Optional Activity 4 – Exploring Group Differences with T-tests and ANOVA

[AdrianPalmer]

1. Did students improve from pre- to post-test?
There is sufficient evidence to conclude that there is a statistically significant different in the means of the pre-test and post-test, where the mean difference is -7.2 (95% CI: -8.36, -6.04) with a p-value of <0.001. This would indicate the students scores improve from pre-test to post-test since the mean of the post-test scores (67.90 ± 4.95) is higher than the pre-test scores (60.70 ± 3.59).

2. Are gender differences in test performance statistically significant?
The independent samples t-test did not detect a statistically significant difference in post-test scores between genders, suggesting that gender does not influence the outcome. Though the mean difference was 5 scores, the 95% CI (-1.49, 11.49) with a p value of 0.802, indicates the possibility of no difference.

3. Which study method led to the highest performance?
Students who used flashcards had the highest post-test score mean of 71.00 ± 1.83, but there was no statistically significant difference with those who did group study which had a mean of 70.00 ± 4.36.

4. What does the Tukey post hoc test tell us about specific group differences?
The One-Way ANOVA test did detect a statistically significant difference in mean post-test scores amongst study methods (F=11.18, p = 0.007), which indicates that study methods did affect post-test scores. The Tukey’s test indicated that students who did group study and use flashcards significantly outperformed those who did reading only, and there was no statistically significant difference between flashcards and group study groups.

5. How can these results be applied in designing student support strategies?
In designing support strategies, it would be good to inform students that for better exam results it is not advised to read only, but they should also incorporate the use of flashcards or participate in group study as a part of their preparation

[Author]

Posts