In contrast to SPSS Statistics 29, older Versions did not allow for a repeated measures ANOVA with a continuous between-subjects factor via the graphical user interface. Since the function has not been implemented for long and does not seem to be used widely, I worry whether there are statistical implications on a design level why one should refrain from utilizing said function. Can anybody help me on that?

I am conducting a study with a differential conditioning paradigm, where people rate 3 different stimuli over the course of two days, with multiple ratings a day. I am interested in a possible relationship of a personality measure (e.g. extraversion) and the change in ratings for said stimuli over time, which is why I would have computed a repeated measures ANOVA with within-subject factors stimuli and time and the continuous between subject factor of extraversion.

As said before, I am a bit unsure whether treating my data and proceeding like that would be statistically wrong, since nobody seems to have approached it that way in the past (people usually seem to perform a median split and divide their participants into pre-defined groups of low and high). Any help on that matter would be greatly appreciated!

Let's say you're conducting a study on the effectiveness of different relaxation techniques (within-subject factors: stimuli and time) on reducing stress levels, and you want to investigate whether the level of extraversion (continuous between-subject factor) moderates the effects of the relaxation techniques on stress reduction. Participants rate their stress levels on a scale of 1 to 10 before and after each relaxation session, which is conducted over two days with multiple sessions per day.

Feasibility:
Treating extraversion as a continuous variable allows you to examine the relationship between extraversion and the change in stress levels over time, while also considering the effects of the relaxation techniques. You can perform a repeated measures ANOVA with within-subject factors (stimuli and time) and the continuous between-subject factor (extraversion) to explore potential interactions and main effects.

Potential Issues:
Assumptions: As with any repeated measures ANOVA, it is important to assess the assumption of sphericity. If the assumption is violated, adjustments such as the Greenhouse-Geisser correction can be used. Additionally, consider other assumptions of ANOVA, such as normality of the data and homogeneity of variances.

Statistical Power: Including a continuous between-subjects factor can reduce statistical power compared to a traditional between-subjects ANOVA. Ensure that your sample size is adequate to detect the effects you expect. Additionally, consider the variability in extraversion scores and its potential impact on statistical power.

Interpretation: Interpreting the results may be more complex when you have a continuous between-subjects factor. In this case, you may observe interactions between extraversion, stimuli, and time. Carefully examine the nature of these interactions and consider conducting post hoc analyses (simple effects analysis, planned contrasts, regression analysis) or plotting the results to understand the specific relationships between extraversion, relaxation techniques, and stress reduction.

While this approach is feasible and allows for a more nuanced examination of the relationship between extraversion and the change in ratings over time, it is crucial to consider the assumptions, statistical power, and interpretation challenges associated with the analysis. Consulting with a statistician or expert in your field can provide valuable guidance and help ensure appropriate analysis and interpretation of your study's findings.

In contrast to SPSS Statistics 29, older Versions did not allow for a repeated measures ANOVA with a continuous between-subjects factor via the graphical user interface. Since the function has not been implemented for long and does not seem to be used widely, I worry whether there are statistical implications on a design level why one should refrain from utilizing said function. Can anybody help me on that?

I am conducting a study with a differential conditioning paradigm, where people rate 3 different stimuli over the course of two days, with multiple ratings a day. I am interested in a possible relationship of a personality measure (e.g. extraversion) and the change in ratings for said stimuli over time, which is why I would have computed a repeated measures ANOVA with within-subject factors stimuli and time and the continuous between subject factor of extraversion.

As said before, I am a bit unsure whether treating my data and proceeding like that would be statistically wrong, since nobody seems to have approached it that way in the past (people usually seem to perform a median split and divide their participants into pre-defined groups of low and high). Any help on that matter would be greatly appreciated!