Table 2 Fit indices for different models with number of profiles ranging from 2 to 7.

From: Multiple air pollutant exposure and lung cancer in Tehran, Iran

 

Log-likelihood

AIC

BIC

aBIC

Entropy

BLRT

2 Profile

-70,802.1

141,678.2

141,878.3

141,760.8

0.93

6922.7***

3 Profile

-69,084.9

138,269.8

138,540.3

138,381.4

0.92

3434.3***

4 Profile

-67,326.8

134,779.7

135,120.5

134,920.4

0.94

3516.1***

5 Profile

-66,087.4

132,326.8

132,738.0

132,496.5

0.96

2478.8***

6 Profile

-65,394.4

130,966.8

131,448.3

131,165.5

0.94

1386.1***

7 Profile

-64,790.5

129,785.1

130,336.9

130,012.8

0.95

1469.5***

  1. Note AIC Akaike’s Information Criterion, BIC Bayesian Information Criterion, aBIC Sample Size Adjusted BIC, and BLRT bootstrap likelihood ratio test. ***P value < 0.001. Models with a low AIC, BIC, and aBIC value and high Entropy value as well as significant Bootstrap Likelihood Ratio Test (BLRT) were preferred.
  2. Bolded Fit Indices indicate that the 3 profile is optimal latent profile model, The selected (3 profile) has been bolded.
Back to article page