. 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 3.

Study 2: Random paring of predictions and IAT D-scores between two participants (A and B).

Parameters	DV: Participant A’s IAT D-score		DV: Participant B’s IAT D-score
Parameters	Simple model estimates	Sim. reg. model estimates	Simple model estimates	Sim.rReg. model estimates
Fixed effects
Participant A’s IAT score pre-dictions		.50^***	.34^***	.17^***
Participant B’s IAT score pre-dictions	.36^***	.18^***		.48^***
A’s predictions × Condition		−.00	−.08	−.02
B’s predictions × condition	−.01	.07		.02

Random effect variances
Participant A’s IAT score pre-dictions		.057	.080^*	.043
Participant B’s IAT score pre-dictions	.026	.054		.055
Residuals	.683^***	.466^***	.646^***	.481^***

Goodness of fit
−2 log likelihood	1124.55	1010.99	1121.67	1018.27

p<.05

^***

p<.001

All level-1 variables, including the dependent IAT scores, are standardized for each individual participant before they are entered in the analysis. Pairing of participants A and B are entirely random, but fixed within condition. Different random pairings would lead to slightly different results.