Skip to main content

Table 4 Model fit characteristics for multi-trajectory model analysis in girls of the MATCH study

From: One size does not fit all: identifying clusters of physical activity, screen time, and sleep behaviour co-development from childhood to adolescence

Trajectory group n Estimated (%, 95 CI) Assigned (%) APP (%) OCC BIC (n = 15,694) BIC (n = 491)
2-group model       − 7491.93 − 7459.01
 1 148 29.4 (25.4–33.4) 29.0 93.6 14.6   
 2 362 70.6 (66.6–74.6) 71.0 97.0 32.3   
3-group model       − 7225.88 − 7175.65
 1 314 60.1 (55.8–64.4) 61.6 94.6 35.6   
 2 114 22.4 (18.8–26.0) 22.5 91.7 22.4   
 3 82 16.1 (12.9–19.3) 16.0 92.9 26.6   
4-group model       − 7131.09 − 7063.53
 1 66 14.6 (11.5–17.7) 12.9 87.7 21.4   
 2 293 54.8 (50.5–59.1) 57.5 91.5 32.3   
 3 108 21.9 (18.3–25.5) 21.2 89.7 26.1   
 4 43 8.7 (6.3–11.1) 8.4 93.0 39.3   
5-group model       − 7083.56 − 6998.68
 1 203 37.2 (33.0–41.4) 39.8 86.4 25.4   
 2 148 30.2 (26.2–34.2) 29.0 85.0 22.7   
 3 47 9.5 (7.0–12.0) 9.2 89.5 34.1   
 4 68 14.3 (11.3–17.3) 13.3 86.5 25.6   
 5 44 8.7 (6.3–11.1) 8.6 91.5 43.1   
6-group model       − 7081.03 − 6978.82
 1 209 36.1 (31.9–40.3) 41.0 82.4 23.4   
 2 44 10.7 (8.0–13.4) 8.6 73.2 13.6   
 3 110 22.4 (18.8–26.0) 21.6 82.2 23.0   
 4 71 15.1 (12.0–18.2) 13.9 87.1 33.7   
 5 32 7.0 (4.8–9.2) 6.3 90.9 49.8   
 6 44 8.7 (6.3–11.1) 8.6 91.5 53.7   
Final modela       − 7154.16 − 7105.65
Non-compliers 228 42.5 (38.2–46.8) 44.7 88.8 23.8   
Decliners 85 18.5 (15.1–21.9) 16.7 86.3 18.9   
Compliers 44 9.0 (6.5–11.5) 8.6 94.0 47.0   
Screen-Compliers 153 30.0 (26.0–34.0) 30.0 89.0 24.3   
  1. Abbreviations: APP Average posterior probability of classification, OCC Odds of correct classification, BIC Bayesian information criterion
  2. aVariations between the distribution of participants in trajectory group results and final model results to facilitate trajectory group presentation