The first hypothesis is the here will be significant differences between the groups on whether J--- would be found guilty. We tested this with the following:
probit guilty i.condition
Iteration 0: log likelihood = -260.14754
Iteration 1: log likelihood = -244.64032
Iteration 2: log likelihood = -243.53838
Iteration 3: log likelihood = -243.51145
Iteration 4: log likelihood = -243.51145
Probit regression Number of obs = 695
LR chi2(6) = 33.27
Prob > chi2 = 0.0000
Log likelihood = -243.51145 Pseudo R2 = 0.0639
------------------------------------------------------------------------------
guilty1 | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
condition |
1 | -1.418403 .4001703 -3.54 0.000 -2.202723 -.634084
2 | -.7840627 .422603 -1.86 0.064 -1.612349 .0442241
3 | -1.13126 .4085279 -2.77 0.006 -1.93196 -.3305604
4 | -1.279296 .4043334 -3.16 0.002 -2.071775 -.4868172
5 | -1.290944 .4025409 -3.21 0.001 -2.07991 -.5019787
7 | -1.520112 .3981991 -3.82 0.000 -2.300568 -.7396558
|
_cons | 2.333769 .3724239 6.27 0.000 1.603831 3.063706
------------------------------------------------------------------------------
Afterwards, we tested whether people would recommend the 'harsher' punishment of prison over probation:
probit prison i.condition
Iteration 0: log likelihood = -458.94825
Iteration 1: log likelihood = -427.35176
Iteration 2: log likelihood = -427.32345
Iteration 3: log likelihood = -427.32345
Probit regression Number of obs = 695
LR chi2(6) = 63.25
Prob > chi2 = 0.0000
Log likelihood = -427.32345 Pseudo R2 = 0.0689
------------------------------------------------------------------------------
prison | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
condition |
1 | -.9045156 .1835222 -4.93 0.000 -1.264212 -.5448187
2 | -.2369455 .179025 -1.32 0.186 -.587828 .113937
3 | -.9614254 .1867252 -5.15 0.000 -1.3274 -.5954509
4 | -.6973651 .1822121 -3.83 0.000 -1.054494 -.3402359
5 | -1.03343 .1859149 -5.56 0.000 -1.397817 -.6690439
7 | -1.03343 .1859149 -5.56 0.000 -1.397817 -.6690439
|
_cons | .3511308 .126917 2.77 0.006 .102378 .5998837
------------------------------------------------------------------------------
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.
CATEGORICAL = guilty prison sc1 ts1 fault blame;
USEVARIABLES ARE
guilty prison fault blame sc1 ts1 tbinospecific genesfor rs16838844 raregene wombtbi tbi;
missing are all (9999);
Analysis:
estimator = WLSMV;
ITERATIONS= 10000;
Model:
guilty ON tbinospecific genesfor rs16838844 raregene wombtbi tbi;
prison ON tbinospecific genesfor rs16838844 raregene wombtbi tbi;
fault ON tbinospecific genesfor rs16838844 raregene wombtbi tbi;
blame ON tbinospecific genesfor rs16838844 raregene wombtbi tbi;
sc1 ON tbinospecific genesfor rs16838844 raregene wombtbi tbi;
ts1 ON tbinospecific genesfor rs16838844 raregene wombtbi tbi;
guilty WITH prison;
fault WITH blame sc1 ts1;
blame WITH sc1 ts1 ;
sc1 WITH ts1;
Output:
Samp stdYX !Mod(All 4)
Residual Cinterval Tech4;
STANDARDIZED MODEL RESULTS
STDYX Standardization
Two-Tailed
Estimate S.E. Est./S.E. P-Value
GUILTY ON
TBINOSPECI -0.448 0.107 -4.199 0.000
GENESFOR -0.246 0.124 -1.985 0.047
RS16838844 -0.351 0.112 -3.126 0.002
RAREGENE -0.397 0.108 -3.658 0.000
WOMBTBI -0.409 0.110 -3.721 0.000
TBI -0.482 0.105 -4.600 0.000
PRISON ON
TBINOSPECI -0.296 0.058 -5.129 0.000
GENESFOR -0.077 0.058 -1.327 0.184
RS16838844 -0.309 0.057 -5.383 0.000
RAREGENE -0.224 0.057 -3.916 0.000
WOMBTBI -0.340 0.058 -5.866 0.000
TBI -0.340 0.058 -5.866 0.000
FAULT ON
TBINOSPECI -0.443 0.054 -8.195 0.000
GENESFOR -0.206 0.059 -3.521 0.000
RS16838844 -0.412 0.053 -7.735 0.000
RAREGENE -0.454 0.051 -8.834 0.000
WOMBTBI -0.414 0.054 -7.702 0.000
TBI -0.408 0.054 -7.538 0.000
BLAME ON
TBINOSPECI -0.304 0.050 -6.049 0.000
GENESFOR -0.037 0.050 -0.753 0.451
RS16838844 -0.266 0.049 -5.461 0.000
RAREGENE -0.230 0.050 -4.621 0.000
WOMBTBI -0.220 0.048 -4.556 0.000
TBI -0.264 0.050 -5.254 0.000
SC1 ON
TBINOSPECI 0.063 0.057 1.122 0.262
GENESFOR -0.083 0.059 -1.408 0.159
RS16838844 0.041 0.056 0.745 0.456
RAREGENE 0.061 0.059 1.037 0.300
WOMBTBI -0.035 0.057 -0.605 0.545
TBI -0.002 0.057 -0.026 0.979
TS1 ON
TBINOSPECI 0.357 0.043 8.323 0.000
GENESFOR 0.017 0.047 0.361 0.718
RS16838844 0.244 0.042 5.751 0.000
RAREGENE 0.267 0.045 5.917 0.000
WOMBTBI 0.294 0.045 6.565 0.000
TBI 0.340 0.046 7.347 0.000
