Table 4: Multivariate growth curve modeling examining estimated effect on students’ baseline physical activity and physical activity trajectories when adjusted for student and school neighbourhood characteristics.

Model 1Model 2Model 3
Student characteristicsSchool neighbourhood characteristics Student and school neighbourhood characteristics
Fixed effect valueFixed effect valueFixed effect value

Estimated effect on baseline physical activity (minutes per day)

Gender
 Female−3.81 (1.27)0.0029−4.20 (1.28)0.0011
 MaleReferentReferent

Baseline physical activity
 Low−46.5 (1.54)<0.0001−46.0 (1.59)<0.0001
 Moderate−25.1 (1.51)<0.0001−25.0 (1.53)<0.0001
 HighReferentReferent

BMI status
 Overweight or obese2.32 (1.81)0.19992.56 (1.80)0.1562
 Healthy weightReferentReferent
 Missing1.48 (1.38)0.28151.41 (1.37)0.3036

School location
 Rural−8.25 (2.49)0.00100.680 (1.48)0.6465
 UrbanReferentReferent

Socioeconomic status
 Low−5.46 (2.50)0.0295−3.47 (1.50)0.0216
 HighReferentReferent

Estimated effect on physical activity trajectories (minutes per day per semester)

Gender
 Female−0.885 (0.608)0.1463−0.735 (0.612)0.2304
 MaleReferentReferent

Baseline physical activity
 Low7.34 (0.741)<0.00017.08 (0.759)<0.0001
 Moderate4.24 (0.735)<0.00014.19 (0.749)<0.0001
 HighReferentReferent

BMI status
 Overweight or obese−1.03 (0.866)0.2343−1.19 (0.866)0.1711
 Healthy weightReferentReferent
 Missing−0.00292 (0.653)0.99640.0645 (0.652)0.9213

School location
 Rural0.942 (0.716)0.1888−0.0316 (0.705)0.9643
 UrbanReferentReferent

Socioeconomic status
 Low1.35 (0.724)0.06291.46 (0.722)0.0432
 HighReferentReferent

Fixed effects of intercepts for each model are not shown.
Standard errors are shown in parentheses.
Referent categories are identified as “referent.”