Tests of Between-Subjects Effects Dependent Variable: GRE in major Source Type I

Question

Tests of Between-Subjects Effects                      
Dependent Variable: GRE in major
Source   Type III Sum of Squares   df   Mean Square   F   Sig.   Partial Eta Squared
Corrected Model   239308.000a   44   5438.818   4.029   .062   .973
Intercept   19618076.430   1   19618076.430   14531.908   .000   1.000
Gender   .000   0   .   .   .   .000
PrepApproach   35266.667   2   17633.333   13.062   .010   .839
GPA   232031.191   35   6629.463   4.911   .041   .972
Gender * PrepApproach   .000   0   .   .   .   .000
Gender * GPA   .000   0   .   .   .   .000
PrepApproach * GPA   .000   0   .   .   .   .000
Gender * PrepApproach * GPA   .000   0   .   .   .   .000
Error   6750.000   5   1350.000          
Total   21514300.000   50              
Corrected Total   246058.000   49              
a R Squared = .973 (Adjusted R Squared = .731)                      

I want to know why some of the factors are blank without any data under significance value. This also affected the Lavene's test so that the F value and Sig. value were blank. Does this mean there are outliers in the data? Is the rest of the information in the table accurate to report then?

Explanation / Answer

this is regression analysis with Gender, PrepApproach and GPA as indpendent variables

and GRE in major as dependent variable.

here independent variable Gender has zero value and so for corresponding interaction term with gender.

that is why some of the factors are blank without any data under significance value. It may be due to the gender may have only one level.

This doesnot mean that there are outliers in data. yes rest is OK, after removing gender and its interaction with other variables.

Related Questions

Navigate

• Browse (All)
• Browse T
• Subjects

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.