Skip to main content
. 2016 Dec 30;14(1):35. doi: 10.3390/ijerph14010035

Table 3.

Hurdle models.

Independent Variables Total Number of Park Users Male Park Users Female Park Users
Logistic Regression a Poisson Model b Logistic Regression a Poisson Model b Logistic Regression a Poisson Model b
OR 95% CI Exp. B 95% CI OR 95% CI Exp. B 95% CI OR 95% CI Exp. B 95% CI
Time of day (ref = morning)
lunch 4.40 *** 1.94–9.99 0.98 0.76–1.29 7.29 *** 3.01–17.68 0.71 0.49–1.02 1.38 0.63–3.03 1.57 * 1.06–2.33
afternoon 6.82 *** 2.96–15.62 2.09 *** 1.65–2.66 15.46 *** 6.21–38.48 1.53 * 1.10–2.12 3.76 *** 1.72–8.23 2.01 *** 1.40–2.88
evening 4.46 *** 1.97–10.09 1.40 ** 1.09–1.81 6.01 *** 2.50–14.42 1.14 0.81–1.61 3.26 ** 1.49–7.10 1.36 0.92–1.99
Type of park area (ref = grassy area with a playground)
trail 13.09 *** 4.24–40.41 1.79 0.90–3.59 15.79 *** 4.38–56.96 2.09 * 1.12–3.92 4.53 * 1.37–14.96 1.43 0.96–2.11
wood 0.05 *** 0.02–0.14 0.45 *** 0.31–0.67 0.06 *** 0.02–0.15 0.65 * 0.43–0.97 0.02 *** 0.00–0.12 0.55 0.18–1.69
Day type (ref = week)
weekend 0.38 * 0.15–0.98 0.96 0.63–1.45 0.72 0.26–1.92 1.09 0.72–1.66 0.46 0.20–1.06 0.75 0.53–1.07
Independent Variables Child/Adolescent Park Users Adult/Senior Park Users
Logistic Regression a Poisson Model b Logistic Regression a Poisson Model b
OR 95% CI Exp. B 95% CI OR 95% CI Exp. B 95% CI
Time of day (ref = morning)
lunch 4.86 ** 1.71–13.79 2.05 0.96–4.34 3.05 ** 1.35–6.88 0.89 0.66–1.21
afternoon 17.81 *** 6.35–49.98 4.91 *** 2.38–10.11 4.54 *** 2.00–10.29 1.17 0.88–1.54
evening 10.78 *** 3.86–30.06 3.28 ** 1.56–6.89 2.45 * 1.09–5.50 1.09 0.82–1.46
Type of park area (ref = grassy area with a playground)
trail 4.26 *** 1.81–10.05 1.66 0.82–3.34 8.15 *** 2.65–25.00 1.51 * 1.08–2.11
wood 0.09 *** 0.03–0.23 0.67 0.43–1.05 0.04 *** 0.01–0.12 0.53 * 0.31–0.92
Day type (ref = week)
weekend 1.37 0.49–3.83 0.82 0.48–1.40 0.29 ** 0.13–0.68 0.96 0.77–1.20

a The logistic regression model estimated the association of the independent variables with the odds of observing park users in the park areas. b The Poisson models (Exp. B) estimated the proportional difference in number of users associated with a one-unit difference in the independent variables for park target areas where at least one person was present. OR = Odds Ratio, CI = Confidence interval, Exp. B = Exponent of B, ref = reference category. * α < 0.05, ** α < 0.01, *** α < 0.001.