Multilevel multivariable logistic regression of school and local authority characteristics on The Daily Mile registration in England (level 1, N=12,214 primary schools; level -2, N=124 local authorities) using scaled parameters
| Model 1: intercept only | Model 2: +school variables | Model 3: +local authority variables | ||||||
|---|---|---|---|---|---|---|---|---|
| # | Parameters | OR (95% CI) | SE | Scaled OR (95% CI) | SE | Scaled OR (95% CI) | SE | Unscaled OR (95% CI) |
| 1 | Local authority–controlled schoolⱡ | 1.06 (0.96, 1.17) | 0.05 | 1.07 (0.96, 1.19) | 0.05 | |||
| 2 | Hamlets and isolated dwellings (rural)ⱡ | 0.95 (0.75, 1.21) | 0.12 | 0.96 (0.76, 1.22) | 0.12 | |||
| 3 | Town and fringe (rural)§ | 1.01 (0.87, 1.19) | 0.08 | 1.03 (0.87, 1.21) | 0.08 | |||
| 4 | Village (rural)§ | 0.88 (0.75, 1.04) | 0.08 | 0.88 (0.74, 1.04) | 0.09 | |||
| 5 | Major conurbation (urban)§ | 1.46‡ (1.24, 1.71) | 0.08 | 1.46‡ (1.22, 1.74) | 0.09 | |||
| 6 | Minor conurbation (urban)§ | 1.04 (0.73, 1.48) | 0.18 | 0.93 (0.63, 1.37) | 0.2 | |||
| 7 | % of pupils whose first language is known or believed to be other than English | 1.00 (0.95, 1.05) | 0.03 | 1.01 (0.95, 1.07) | 0.03 | 1.23 (0.33, 4.59) | ||
| 8 | % of pupils reaching the expected standard in reading, writing and maths | 0.99 (0.95, 1.04) | 0.02 | 1.00 (0.96, 1.05) | 0.02 | 0.98 (0.50, 1.91) | ||
| 9 | % of disadvantaged pupils | 1.01† (1.00, 1.02) | 0 | 1.01* (1.00, 1.02) | 0.004 | 1.16 (1.02, 1.33) | ||
| 10 | % of overweight or obese adults | 0.97 (0.86, 1.11) | 0.07 | 0.89 (0.51, 1.52) | ||||
| 11 | % of physically active adults | 0.93 (0.83, 1.05) | 0.06 | 0.73 (0.44, 1.22) | ||||
| 12 | % of physically active children | 1.02 (0.93, 1.11) | 0.04 | 1.04 (0.85, 1.26) | ||||
| 13 | % of children who are overweight or obese | 0.97 (0.88, 1.07) | 0.05 | 0.92 (0.72, 1.17) | ||||
| Model summary | ||||||||
| Intercept | 0.29‡ (0.26, 0.32) | 0.05 | 0.22‡ (0.18, 0.26) | 0.088 | 0.22‡ (0.18, 0.26) | 0.09 | ||
| ICC | 0.09 | 0.07 | 0.06 | |||||
| Likelihood ratio χ2 test (ordinary logistic vs multilevel logistic model) | 557.42‡ | 302.24‡ | 247.12‡ | |||||
| AIC/BIC | 16 185.3 (BIC) | 14 290.5 (BIC) | 12 688.2 (BIC) | |||||
| 16 170.0 (AIC) | 14 207.8 (AIC) | 12 577.0 (AIC) | ||||||
*P≤0.05.
†P≤0.01.
‡P≤0.001.
§Reference group=city and town.
ⱡ Reference group = Academy
Parameters from 1 to 13 all are fixed-effects estimates. AIC, Akaike information criterion; BIC, Bayesian information criterion; ICC, intraclass correlation coefficient.