Feasibility and validity of a statistical adjustment to reduce self-report bias of height and weight in wave 1 of the Add Health study

Table 2 Final multiple regression models predicting measured height and weight at Wave 2 from self-reported height, weight, and other candidate factors at Wave 2

	Model 1					Model 2
	Measured weight at wave 2					Measured height at wave 2
	B	SE	95 % CI		P	B	SE	95 % CI		P
Intercept	8.20	1.47	5.31	11.09	<.0001	9.53	0.27	9.01	10.06	<.0001
SR Height W2						0.85	0.00	0.84	0.86	<.0001
SR Weight W2	0.95	0.00	0.94	0.96	<.0001
Age in months	-0.01	0.01	-0.02	0.00	0.049	0.00	0.00	0.00	0.00	<.0001
Puberty status	0.22	0.10	0.03	0.41	0.021	0.05	0.01	0.03	0.07	<.0001
Weight status by SR (HW)
UW	3.67	0.59	2.51	4.83	<.0001	-0.92	0.07	-1.06	-0.79	<.0001
OW	2.46	0.35	1.78	3.14	<.0001	0.17	0.04	0.10	0.25	<.0001
OB	2.86	0.51	1.86	3.87	<.0001	0.40	0.04	0.32	0.49	<.0001
Body size estimation (About right)
Underweight	-1.76	0.30	-2.35	-1.17	<.0001	0.13	0.04	0.06	0.20	<0.001
Overweight	3.68	0.28	3.13	4.23	<.0001	-0.11	0.03	-0.17	-0.04	0.001
Dieting to lose weight	0.93	0.31	0.32	1.54	0.003	-0.08	0.04	-0.15	-0.01	0.029
Depressive symptoms^b	-0.04	0.03	-0.10	0.01	0.115	-0.01	0.00	-0.02	-0.01	0.000
Self-rated health status	-0.39	0.12	-0.64	-0.15	0.002	-0.04	0.01	-0.07	-0.01	0.009
Self-esteem^b	0.35	0.20	-0.04	0.75	0.079	-0.06	0.02	-0.10	-0.01	0.014
Race/ethnicity (White)^a
Hispanic/Latino						-0.23	0.04	-0.30	-0.15	<.0001
Black/African American^c						0.03	0.03	-0.04	0.10	0.365
Asian						-0.29	0.06	-0.41	-0.17	<.0001
Other^c						0.02	0.16	-0.29	0.34	0.882
Sex (Male)	^a					0.65	0.03	0.59	0.71	<.0001
Parent education (College degree)	^a					0.08	0.02	0.03	0.13	0.001
R²	0.91					0.89

Reference groups are shown in parentheses. Design-based sample weights were applied prior to analyses. Two-way interactions were tested for all significant predictors but none were retained as contributing factors based on AIC selection
SR, self-reported; HW, healthy weight; UW, underweight; OW, overweight; OB, obesity
^aIn Model 1, candidate predictors with empty cells were removed during the AIC selection procedure as non-contributing factors
^bIn Model 1, depressive symptoms and self-esteem were not significant alone but were retained during AIC selection as contributing factors
^cIn Model 2, African American and Other race/ethnic categories were not significant (shown above) but were retained during AIC selection because the class variable overall was a contributing factor

ISSN: 1471-2288