





|
Assignment 3 2005-2006: Crosstabulation
Date Assigned: October 18/20
Title of Assignment: Crosstabulation
Due Date: October 25/27
Points: 10
This assignment requires that you examine four
hypotheses by means of crosstabulations. The variables for your crosstabulations
may be selected from any available data file. To complete this assignment you
must:
- Examine
the relevant questionnaire or codebook and choose two dependent variables
whose variation you are interested in explaining.
- Choose
two independent variables from the same data set that you think may help
explain variation on each of the selected dependent variables.
- Write
hypotheses stating the relationships you expect to find between each of your
dependent variables and the two independent variables, yielding four
separate hypotheses. Include these hypotheses on the first page of your
output. You will need to begin each successive assignment with your
hypotheses as well.
- In
order to produce clear, interpretable contingency tables, some of the
categories of the variables may have to be combined or rearranged. So
consider whether any of your variables need to be recoded. For example, the
variables may have numerous codes or response categories, requiring
re-grouping into a few larger categories or the codes for the variables may
not be in a logical order (e.g., running medium/low/high rather than
low/medium/high or high/medium/low instead of low/medium/high).
- Declare
any values that should be treated as "missing data" on the
variables so that they are not used in constructing the crosstabulations.
-
Crosstabulate each dependent variable with the two independent variables,
selecting an appropriate measure of association to summarize each table and
checking to make sure that there are no obvious errors in your recodes and
labels.
-
Print your results.
-
Focusing on column percentages (i.e., compare you column percentages
across the rows to identify the trends in the relationship) and the value of measures of
association, write a short paragraph describing each contingency table and
whether the evidence supports your hypothesis. This will be the bulk of your
write-up so take care to explain exactly what is going on in each
crosstabulation. It is not sufficient to have a perfect output if you do not
communicate what the output tells us.
-
At the end of the output, briefly state which of the two independent
variables best explains variation in each dependent variable and why you think
that this is so.
- Hand in you printout along with your answers.

Assignments
|