Skip to main content

Table 2 Fit statistics of the Latent Growth Curve Model and Growth Mixture Models

From: The impact of physical activity on healthy ageing trajectories: evidence from eight cohort studies

  Linear growth - Number of latent classes
  1 class 2 classes 3 classes 4 classes 5 classes
Loglikelihood
(N of parameters)
−1,661,740 (6) −1,657,724 (9) −1,656,300 (12) −1,655,481 (15) −1,654,823 (18)
BLRT   8032.7*** 2846.4*** 1639.3*** 1314.5***
Adj. LMR-LRT   7811.6*** 2768.1*** 1594.2*** 1278.3***
BIC 3,323,552 3,315,554 3,312,743 3,311,139 3,309,860
SABIC 3,323,532 3,315,526 3,312,705 3,311,091 3,309,803
Entropy   0.644 0.692 0.661 0.690
Group size (%)
 Class 1 100 73.5 4.1 54.1 17.2
 Class 2   26.5 25.4 9.9 2.1
 Class 3    70.5 3.6 43.1
 Class 4     32.4 34.1
 Class 5      3.5
  Quadratic growth - Number of latent classes
  1 class 2 classes 3 classes 4 classes 5 classes
Loglikelihood (N of parameters) −1,660,737 (10) −1,656,703 (14) −1,655,111 (18) −1,654,139 (22)  
BLRT   8067.8*** 3184.6*** 1943.9***  
Adj. LMR-LRT   7900.1*** 3118.4*** 1903.5***  
BIC 3,321,592 3,313,571 3,310,434 3,308,537  
SABIC 3,321,560 3,313,527 3,310,377 3,308,467  
Entropy   0.644 0.722 0.668  
Group size (%)
 Class 1 100 73.5 25.2 34.2  
 Class 2   26.5 3.4 12.2  
 Class 3    71.4 3.3  
 Class 4     50.3  
 Class 5      
  1. Note. *** p < .001. The model with quadratic growth and 5 classes provided an improper solution and therefore is not displayed in the table. BLRT Bootstrap likelihood ratio test; Adj. LMR-LRT Adjusted Lo-Mendell-Rubin likelihood ratio test; BIC Bayesian Information Criterion; SABIC Sample-size-Adjusted Bayesian Information Criterion