three-way mixed factor ANOVA

*** Apologies for cross-posting this message. I just thought that people who don't use SPSS may still be able to help.

This is a question about within-subject factors in a three-way mixed factor ANOVA. I am terribly confused and I need some help.

I am trying to run a three-way mixed factor ANOVA with group (patients, controls) as a between-subjects factor and two within-subjects factors: familiarity and syntactic complexity. I have organised the data on SPSS as follows: one column for scores on familiar texts with easy syntax, another for familiar text with hard syntax, another for unfamiliar text with easy syntax and another with unfamiliar text with hard syntax. I thought that the familiarity factor has two levels (familiar, unfamiliar) and syntactic complexity has two levels two (easy, hard). The trouble is that the factor of familiarity creeps in the syntactic complexity variable and also the other way around. If you take, for example, the score of one participant on the familiar text with easy syntax you have both variables interacting. In other words, every measure has a level of familiarity (familiar or unfamiliar) and every measure has a level of syntactic complexity (easy or hard). It is, therefore, impossible to define two factors with two levels each. Would defining a broader factor named 'condition' with four levels make more sense? I could run simple effects and get a better picture in relation to each subcondition. Would that make sense?

I am really stuck at the moment, because I feel that if I run these as one condition with four levels (familiar text with easy syntax, familiar text with hard syntax, unfamiliar text with easy syntax, unfamiliar text with hard syntax), I won’t be able to see clearly the effect of familiarity and syntactic complexity and if I run it as two separate variables with two levels each all columns need to be inputted in the ‘Repeated measures define factors’ window on SPSS.

Could you please help?
Last edited:
I would define two factors in SPSS: familiarity(2) and complexity(2). The output will provide you the main effects of each, then you will also see the interaction effect of familiarity x complexity. So you will be able to view the impact of each main effect as well as their interaction.