The first hypothesis is the here will be significant differences between the groups on whether J--- would be found guilty. We will test this with the following:
probit guilty i.tbi
Afterwards, we will test, whether people would recommend the 'harsher' punishment of prison over probation:
probit prison i.tbi
Finally, we will run all seven conditions together on all four dependent variables, allowing all four to correlate with one another. This will provide the most conservative test of the overall hypotheses. This will be done in MPlus, treating all dependnet variables as categorical variables, and dummy coding each of the groups with the control group being the reference group. All covariances between the dummy-coded variables will be set to zero.