed across time? Use your sociological imagination to develop an answer.

Part II:

While it is important to establish descriptive measures of data, ultimately what we want to do is explore why particular patterns occur. In this part of the exercise we are going to probe some social factors that may influence attitudes toward abortion. First, let's consider gender differences in abortion attitudes; again, we can pose a comparative question, "How do the attitudes of men differ from those of women?" We can go further and develop a hypothesis that men will have more negative views on abortion than women. This basis for this hypothesis could be that because in the United States children have carried their father's name and inherit their father's property, and in some subcultural groups where a man's worth is measured by the number of children he fathers, there may be a basis to expect more men (than women) to be willing to have women bear children. In addition, since men do not directly assume any of the physical risks of childbearing, it may also make it easier for them to expect women to carry their pregnancies to term.

In contrast, women are at physical risk by having children; they also have historically assumed a disproportionately greater burden in providing child care. Additionally, reproductive freedom has become an important part of the feminist social consciousness, emerging during the revival of feminism beginning in the late 1960s. It seems logical to assume that there would be more women feminists than male feminists. Thus we may have reason to expect women to be more accepting of the right to have abortions than men. But what do the data show? Table 2 provides the survey results for the year 2000.

When you read a table like the one presented above, you first want to c ompare percentages across the table. How large should the subgroup percentage differences have to be in order for us to claim that a relationship exists? As a general rule, a 10% difference is a must.

Writing Assignment 2:

As you look at the numbers in the preceding table, what conclusions can you draw about the differences in attitudes between men and women? Did you find any support for the hypothesis that men are less inclined to support a woman's right to an abortion?

Actually, these results indicate little difference between men and women on the abortion issue and, therefore, no support for our hypothesis. As you can see, there is less than a 1 percent difference between the two subgroups, so on the basis of percentages alone we can conclude that the differences are not significant.

Part III:

Another way to determine whether differences are significant is to run a statistical test . One test that social scientists use to measure for an association between variables is the chi-square test. The test tells us the probability that two variables are associated with one another. If the probability of obtaining a particular chi-square value is less than .05 , we can say that the differences among the groups we're comparing are statistically significant. If the probability exceeds .05 then we have insignificant differences. In Table 1, the chi-square value was 0.0 and probability was .97, which was far in excess of the cut-off for statistical significance, so once again, we can conclude that the relationship between gender and abortion attitudes is not significant.

Having possibly shattered one myth about abortion, we now want to explore whether there may be other factors that are related to these attitudes. First, let's look at the impact that education may have. Are there significant differences based on the amount of education a person has acquired? What would you hypothesize about the relationship between educational achievement and attitudes toward abortion?

Writing Assignment 3:

As you explore the table, how would you summarize the relationship between educational achievement and abortion attitudes? What happens to the percentage of people who responded in the affirmative as the level of education increases? Consider the chi-square statistic, which measures the strength of the relationship between these two variables.

Unlike the previous test of gender differences, here we see that there is a statistically significant relationship between attitudes and education; you will notice that the probability of obtaining a chi-square value of 31.08 is 0.000, which is statistically significant. Thus, we can conclude that education is related to abortion attitudes, with the more highly educated more inclined to accept abortion then the less well educated. The most pronounced shift appears to be between those with a high school education or less and those with some college and more. Before moving on to the last set of analyses, take a minute and consider why the more highly educated members of our society are more accepting of abortion than the less well educated.

Finally, let's look only at the responses of the women in our sample to see whether or not employment status affects their attitudes about abortion. What do you think? Before analyzing the data, develop a hypothesis that would express a possible relationship between these variables.

Writing Assignment 4:

Now, looking at the results that are presented in Table 4, what conclusions can you draw about the relationship between employment and abortion attitudes for this sample of women? As you look at the numbers more closely, can you come up with a possible explanation for why housewives and retired women showed significantly less support for abortion than women who were in the workforce?

As you can see, the probability for the chi-square test was .04, bringing this within the range of statistical significance. Therefore, we can conclude that among the women in this sample, employment status is related to attitudes toward abortion. 

Part IV:

Writing Assignment 5/Essay:

The final part of this exercise is to write a brief statement of what you learned about attitudes toward abortion among the U.S. population. When researchers publish their findings in research journals, they will generally include an "abstract" or brief statement about the research that appears at the beginning of the article. This short summary includes (1) some description of the sample of respondents; (2) a statement about what the research questions or research hypotheses are; (3) a statement about the methods used; and (4) a short summary of the findings. Prepare an abstract of this research, including all four of these points.