**” Evaluating the effectiveness of interactive graphs in improving students’ performance compared to static graphs”**

The hypothesis for the gain score is:

Null Hypothesis: There is no difference between gain scores from pre-test to post-test between participants who used instructional materials with interactive graphs and participants who used instructional materials with static graphs.

Mt=Mc

**Alternative hypothesis:**

There is a statistically significant difference in gain score from pre-test to post-test between participants who used instructional materials with interactive graphs and participants who used instructional materials with static graphs.

Mt not equal to Mc

where mt=mean gain scores in Group E and Mc= mean gain scores in Group C

The hypothesis for the ANCOVA is:

**Null Hypothesis:**

There is no difference in the post-test scores between participants who used instructional materials with interactive graphs and participants who used instructional materials with static graphs.

**Alternative hypothesis:**

There is a statistically significant difference in the post-test scores between participants who used instructional materials with interactive graphs and participants who used instructional materials with static graphs.

**Evaluating the effectiveness of interactive graphs in improving students’ performance compared to static graphs**

**This is the plan:**

1- Plot pretest scores and posttest scores for each group to assess whether the data is normally distributed and whether there are outliers and deal with violation of normality and outliers, if exist.

2- Calculate the reliability of the pretest scores and posttest scores in each group using (KR-20) measure.

3- Run f-test on pretest scores of the two groups to assess the homogeneity of variances and deal with the violation if exist.

4- Run f-test on pretest scores and posttest scores for the experimental group and another f-test on pretest scores and posttest scores for the control group to check similarity of variances between pretest scores and posttest scores in each group.

5- If the pretest scores and the posttest scores in each group have different reliability and variances, then run f-test on pretest scores of the two groups to assess the homogeneity of variances and deal with the violation if exist.

6- If variance assumption between the pretest scores of the two groups is confirmed then, plot posttest scores against pretest scores for each group to determine if the relationship between posttest scores and pretest scores within each group is approximately linear.

7- If the linearity assumption is confirmed then conduct t-test to determine if the correlation coefficient is significantly different from zero, and, hence that there is evidence of a correlation between the pretest and posttest within the groups.

8- If the correlation coefficient between pretest and posttest in both groups is high (approximately 1.00), then measure the improvement (gain) from pretest to posttest for each participant by subtracting each participant’s pretest score from his or her posttest score.

9- Calculate means gain scores for the experimental and control groups.

10- Run single factor ANOVA on the mean gain scores to compare the improvement of the two groups.

11- If there is a statistically significant difference between the mean gain scores of the two groups, then calculate the effect size and the standard error of the difference between the gain scores means.

12- If the pretest scores and posttest scores have similar reliability and variances and the correlation coefficient is low then run ANCOVA test between the groups instead of the gain score test.

13- Calculate the effect size, if exist.

Last Updated on July 23, 2020 by Essay Pro