More Pretest-Posttest Design

Overview

The three approaches for Pretest-Posttest analysis described in Pretest-Posttest Design can be used with more than two between-subject groups, except that Gain can be analyzed using one-way ANOVA instead of a t-test. We demonstrate this using the data in Figure 1.

Three between-subjects groups

Figure 1 – Three between-subjects groups

Using ANOVA

Figure 2 shows the ANOVA analysis of the Gain (using the data in range F3:G18 in standard format).

Gain analysis using ANOVA

Figure 2 – ANOVA on Gain

Using Repeated Measures Analysis

Figure 3 displays the repeated measures analysis.

Repeated measures ANOVA

Figure 3 – Repeated Measures ANOVA

Using ANCOVA

Figure 4 shows the ANCOVA analysis.

ANCOVA

Figure 4 – ANCOVA approach

We also perform Tukey’s HSD to determine which comparisons are significant, as shown in Figure 5. We see that Treat 1 vs. Control and Treat 1 vs. Treat 2 are significantly different, while Treat 2 vs. Control is not significant.

Tukey's HSD post-hoc test

Figure 5 – Tukey’s HSD

Examples Workbook

Click here to download the Excel workbook with the examples described on this webpage.

Reference

Dimitrov, D. M. and Rumrill Jr., P. D. (2003) Pretest-posttest designs and measurement of change
https://www.academia.edu/6629291/Pretest_posttest_designs_and_measurement_of_change

Leave a Comment