Skip to main content
. Author manuscript; available in PMC: 2022 Jul 1.
Published in final edited form as: Psychol Assess. 2021 Apr 1;33(7):637–651. doi: 10.1037/pas0000906

Table 3.

Model Fit of the Two-level 1-PL and 2-PL for the EDS and REDI

Model Number of
Parameters
Log-likelihood AIC BIC Likelihood Ratio Test
Comparison Δχ2 Δdf p value
EDS Discrimination Frequency (EDS-DF)
1-PL 7 −1866.15 3746.31 3790.00
2-PL 15 −1851.09 3732.17 3825.81 1-PL 20.98 8 .007
2-PL(D) 11 −1855.90 3733.79 3802.46 2-PL 9.22 4 .056
2-PL(P) 11 −1852.08 3726.16 3794.83 2-PL 1.36 4 .851
EDS Attribution to Ethnicity/Race (EDS-AER)
1-PL 7 −1171.97 2357.93 2401.63
2-PL 15 −1154.33 2338.66 2432.30 1-PL 28.71 8 < .001
2-PL(D) 11 −1158.49 2338.98 2407.65 2-PL 6.89 4 0.142
2-PL(P) 11 −1161.49 2344.98 2378.70 2-PL 12.00 4 0.017
Racial/Ethnic Discrimination Index (REDI)
1-PL 8 −1685.45 3386.90 3436.81
2-PL 18 −1639.29 3314.58 3426.87 1-PL 38.20 10 < .001
2-PL(D) 13 −1672.27 3370.54 3451.65 2-PL 1467.91 5 < .001
2-PL(P) 13 −1650.83 3327.67 3408.77 2-PL 12.59 5 .028

Note. 2-PL(D) is a variant of the 2-PL model in which the daily-level item loadings are constrained to be equal across all items. 2-PL(P) is a variant of the 2-PL model in which the participant-level item loadings are constrained to be equal across all items. AIC is Akaike Information Criterion. BIC is Bayesian Information Criterion (BIC). Δχ2 is the χ2 difference. Δdf is the difference of degrees of freedom.