Skip to main content

Table 4 Interaction effects between parent-perceived neighbourhood physical activity places and facilities, safety-related attributes and physical-activity-related parenting practices on preschool-aged children’s physical activity and screen time

From: Parent-perceived neighbourhood environment, parenting practices and preschool-aged children physical activity and screen time: a cross-sectional study of two culturally and geographically diverse cities

Neighbourhood environmental attribute

Moderator

Outcome

Association at below average value of moderator

Association at average value of moderator

Association at above average value of moderator

   

Reg. coef. (95% CI)

Reg. coef. (95% CI)

Reg. coef. (95% CI)

Availability of places for children’s PA

Informal social control - civic engagement for neighbourhood enhancement

MVPA – weekday

eb = 0.99 (0.96, 1.01)

eb = 1.01 (0.99, 1.02)

eb = 1.03*(1.01, 1.05)

Availability of places for children’s PA

Informal social control - civic engagement for neighbourhood enhancement

Total PA – weekend day

b = 0.24 (−2.84, 3.33)

b = 2.91*(0.47, 5.35)

b = 5.58**(1.86, 9.31)

Availability of active-play equipment

Traffic hazards

Total PA – weekend day

b = 6.36*(0.37, 12.34)

b = 2.02 (−1.46, 5.51)

b = −2.31 (−6.34, 1.72)

Availability of places for children’s PA

Parental engagement

Total PA - weekday

b = 3.32*(0.69, 5.94)

b = 1.02 (−1.14, 3.18)

b = −1.28 (−4.22, 1.66)

Availability of active-play equipment

Parental engagement

MVPA - weekday

eb = 0.96*(0.94, 0.99)

eb = 1.00 (0.98, 1.02)

eb = 1.05*(1.00, 1.11)

Signs of physical and social disorder

Restrictions for safety concerns

Screen time – weekend day

b = −0.46 (−1.19, 0.27)

b = 0.07 (−0.38, 0.53)

b = 0.61*(0.11, 1.10)

  1. b, estimate of unstandardised regression coefficient; eb, exponentiated estimate of unstandardised regression coefficient; CI confidence interval, PA physical activity, MVPA moderate-to-vigorous physical activity
  2. *p < .05; ** p < .01. All models were adjusted for socio-demographic characteristics. Models of total PA and MVPA were adjusted for average accelerometer wear time. Regression coefficients of neighbourhood environmental attributes moderated by other environmental attributes were not adjusted for PA-related parenting practices. Below and above average value of moderator means respectively 1 standard deviation below and above the moderator’s mean. Only statistically significant (p < .05) interaction effects are reported in this table. Table S3 reports all tested interaction effects (statistically significant and not significant)