Calibratio curve

#1
Hello, i am looking for help, for the following situation. I want to do a calibration curve in SPSS. In the study, the following variables are dependent - accuracy, confidence, overconfidence and slope. Accuracy at the trial level was a binary variable (0 - incorrect response/ 1- correct response). Confidence was measure by two choices (50% and 100), in SPSS confidence was introduced as 0.5 and 1.
Now, i need to calculate the proportion of correct responses. How can I do that? Also I want to mention that total score for accuracy was computed for all subjects (N= 119). For confidence was computed average score. How can i do calibration curve which will show how much overconfident or under confident respondents were? Thank you.
 

hlsmith

Less is more. Stay pure. Stay poor.
#2
I believe calibration plot are primarily used with binary dependent values and consists of plot the expected (predicted probability) against the observed (actual in sample probability).. I was unable to follow your description, but if you are using logistic regression it should not be that difficult. Start by outputting your y-hat probabilities or scoring your data.