Skip to main content

Table 2 (Aim 1 Results).

From: Patterns of perceived barriers to medical care in older adults: a latent class analysis

  Number of Classes
Fit statistic 1 2 3 4 5
Log-likelihood -17571 -15773 -15450 -15316 -15292
# of estimated parameters 9 19 29 39 49
BIC 35220 31710 31149 30968 31005
P-value (k-1 vs. k) N/A P < .001 P < .001 P < .001 P = 0.058 (NS)
Entropy N/A 0.79 0.82 0.84 0.84
# of bivariate residuals z > 1.96 36 28 15 7 5
  1. Fit Statistics for Latent Class Model of Perceived Barriers in Access to Medical Care Indicating Optimal 4-Class Solution.
  2. For Bayesian information criterion (BIC), a smaller value suggests a better model fit. The likelihood ratio test tests significance in the -2 times Log-likelihood difference between the model with k and k - 1 (H0) classes. Entropy values range from zero to one, with values closer to one indicating clearer delineation of latent classes. Reported p-values are for Lo-Mendell-Rubin adjusted log ratio test.
Back to article page

BMC Health Services Research

ISSN: 1472-6963

Contact us