I need to transform the score of a variable to z-scores, because of two different scales that were used according to the subjects' age. However, there are two groups: control and diseased. Do I compare the variable score to the mean and standard deviation of the control group i.e. z-score = (score – control mean)/SD in control group for each scale?
Solved – z-score of one variable, but for two groups
spssstandardization
Related Solutions
For the most part it is similar to a standard pre-post treatment-control design. As such most of the responses to this question would be relevant. The main difference is that you have three age groups.
I don't quite understand how you can have unit scores taken at different intervals and still have a pure pre-post design.
Nonetheless, one way of implementing your analysis is as a 3 x 2 x 2 mixed ANOVA with 3 levels of age, 2 levels of condition, and 2 levels of time.
This can readily be tested using a GLM - repeated measures
in SPSS with time as your repeated measures factors and age and condition as your between subjects variables.
Furthermore, it sounds like you have two different dependent variables. One for Subject A and one for subject B. If you are happy you could just run two separate ANOVAs, one for each DV.
Regarding your specific questions:
- Presumably you are actually interested in the pre-post change scores, which would be represented by the condition by time interaction.
- The main effect of age and possibly with some post hoc tests would test this.
- I don't quite understand how your unit scores work and how these relate to pre-post measures. Are they different types of tests? Or are they just a way of measuring change over time?
- You could just add gender as a between subjects factor and look at the effect of gender.
More generally, you might find it useful to read through a book like Statistical Methods for Psychology. This book provides extensive discussion of how to analyse complex psychological experiments including mixed ANOVAs and follow-up tests.
I should also mention that these are just some quick thoughts. Your data is of moderate complexity and basically you are asking about how to design an analysis plan for an entire data analysis project. This is a complex undertaking requiring many reasoned decisions. This site generally works best when you ask about more discrete questions.
Best Answer
It depends on what your purpose is. If you want to compare individuals with different scores within the same treatment group, you would use the mean and standard deviation of the treatment group they're in. If you want to compare the average score between treatment groups, you would use the overall mean and standard deviation. If you standardized within each group, the mean difference would be exactly equal to 0 (i.e., because the means of the Z-score within each group would be equal to 0). To retain the differences between the treatment groups, the Z-scores should be computed with respect to the whole sample.