Exercise 10: Attitudes on Trade and Immigration and the Presidential Vote

Another example of the joint influence of two independent variables involves the impact of two different issue orientations on the vote: attitudes on trade and on immigration. Two variables that measure attitudes on these issues are: J07, whether or not the U.S. should limit imports to preserve jobs; and K11, whether the current immigration level should be increased or decreased.

Attitudes on Trade and Immigration

Trump’s presidential campaign emphasized two issues that made him stand out, even from other Republican candidates. First, he relentlessly criticized past trade policy as devastating to manufacturing jobs in the U.S., and he promised to revise existing trade deals to prevent further loss of jobs to other countries, such as Mexico or China, and even to bring manufacturing jobs back to the U.S. Second, he stressed the dangers to this country from current immigration policies, focusing on both the threats from illegal immigrants coming from Mexico and from Muslim refugees from the Middle East. Trump repeatedly talked about building a wall along the entire border with Mexico, deporting many of those living here illegally, and banning Muslim refugees from entering the U.S., at least temporarily.

Step A. Create and interpret Tables 10A and 10B

To examine the effects of these variables on the presidential vote create Tables 10A and 10B, two basic two-variable tables, one table for the relationship between trade policy attitudes (J07) and the presidential vote (Table 10A) and one table for the relationship between immigration attitudes (K11) and the presidential vote (Table 10B). For the reasons suggested in exercise one, 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 Tables 10A and 10B as suggested, you should have two tables. Table 10A (vote by trade policy) should have two columns and two rows. Table 10B (vote by immigration policy) should have four columns and two rows. In each case, the attitude (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. In Table 10A, how much difference in their presidential voting is there between those who favor limiting imports and those who do not?
  2. In Table 10B, how much difference in their presidential voting is there between those who want to limit immigration and those who do not? Looking at all four categories of K11, what is the overall relationship between these two variables?
  3. Is the relationship in Table 10A stronger, weaker, or about the same as the relationship in Table 10B?

These tables show that both attitudes are related to the vote. Trump did better among both those who wanted to limit imports and those wanted to reduce immigration. This raises an interesting question: what if a voter agreed with Trump on one of these two issues, but disagreed with him on the other? Did both of these variables equally affect the vote, or was one a more importance influence?

Step B. Create Table 10C

To examine these questions, create Table 10C that uses attitudes on both issues as independent variables and presidential vote as the dependent variable. To simplify the table, recode K11 so that it is dichotomized into those who want to decrease immigration and those who do not. In this case, it does not matter whether J07 is treated as the independent variable and the recoded version of K11 is the control variable, or the reverse, since we are considering both to be independent variables that jointly affect the vote.

K11 has four categories. To simplify the three-variable table that we want to create, it would be helpful to collapse these four categories down to just two. One way to recode K11 would be to simply be dichotomized it by combining the top two categories and the bottom two categories, thus dividing respondents into those who: (a) want to keep immigration levels as they are now or to even increase them; and (b) those who want them decreased.

Step C. Interpret Table 10C

If you ran Table 10C as suggested, you should have a table that consists of two subtables. Each subtable should have two columns and two rows. If you used trade policy as the independent variable and immigration policy as the control variable in generating the table, then the subtables should have trade policy 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 two subtables: one for each of the two categories of the desirable immigration policy. In reading your subtables, take care to properly interpret the percentages, remembering that they are column percentages, not row percentages.

It does not make any difference whether you: (a) treat income as the independent variable and church attendance as the control variable; or (b) you reverse these two variables, as both are really independent variables in this analysis. Either way, you will have nine groups of voters, defined by the possible combinations of being high, medium, or low in income or church attendance.

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

  1. Among those who both favor limiting imports and reducing immigration, what percentage of respondents voted for Trump? Among those who both oppose limiting imports and do not want to reduce immigration, what percentage voted for Trump?
  2. Among those who oppose limiting imports but favor reducing immigration, what percentage of respondents voted for Trump? How about among those favor limiting imports but do not want to reduce immigration?
  3. In examining the joint impact of these two variables on the vote, does the combination of the two variables provide a better explanation that either variable alone? Does one variable seem to have a greater effect, or do both have a similar effect?