. Author manuscript; available in PMC: 2013 Oct 1.

Published in final edited form as: Demography. 2012 Aug;49(3):989–1009. doi: 10.1007/s13524-012-0113-0

Table 2.

Test for recall bias: effect of survey year on reported miscarriage controlling for linear time trend in incidence of miscarriage: Females aged 13–25 and 13–35, 1970–2000

	Aged 13–25			Aged 13–35
	(1)	(2)	(3)	(4)	(5)	(6)
Year of Pregnancy	0.005 (0.005)	0.006^* (0.003)		0.004 (0.004)	0.007^** (0.003)
1995 NSFG	0.034 (0.072)		0.068 (0.060)	0.084 (0.055)		0.113^* (0.045)
2002 NSFG	0.035 (0.102)		0.104 (0.068)	0.073 (0.084)		0.133^* (0.054)
χ²	250.43	248.03	248.77	351.84	349.81	343.51
AIC	59,378.47	59,374.91	59,378.23	103,772.8	103,773.2	103,772.8
N	24,541	24,541	24,541	38,119	38,119	38,119

Notes: Standard errors clustered by respondent are reported in parentheses. All specifications control for race, religion, own and maternal education, prior miscarriage, whether pregnancy occurred at “the right time,” and age at conception.

Source: Authors calculations from National Survey of Family Growth data.

p <.05;

^**

p <.01