Skip to main content

Table 2 Criteria to assess model fit for latent class analysis models

From: Identifying heterogeneous health profiles of primary care utilizers and their differential healthcare utilization and mortality – a retrospective cohort study

Number of Classes (k) Class sizes Akaike (AIC) Bayesian (BIC) Sample-Size Adjusted BIC
2 Class 1 = 65,012 (64.5%)
Class 2 = 35,735 (35.5%)
672,123 672,551 672,409
3 Class 1 = 64,813 (64.3%)
Class 2 = 32,519 (32.3%)
Class 3 = 3415 (3.4%)
662,674 663,321 663,105
4 Class 1 = 64,091 (63.6%)
Class 2 = 32,389 (32.2%)
Class 3 = 3013 (3.0%)
Class 4 = 1254 (1.2%)
658,999 659,866 659,576
5 Class 1 = 61,792 (61.3%)
Class 2 = 24,225 (24.1%)
Class 3 = 2510 (2.5%)
Class 4 = 1209 (1.2%)
Class 5 = 11,011 (10.9%)
656,030 657,115 656,753
6 Class 1 = 58,213 (57.8%)
Class 2 = 26,309 (26.1%)
Class 3 = 2964 (2.9%)
Class 4 = 1104 (1.1%)
Class 5 = 11,122 (11.0%)
Class 6 = 1035 (1.0%)
653,089 654,394 653,958