Residential self-selection bias in the estimation of built environment effects on physical activity between adolescence and young adulthood

Table 3 Baseline and changes in built and socioeconomic environment characteristics between adolescence (Wave I, 1994-95) and young adulthood (Wave III, 2001-02), by residential relocation status¹

	Movers (n = 8,525)		Non-movers (n = 4,176)
Measure (geographic area²)	mean (SE)	median (IQR)	mean (SE)	median (IQR)	P ³
Landscape diversity (1 k)
Baseline	0.53 (0.01)	0.58 (0.43, 0.67)	0.54 (0.01)	0.58 (0.46, 0.67)	0.3
Change (Wave III-Wave I)	-0.01 (0.01)	-0.02 (-0.15, 0.12)	-0.04 (0.01)	-0.03 (-0.14, 0.06)	0.002
Pay facility availability (count/10,000 population) (3 k)
Baseline	2.64 (0.23)	1.71 (0.00, 3.71)	2.43 (0.20)	1.61 (0.34, 3.40)	1.0
Change (Wave III-Wave I)	2.00 (0.21)	1.38 (-0.05, 3.98)	2.10 (0.23)	1.02 (0.06, 3.04)	0.7
Public facility availability (count/10,000 population) (3 k)
Baseline	0.30 (0.05)	0.00 (0.00, 0.29)	0.28 (0.05)	0.00 (0.00, 0.31)	0.7
Change (Wave III-Wave I)	0.33 (0.05)	0.00 (0.00, 0.54)	0.18 (0.05)	0.00 (0.00, 0.30)	0.02
Alpha street connectivity (1 k)
Baseline	0.31 (0.02)	0.30 (0.22, 0.38)	0.33 (0.02)	0.30 (0.22, 0.38)	0.5
Change (Wave III-Wave I)	-0.002 (0.019)	-0.005 (-0.097, 0.077)	-0.018 (0.016)	-0.003 (-0.023, 0.012)	0.5
Median household income, 1,000's U.S. dollars (CT) ⁴
Baseline	38.9 (1.3)	37.3 (27.5, 46.7)	41.2 (1.3)	39.8 (28.7, 51.5)	0.002
Change (Wave III-Wave I)	0.3 (1.0)	1.9 (-7.8, 9.8)	2.4 (0.3)	2.5 (-1.6, 5.8)	0.02
Crime, per 100,000 population (Co)
Baseline	5,300 (247)	5,369 (3,072, 6,975)	5,547 (238)	5,528 (3,647, 6,459)	0.05
Change (Wave III-Wave I)	-551 (171)	-669 (-1,950, 309)	-879 (161)	-1,081 (-1,645, -350)	0.005

¹ National Longitudinal Study of Adolescent Health (United States). Residential relocation defined as > 1/4 mile Euclidean distance between Wave I and Wave III residential locations
² Geographic areas consistent with the strongest associations with MVPA in a previous study were selected for each variable.
³2-sided test of difference between movers and non-movers in males and females determined from adjusted Wald tests (continuous variables) and design-based F-tests (categorical variables), weighted and corrected for clustering. Statistical tests were performed on natural log-transformed pay facilities, public facilities, and median household income to correct for skewness.
⁴ Wave I values inflated to 2000 U.S. dollars
1 k and 3 k, radius of Euclidean neighborhood buffer in kilometers (k); CT, Census Tract; Co, County; IQR, Interquartile Range; SE, standard error

ISSN: 1479-5868