Normalizing Data

Hi,

I have a repeated measures design and I am trying to analyze the data. The problem is that there is a different number of stimuli in each condition because the software as designed to be interactive with the subject. Every time a stimulus is presented, I have recorded whether or not the subject makes a response. This is what I call the response likelihood of the subject - total number of subject responses/ total stimulus presentations. However because I have a different number of stimuli in each condition, I need to normalize the data in some way because the variable of interest is calculated based on something that changes depending on the condition. Any insights?

Thank you,
Cam