Table 4.
Linear mixed-effects model of log-transformed activity as a function of owner and environmental variables using the all-visits dataset.
| Variables | Estimate | Error | 2.5% CL | 97.5% CL | p-value | Marginal mean (LM) | 2.5% CL | 97.5% CL | Marginal mean (BART) | 2.5% CL | 97.5% CL |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Intercept | 28.650 | 0.038 | 26.581 | 30.874 | < 0.001 | – | – | – | – | – | – |
| Latitude | 0.072 | 0.009 | 0.054 | 0.090 | < 0.001 | – | – | – | – | – | – |
| Climate | |||||||||||
| Cold | Reference category | 41.8 | 41.2 | 42.4 | 40.4 | 40.0 | 41.8 | ||||
| Hot humid | −0.006 | 0.010 | −0.026 | 0.014 | 1.000 | 41.5 | 41.1 | 41.9 | 41.3 | 40.6 | 42.2 |
| Hot dry | 0.060 | 0.013 | 0.035 | 0.087 | < 0.001 | 44.2 | 43.5 | 45.0 | 44.1 | 42.8 | 45.7 |
| Marine | 0.061 | 0.014 | 0.034 | 0.090 | < 0.001 | 44.6 | 43.5 | 45.7 | 44.3 | 41.7 | 47.1 |
| Season | |||||||||||
| Spring | Reference category | 44.1 | 43.7 | 44.5 | 42.7 | 42.2 | 43.4 | ||||
| Summer | −0.009 | 0.003 | −0.014 | −0.003 | 0.083 | 43.0 | 42.6 | 43.4 | 40.2 | 39.7 | 40.8 |
| Autumn | −0.012 | 0.003 | −0.017 | −0.006 | 0.004 | 43.3 | 42.9 | 43.7 | 41.7 | 41.1 | 42.3 |
| Winter | −0.082 | 0.003 | −0.087 | −0.077 | < 0.001 | 41.6 | 41.2 | 42.0 | 41.8 | 41.2 | 42.4 |
| Day | |||||||||||
| Weekday | Reference category | 41.3 | 41.0 | 41.7 | 40.6 | 40.2 | 41.1 | ||||
| Weekend | 0.114 | 0.002 | 0.110 | 0.119 | < 0.001 | 44.7 | 44.3 | 45.1 | 44.1 | 43.6 | 44.6 |
| Location | |||||||||||
| Rural | Reference category | 43.6 | 43.1 | 44.1 | 42.6 | 41.9 | 43.3 | ||||
| Urban | −0.027 | 0.006 | −0.038 | −0.014 | < 0.001 | 42.4 | 42.0 | 42.8 | 41.9 | 41.2 | 42.6 |
| Owner age | |||||||||||
| 18–30 | Reference category | 44.6 | 44.1 | 45.1 | 37.8 | 39.0 | 39.8 | ||||
| 30–45 | −0.031 | 0.005 | −0.040 | −0.022 | < 0.001 | 42.8 | 42.4 | 43.2 | 38.2 | 39.1 | 40.9 |
| 45–60 | −0.027 | 0.006 | −0.039 | −0.016 | < 0.001 | 42.7 | 42.3 | 43.2 | 39.9 | 41.1 | 42.3 |
| 60+ | −0.029 | 0.008 | −0.044 | −0.014 | 0.002 | 41.9 | 41.3 | 42.5 | 42.8 | 44.7 | 46.6 |
| Day-owner age interaction; day = weekend (reference category = weekday) | |||||||||||
| 18–30 | Reference category | – | – | – | – | – | – | ||||
| 30–45 | −0.018 | 0.003 | −0.023 | −0.0013 | < 0.001 | – | – | – | – | – | – |
| 45–60 | −0.030 | 0.003 | −0.036 | −0.025 | < 0.001 | – | – | – | – | – | – |
| 60+ | −0.064 | 0.004 | −0.070 | −0.057 | < 0.001 | – | – | – | – | – | – |
| Season-climate interaction (season reference category = spring; climate reference category = cold) | |||||||||||
| Summer: hot humid | −0.039 | 0.003 | −0.046 | −0.033 | < 0.001 | – | – | – | – | – | – |
| Summer: hot dry | −0.047 | 0.005 | −0.056 | −0.038 | < 0.001 | – | – | – | – | – | – |
| Summer: marine | 0.025 | 0.006 | 0.013 | 0.037 | 0.003 | – | – | – | – | – | – |
| Autumn: hot humid | −0.004 | 0.004 | −0.011 | 0.003 | 0.850 | – | – | – | – | – | – |
| Autumn: hot dry | −0.012 | 0.005 | −0.022 | 0.003 | 0.249 | – | – | – | – | – | – |
| Autumn: marine | −0.010 | 0.006 | −0.022 | 0.002 | 0.612 | – | – | – | – | – | – |
| Winter: hot humid | 0.043 | 0.003 | 0.036 | 0.050 | < 0.001 | – | – | – | – | – | – |
| Winter: hot dry | 0.059 | 0.005 | 0.049 | 0.070 | < 0.001 | – | – | – | – | – | – |
| Winter: marine | 0.010 | 0.006 | −0.002 | 0.002 | 0.647 | – | – | – | – | – | – |
The marginal means for the linear model (LM) and the BART model are displayed. The details include intercept, fixed effects coefficients and their 2.5 and 97.5% confidence limits (CL) and p-values. Estimated marginal mean activity is displayed for values of categorical variables, where all other variables are adjusted for mean values (dog age of 6.0 years and owner age of 41.7 years). The model had an R2 of 0.533 and a RMSE of 0.447 on log-transformed active minutes, whilst pet ID as a random effect accounted for 0.246 of the variance in the data.