Exercise 4. Attitude on Abortion and the Presidential Vote

Step A. Create and interpret Table 4A

Another issue that might have influenced how people voted in the 2016 presidential election is the issue of abortion. To examine whether attitudes on this issue affected voting, we can create Table 4A and look at attitudes toward abortion to the presidential vote. Respondents in the 2016 ANES were asked about the conditions under which abortion should be legal. We would hypothesize that being more willing to allow abortions would make one more likely to vote for Clinton. To examine that possibility, we can look at a table that relates attitude toward abortion (K01) to the presidential vote. For the reasons suggested in exercise 1, you should use the recoded version of A02 that you created for that exercise, so that you examine only the major-party vote (i.e., only the Clinton and Trump voters).

If you ran Table 4A as suggested, you should have a table with four columns and two rows. Attitude toward abortion (the independent variable) should be on the top of the table (the column variable), and the two-party presidential vote (the dependent variable) should be on the side of the table (the row variable). Percentages should be calculated by column (i.e., they should sum to 100% for each column). In reading your table, take care to properly interpret the percentages, remembering that they are column percentages, not row percentages.

You should attempt to answer these questions to see if you are able to correctly read the table and interpret the data:

  1. What percentage of the voters who thought that abortion should never be permitted voted for Trump? What percentage of the voters who thought that abortion should always be permitted cast a ballot for Trump? Looking at all four columns of this table, is the relationship between the two variables clear and consistent?

    As discussed previously, it does not matter whether you focus on the vote for Trump or for Clinton. If voters who were anti-abortion were more likely to vote for Trump, then those who were pro-abortion must have been more likely to vote for Clinton.

  2. Overall, how strong of a relationship is there between these two variables? How does the relationship of this issue to voting compare with the relationship that you found in Table 3A?

After examining Table 4A, you should conclude that those who were more pro-abortion had a much greater propensity to vote for Clinton than those who were more anti-abortion. However, as we learned from Exercise 3, this does not necessarily mean that attitudes on this issue really had a significant effect on the vote, meaning that people truly voted on the basis of this issue. The relationship in Table 4A could be the result of the actions of an confounding variable.

Step B. Create a three-variable table

As in Exercise 3, party identification is a possible confounding variable, which should be examined to better understand why our independent and dependent variables are related. To do so, you need to create Table 4B, a three-variable table that shows the relationship between attitude toward abortion, presidential vote, and party identification. To ensure that you have a sufficient N for each column, you should recode K01 so that it has just two categories (favor and oppose) and use the recoded version of party identification that you created for exercise one (Democrats, independents, and Republicans).

Variable K01 has four response categories. Respondents could have said that:

  1. Abortion should never be allowed;
  2. Abortion should be allowed only in limited circumstances, such as rape, incest, or when the life of the mother is in danger;
  3. Abortion should be allowed in more circumstances, but the need for abortion should be established;
  4. Abortion should always be allowed.

The first two categories are best classified as anti-abortion or pro-life positions. Even allowing abortion in very limited circumstances is generally accepted as being opposed to abortion. For example, President George W. Bush favored allowing abortion in very limited circumstances, and he was always considered to be opposed to legalized abortion. The last category is clearly a pro-abortion or pro-choice position. The third category is somewhat ambiguous. Permitting abortion in a wider set of circumstances does not fit with the perspective of those opposed to abortion, but stating that a need must be established does not fit with the perspective of those who think that it should be the decision of the mother and her doctor. If we want to recode this variable into just two categories (pro and anti-abortion), then we must decide where this third category fits. We suggest that it makes more sense to combine the third and fourth categories, which will divide those who would allow abortion either not at all or only in very limited circumstances from those who would allow abortion in a broader set of circumstances.

An alternative would be to recode K01 into three categories, putting the first two categories into a clearly anti-abortion group, the fourth category into a clearly pro-abortion group, and the third category into a middle group. That is a legitimate option, and you could choose to do that. As we have discussed in earlier exercises, there often is more that one reasonable way to recode a variable. While we could keep the third group as a separate category, in order to make the tables for this exercise as simple as possible, we suggest combining the third and fourth categories into one that is more favorable to abortion.

To create a three-variable table, you must specify a row variable, a column variable, and a control variable. You normally would set up the table by putting your independent variable as the column variable (this will put it on the top of each subtable), your dependent variable as the row variable (this will put it on the side of each subtable), and your control variable so that it specifies each subtable (there will be one subtable for each category of the control variable). When you are in the SDA crosstabulation program, enter your variables in the row, column, and control dialog boxes. You also will have to enter your recoding instructions for each variable in the appropriate dialog box.

In this example, your control variable should be party identification. You want to recreate Table 3A for each category of party identification—Democrats, independents, and Republicans. You want to use your recoded version of party identification, which has only three groups, because if you use the full seven-category version of party identification, you will have too many subtables, and the result will be many columns with small Ns and a fairly complex table that will be difficult to interpret.

If the table is set up in the above fashion, you normally would want percentages by columns. You should click this option under table options.

You should be sure to have the weight on and that you have selected the weighted Ns to appear in the table. The unweighted data are not a representative sample of the electorate, so be careful not to mistakenly use them.

Because the data are weighted, which means that individual respondents may count as more or less than one person (e.g., as .75 or 1.35 persons), the number of respondents in each cell (the Ns) probably will not be whole numbers. If you prefer to have Ns that are whole numbers, you can revise the output to do that by using the “revise the display” options that appears to the left of the table that you generated.

If statistics are desired, that option should be checked under table options. For a discussion of the statistics that are commonly used for contingency tables, see the section on data analysis. In these exercises, we have not asked you to generate statistics, but your instructor may suggest doing so.

The SDA crosstabulation program will produce both a table and a chart, but the chart is not necessary, as all of the information that you need will be contained in the table that you generate. You can revise the output to drop the chart if you like by using the “revise the display” option.

Step C. Interpret Table 4B

If you ran Table 4B as suggested, you should have a table that consists of three subtables. Each subtable should have two columns and two rows, with the recoded version of attitude toward abortion (the independent variable) on the top of the table (the column variable), and the two-party presidential vote (the dependent variable) on the side of the table (the row variable). Percentages should be calculated by column (i.e., they should sum to 100% for each column). There should be one subtable for Democrats, one for independents, and one for Republicans. In reading your subtables, take care to properly interpret the percentages, remembering that they are column percentages, not row percentages.

You should attempt to answer these questions to see if you are able to correctly read the table and interpret the data:

  1. What is the relationship between attitude toward abortion and presidential vote among Democrats? How about among independents? Among Republicans? Is there still a clear relationship between this attitude and voting?
  2. How does the relationship between attitude toward abortion and the presidential vote in each subtable compare to the original relationship in Table 4A? Is it stronger, weaker, or about the same? What conclusion do you draw from this?

In interpreting Table 4B, focus on the relationship within each subtable. For example, among Democrats, there is about a 10 percentage point difference in voting for Clinton (or for Trump) between those who approve of abortion (94 percent voted for Clinton) and those who disapprove (84 percent voted for Clinton). Both groups were heavily for Clinton, but the first group was more so, which means that these two variables are related among Democrats. The same is true for Republicans: attitude toward Obamacare and the presidential vote are related in this subtable as well, and the difference between the favorable and unfavorable groups also is about 10 percentage points. Among independents, the relationship is even stronger, with nearly a 20 point difference between the favorable and unfavorable groups.

In interpreting the subtables, pay attention to the N in each column. A small N is cause for caution, as the column percentages could be unrepresentative due to random sampling error. In this case, the Ns in each column are large. In only one case is the N less than 100.

In this example, the relationship between attitude toward abortion and the presidential vote persists even after we control for party identification. This indicates that attitudes on this issue did influence how people voted. The association between the independent and dependent variables is not as strong in Table 4B as it was in Table 4A, so we would conclude that party identification did inflate the bivariate association between the two variables, but the important point is the fact that the two variables still are clearly and substantially associated even when party identification is controlled for. Among Democrats and Republicans, there is about a 10 point difference in the presidential vote between those who basically favor and those who basically oppose legalizing abortion; among independents, the difference is almost 20 points. These are meaningful differences.