This section has a dual purpose. One purpose is to review the basics of computing and interpreting a correlation coefficient using SPSS. The second purpose is to explain the basics of entering data into the SPSS program.

A Review of Correlation

Remember that a correlation coefficient provides a measure of the degree of linear relationship between two variables. Generally, correlations are computed between two different variables that have each been measured on the same group of people. Each person in the sample provides a score on each of the two variables. For example, a researcher might be interested in the relationship between current college GPA and the number of hours the student studies in an average week in the middle of the school year. Hopefully, the result would be a positive correlation with higher GPAs associated with more hours of study time. A data from designed to summarize data for such a study might look like this:

Notice that there are eleven participants, each having a score on both the GPA variable and the Weekly Study Time variable. Participants in such studies are normally given a number as shown here in order to protect their confidentiality. SPSS also uses the spreadsheet format shown here. Each row of the spreadsheet is called a CASE which is almost always one of the participants in the study. Each column of the spreadsheet is used to store a particular bit of information about the participant, such as GPA and Weekly Study Hours, as shown here, or any other information relevant to the study. Thus, each column has a different variable with a value for each person or case in the study. Complex studies may have thousands of participants and hundreds of variables.