Molly McPup

Introduction to Research Methods in Political Science:
The POWERMUTT* Project
(for use with SPSS)

*Politically-Oriented Web-Enhanced Research Methods for Undergraduates — Topics & Tools
Resources for introductory research methods courses in political science and related disciplines



From the menu bar, click on “Analyze,” then on “Compare Means,” and on “Means….”  In the left window, select the continuous variable (usually the dependent variable) you wish to analyze, and click on the top right arrow.  Then select the categorical variable(s) (usually the independent variable(s)) you wish to use as independent variables, and click on the second right arrow.  If you wish to analyze two or more categorical variables simultaneously (e.g., “Northeast Republicans,” “Northeast Democrats,” etc.), click on “Next” between the two right windows, select your second layer of categorical variables, and click on the second right arrow.  Do the same for any additional layers of categorical variables.  Note: COMPARE MEANS will produce descriptive statistics for each combination of variables specified, but will only carry out analysis of variance and a test for linearity for first layer variables.

Click on “Options….”  By default, COMPARE MEANS provides means, number of cases, and standard deviations, but a number of additional statistics are available.  Add or remove statistics by selecting the statistic(s) you wish to add or remove and clicking on the arrow in the center of the box.  Check “Anova table and eta” and/or “Test for linearity” as desired.  Click on “Continue” and on “OK.”

If you want to “Display clustered bar charts,” and/or “Suppress tables,” check the appropriate boxes before clicking on “OK.”


Last updated April 28, 2013 .
© 2003---2013  John L. Korey.  Licensed under a  Description: Creative Commons License Creative Commons Attribution-Noncommercial-Share Alike 3.0 Unported License.