 
One-Way
Within Subjects ANOVA
When the participants in an expeirment are involved in all the levels of the independent variable (assuming that there are three or more levels of the independent variable), it is called a "within subjects" or "repated measures" design and calls for a within subjects ANOVA. Using an example looking at the effects of drugs on behavior, the procedures for doing this type of design using SPSS are outlined below. The purpose of the experiment was to study the effects of four drugs upon reaction time to a series of standardized tasks. All participants were given extensive training on these tasks prior to the experiment. The ten participants used in the experiment could be considered a random sample from a population of interest to the experimenter. Each participant was observed under each of the drugs; the order in which a participant was administered a given drug was randomized. A sufficient time was allowed between the administration of the drugs to avoid the effect of one drug upon the effects of subsequent drugs. On the data sheet below note the way data was entered: Each row represents the data from each participant with each of the drug conditions labeled. ID numbers are assigned to each subject just to help keep track of the data.
Select Analyze/General Linear Model/GLM-Repeated Measures to see the following window:
In place of "factor 1," type the name of the within-subjects factor (IV). In this example, 'drug' was used as the name of the within-subjects factor. Now enter the number of levels (4 in this case) followed by clicking the Add button. If you had additional within-subjects factors you would repeat this process for each factor. Once you have completed identifying the within-subjects factors, click the Define button to see the following window
In the left portion of this window you will see a list of the variables contained in the spreadsheet. Highlight each dependent variable that corresponds to each level of the within-subjects factor and click the arrow button to the left of the "Within-Subjects Variables" window. You can change the order of the variables if needed by pressing the up and down arrow buttons. The order the variables appear in this window must correspond to the levels of the within-subjects factor. Select Contrast and highlight each variable in the top window that you would like to do a post hoc analysis (contrasts). By pulling the arrow to the right of the bottom window, scroll until you see Repeated, click Change, and then Continue
Select Plot from the GLM-Repeated Measures window. To obtain a line, highlight drug and click the arrow beside "Horizontal Axis" and then press Add. Press Continue in order to return to the GLM- Repeated Measures window. Select Options from the GLM-General Factorial window and select the following options to obtain descriptive statistics and effect size, then select Continue. The program will provide a measure of homogeneity of variance without requesting it
The completed GLM-Repeated Measures window should look like the following window. If it does, click OK to run the requested analyses.
The output from this procedure is quite extensive. Below is a summary of the key bits of output that are essential to understanding the results. The first window, "Within-Subjects Factors" shows the coding for each categorical variable and the corresponding values. The second window contains the descriptive statistics for each condition.
The homogeneity of variance test is next. SPSS provides use with a measure of Sphericity for within-subjects designs. The is just a more sophisticated test of homogeneity of variance. Look at the Sig value to determine if the assumption of homogeneity of variance was met and interpret it as we have done previously.
    ·
5/00
|
·
.
