Hello,
I need help with choosing a statistical method. I have two groups, each was tested by using a test that is compiled of three subtests. First, I need to estimate the effect of subtest1 on the total test score. Second, I need to compare if that effect is different between groups.
I would really appreciate your help.
HI
Is it possible to explain more clearly the following sentence that you wrote?
؛؛I need to estimate the effect of subtest1 on the total test score.
Test score is a continuous variable to be compared between two groups. In this case t-test is appropriate. This maybe independent and paired t-test depending on your groups. https://www.sagepub.com/sites/default/files/upm-binaries/43352_13.pdf
Use t test between groups, while as there are three sunsets within a group, go for ANOVA. Above answers will help you further.
Thank you for your answers! It seems that I didn't explain well, so I will try again.
The aim is to estimate the effect of subtest1 on the total test score in comparing to other subtests. When I estimate the effect of subtest1 on the total score (and hypothetically I find that it has a bigger effect than two other subtests), then I need to estimate if the level of that effect is different in the control group. All subtests are numerical variables.
It was previously shown that in the general population subtest1 has the biggest effect on total score, the other two subtests have the lower effect. I have to examine is it different in the clinical population. The assumption is that in clinical population it will also have the biggest effect, but I want to examine if there is difference between group in the size of that effect.
I hope that I managed to explain it now.
Hi! Let me see if I got it. To calculate the effect you may use the Cohen d effect size of subtest 1 for both groups. Then, you can check if the effect sizes differ between the 2 groups using X2.
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Techniques
- Structural Equation Modeling