Skip to main content
. Author manuscript; available in PMC: 2016 Jan 1.
Published in final edited form as: Biodemography Soc Biol. 2015;61(1):81–97. doi: 10.1080/19485565.2014.981794

Table 3. Estimates from best fit polynomial growth curve models for longitudinal measurements of BMI z-scores.

White Black
Variable Coef SE p-value Coef SE p-value
Intercept 0.424 0.088 <.0001 0.460 0.121 0.0002
Early maturity 0.0042 0.0999 0.97 0.117 0.104 0.26
Low family SES 0.330 0.103 0.001 0.162 0.100 0.105
Neighborhood median family income ($1000 increase) -- -- -- 0.0048 0.0014 0.0008
Age -0.0014 0.0064 0.83 -0.037 0.022 0.09
Age × low family SES -0.0055 0.0116 0.63 0.029 0.014 0.03
Age × neighborhood median family income 0.000005 0.00051 0.99
Age2 -0.0073 0.0015 <0.0001 -0.0045 0.0037 0.22
Age2 × low family SES -0.0032 0.0027 0.24 -0.0072 0.0022 0.001
Age2 × neighborhood median family income -- -- -- 0.000033 0.000092 0.72
Age3 -- -- -- 0.0025 0.0010 0.01
Age3 × low family SES -- -- -- -0.0010 0.0006 0.08
Age3 × neighborhood median family income -- -- -- -0.00001 0.000024 0.56

Random effect variance parameter estimates Estimate SE p-value Estimate SE p-value
Intercept 0.68 0.06 <0.0001 0.84 0.07 <0.0001
Age 0.0050 0.0008 <0.0001 0.004 0.0007 <0.0001
Age2 0.00011 0.00004 0.006 0.00004 0.00003 0.11
AR-1 0.47 0.06 <0.0001 0.56 0.07 <0.0001
Residual 0.13 0.02 <0.0001 0.12 0.02 <0.0001

AR-1, first order autoregressive process correlation parameter; SE, standard error; SES, socioeconomic status;