### mixed model output not generating interactions between factors

Posted:

**Mon Mar 13, 2017 8:33 am**Hi

I am relatively new to SPSS. I am using version 24. I am analyzing data from an experiment that has used a 2x2x2 mixed factorial design (two factors are nominal and one is ordinal). Two of the factors are within subjects and one factor is between groups. The dependent variable is defined as scale (it is a voltage). On the advice of a statistician I am using a linear mixed model to analysis the data.

All of the factors are being treated as fixed effects, and identified as such in the "fixed effects" window. As well, in the "fixed effects" window I have clicked the "build terms" radio button, and the box between the "factors and covariates:" and "model:" boxes with "factorial" displayed. "Include intercepts" is ticked and the sum of squares is Type III.

I have left the "random effects" window unchanged, i.e., nothing entered into the "model" or "combinations' boxes.

I have left the "estimation" window at the default settings.

When i run the analysis the output displays type III tests of fixed effects, including numerator df, denominator df, F values and significance values for the intercept and each of the three factors. But it will not display any interactions, i.e., it will not display any values for Factor 1*Factor 2, Factor 1*Factor 3, Factor 2*Factor 3, or Factor 1*Factor 2*Factor 3. From what i have read it should have reported these after I left "factorial" displayed in the "fixed effects" window. Even when I change this window to display "All 2 way" or "All 3 way" it doesn't make a difference - still no interactions reported.

How do I get the output to display tests for interactions?

Kind regards, Alex.

I am relatively new to SPSS. I am using version 24. I am analyzing data from an experiment that has used a 2x2x2 mixed factorial design (two factors are nominal and one is ordinal). Two of the factors are within subjects and one factor is between groups. The dependent variable is defined as scale (it is a voltage). On the advice of a statistician I am using a linear mixed model to analysis the data.

All of the factors are being treated as fixed effects, and identified as such in the "fixed effects" window. As well, in the "fixed effects" window I have clicked the "build terms" radio button, and the box between the "factors and covariates:" and "model:" boxes with "factorial" displayed. "Include intercepts" is ticked and the sum of squares is Type III.

I have left the "random effects" window unchanged, i.e., nothing entered into the "model" or "combinations' boxes.

I have left the "estimation" window at the default settings.

When i run the analysis the output displays type III tests of fixed effects, including numerator df, denominator df, F values and significance values for the intercept and each of the three factors. But it will not display any interactions, i.e., it will not display any values for Factor 1*Factor 2, Factor 1*Factor 3, Factor 2*Factor 3, or Factor 1*Factor 2*Factor 3. From what i have read it should have reported these after I left "factorial" displayed in the "fixed effects" window. Even when I change this window to display "All 2 way" or "All 3 way" it doesn't make a difference - still no interactions reported.

How do I get the output to display tests for interactions?

Kind regards, Alex.