ANOVA or ANCOVA???

[statanx]

In a QUASI-EXPERIMENTAL PRE-POST TEST STUDY (factorial design 3x2, instructional method x gender), 6 intact classes were taken & 2 instructors were involved. Out of 6 classes, 2 were randomly assigned to instructional method X, other 2 to method Y & the remaining 2 to method Z. Each instructor was randomly assigned to teach 3 classes by following a different method so as to control teacher differences.

At pretest stage, it was found that all the 6 sections (comprising 3 different groups) were not significantly different from each other as far as their scores on relevant variables (covariates) was concerned.

Please suggest: which one is more appropriate to use: ANOVA or ANCOVA?
(I noticed in a no. of studies that researchers have used ANCOVA preferably in such a case.)

[victorxstc]

Depending on the types of your variables, you might prefer one approach. ANCOVA tries to find the relationship of two continuous variables while adjusting for a categorical variable. In your case, your variables are not continuous. Your response variable seems to be ordinal (but I don't see you have clarified that). If the response variable is continuous, but your independent variables are categorical, ANOVA should be used. Here you have two independent variables: gender and instructional method. Both seem categorical. So I think ANOVA is better.

Besides, you should check whether ANOVA requirements are met or not. If not, you should consider a non-parametric alternative.

[statanx]

Depending on the types of your variables, you might prefer one approach. ANCOVA tries to find the relationship of two continuous variables while adjusting for a categorical variable. In your case, your variables are not continuous. Your response variable seems to be ordinal (but I don't see you have clarified that). If the response variable is continuous, but your independent variables are categorical, ANOVA should be used. Here you have two independent variables: gender and instructional method. Both seem categorical. So I think ANOVA is better.

Besides, you should check whether ANOVA requirements are met or not. If not, you should consider a non-parametric alternative.

Thanks for your reply.
Dependent variables are three: achievement, attitude towards the teaching subject, attitude towards the instructional method.

Yeah, data satisfy all the assumptions of ANOVA & even ANCOVA as well.

[statanx]

What about ANOVA with repeated measures????????

[victorxstc]

I thought you have determined the type of ANOVA already.

Repeated-measures ANOVA is fine since you have pre- and post-treatment design. However, if you want to exactly determine the type of the analysis, you need to perform a repeated-measures two-way ANOVA.

[statanx]

I thought you have determined the type of ANOVA already.

Repeated-measures ANOVA is fine since you have pre- and post-treatment design. However, if you want to exactly determine the type of the analysis, you need to perform a repeated-measures two-way ANOVA.

Thanks a lot for your reply.

Yeah, I determined the type of ANOVA.
I need to go for 2-way mixed design ANOVA with repeated measures as there are 2 between-subjects factors (Group and Gender) and 1 within-subjects factor (Time: Pre & Post).


But still I have some problems. I don't know what all assumptions of 2-way mixed design ANOVA with repeated measures I need to test & and how to test them in SPSS & interpret them.

I know step-wise procedure for ANOVA with repeated measures but not very sure about interpretation & reporting its results in APA style.

Any suitable link please ............

One more thing... sample size is 210 (70 in each group).