Skip to main content
. Author manuscript; available in PMC: 2013 Jan 1.
Published in final edited form as: Health Place. 2012 Jan;18(1):39–45. doi: 10.1016/j.healthplace.2011.09.008

Table 4.

Difference of differences analysis for the observation data (with control parks) and Propensity score analysis for self-reported data (with control parks)

Difference of differences analysis for the observation data
Variable Parameter estimate
(S.E.)
P value
Model 1 (Observation of users)
Intercept Comparison parks at baseline 919.1 (174) .0001
Fitness Zone Parks at baseline −305.3 (247) .23
Change in users in comparison parks −14.3 (141) .92
Change in users in Fitness Zone parks 207.3 (199) .31
Model 2 (Expenditure of METS)
Intercept (comparison parks at baseline) 2191 (425) .0001
Fitness Zone baseline −687 (601) .27
Change in METS for the comparison parks −100 (350) .78
Change in METS at Fitness Zone parks 685 (496) .18
Propensity score analysis: self-reported exercise and park use
Average #
exercise
sessions/week
New user in the
past 6 months (%)
New user in the
past one month
(%)
Uses park use
1x/week or
more(%)
Control
parks
FZ
parks
Control
parks
FZ parks Control
parks
FZ parks Control
parks
FZ
parks
Baseline 2.13 2.36 8.3 7.1 5.1 3.6 85.3 79.3
1st follow-
up
2.17 2.50
(p=.49)
6.0 11.4
(p=.014)
2.6 6.3
(p=.007)
81.2 81.6
(p=.081)