. Author manuscript; available in PMC: 2014 Jun 1.

Published in final edited form as: J Exp Psychol Gen. 2013 Dec 2;143(3):1369–1392. doi: 10.1037/a0035028

Table 1.

Study 1: IAT D-scores regressed on IAT score predictions (left) and explicit thermometer ratings (right)

Parameters (DV: IAT D-scores)	Prediction model estimates	Imp.-exp. model estimates
Fixed effects
IAT score predictions	.53^***
Predictions × condition	.09^†
Explicit therm. ratings		.01

Random effect variances
IAT score predictions	.041
Explicit therm. ratings		.077
Residuals	.550^***	.741^***

Goodness of fit
−2 log likelihood	750.43	850.39

^†

p < .11

^***

p<.001

All level-1 variables are standardized for each individual participant before they are entered in the analysis. Accordingly the intercept in this model would be 0 and is not included in the model. Similarly, the main effect of condition on these centered IAT scores is not included either. “Condition” represents a level-2 condition assignment. It is coded “−1” for the “cultural associations” condition, and “1” for the “true attitudes” condition.