Fall, 2001 (Roos, Soc. 311)

Assignment 3: Elaboration Exercise (due Friday, October 5th)

Because we will talk about this assignment in class on October 5th, late assignments will not be accepted.

Continuing on our political theme, presented below are data from the 1998 General Social Survey describing a sample of men and women in the U.S. who voted for either Clinton or Dole in the 1996 election. To keep it simple, those who didn't vote, or who voted for other candidates (e.g., Perot) were deleted from this analysis. The variables are (1) sex, (2) whether or not the respondent was currently working, and (3) whether the respondent voted for Clinton or Dole in 1996 (yes=voted for Clinton, no=voted for Dole)..

Your assignment is to use the raw data presented below to test fully the following hypothesis:

Hypothesis: Women were more likely than men to vote for Bill Clinton in 1996

To test this hypothesis using the elaboration model, follow these steps (show all your work and don't forget to title and label each table):

1. Which of the variables is the independent variable? the dependent variable? the test variable?

2. Is the test variable antecedent or intervening?

3. Make a bivariate percentage table showing the "original relationship" (round your percentages to the nearest 1/10th of a decimal place).

4. Make the two other bivariate tables (rounding as above).

5. Make the trivariate percentage table illustrating the elaboration of the original relationship (rounding as above).

6. Summarize in several paragraphs the results of your analyses, indicating how they reflect on the original hypothesis. Focus in your discussion on the original and trivariate tables (it's best to consult the "other two" bivariate tables to understand what is going on, but there is no need to interpret them). Which elaboration model best describes your findings? Explain your answer.

ASSIGNMENTS MUST BE TYPED