The 2006–2010 National Survey of Family Growth: Sample design and analysis of a continuous survey

Institute for Social Research, University of Michigan, USA.
Vital and health statistics. Series 2, Data evaluation and methods research 06/2010; 150(150):1-36.
Source: PubMed


The National Survey of Family Growth (NSFG) collects data on pregnancy, childbearing, men's and women's health, and parenting from a national sample of women and men 15-44 years of age in the United States. This report describes the sample design for the NSFG's new continuous design and the effects of that design on weighting and variance estimation procedures. A working knowledge of this information is important for researchers who wish to use the data. Two data files are being released--the first covering 2.5 years (30 months) of data collection and the second after all data have been collected. This report is being released with the first data file. A later report in this Series will include specific results of the weighting, imputation, and variance estimation.
The NSFG's new design is based on an independent, national probability sample of women and men 15-44 years of age. Fieldwork was carried out by the University of Michigan's Institute for Social Research (ISR) under a contract with the National Center for Health Statistics (NCHS). In-person, face-to-face interviews were conducted by professional female interviewers using laptop computers.
Analysis of NSFG data requires the use of sampling weights and estimation of sampling errors that account for the complex sample design and estimation features of the survey. Sampling weights are provided on the data files. The rate of missing data in the survey is generally low. However, missing data were imputed for about 600 key variables (called "recodes") that are used for most analyses of the survey. Imputation was accomplished using a multiple regression procedure with software called IVEware, available from the University of Michigan website.

5 Reads
  • Source
    • "We utilize six data sources in the current analysis. To study the contraceptive behavior of cohabitors in the United States, this research relies on data from the 1995 and 2006/10 National Surveys of Family Growth (NSFG) (Lepkowski et al. 2010). The samples are large, including 10,847 women in 1995 and 7,356 women in 2006/10. "

    Demographic Research 01/2015; 32:147-182. DOI:10.4054/DemRes.2015.32.5 · 1.20 Impact Factor
  • Source
    • "We included sociodemographic variables often associated with contraceptive use [1], including withdrawal [8]: respondent's age at the time of the interview [15] [16] [17] [18] [19] [20] [21] [22] [23] [24], both respondent's and their mother's highest level of education (no high school diploma/GED, high school diploma/GED, and any college or more), race/ethnicity (non-Hispanic white, non-Hispanic black, non-Hispanic other and Hispanic ), union status (married, cohabitating and single/not cohabitating), reception of public assistance in the last year (yes, no), religious affiliation (no religion, Catholic, Protestant or other) and health insurance status (none, any private insurance or any public insurance). "
    [Show abstract] [Hide abstract]
    ABSTRACT: Use of withdrawal (coitus interruptus) has consequences for reproductive health, but few nationally representative studies exist. We 1) examined patterns of withdrawal among 15-24 year-old women and men, and 2) explored withdrawal’s associations with socio-demographic, psychological, and sexual factors.
    Contraception 12/2014; 91(4). DOI:10.1016/j.contraception.2014.12.005 · 2.34 Impact Factor
  • Source
    • "All models adjust for sampling design effects, as explained in the next section. Survey comparability and sampling design effects The 2006–2010 continuous survey was designed to support comparisons to 2002 (Lepkowski et al. 2010). Both surveys drew on the same sampling frame, and although there were some differences in questionnaire content across surveys (and across years within the continuous survey), these differences were irrelevant to the analyses presented here. "
    [Show abstract] [Hide abstract]
    ABSTRACT: Most research on lesbian families draws on either nonrepresentative samples or on representative samples of female-partner households. In contrast, this article uses individual-level, nationally representative survey data to provide a demographic description of lesbian parents in the United States. Pooling data from the 2002 and 2006–2010 rounds of the National Survey of Family Growth yielded a sample of 15,784 women aged 20–44 years, about 1.3 % of whom are lesbians. Defining parents broadly to include legal and social parents, we find that about 23 % of lesbians are parents, compared to about 68 % of heterosexual and 56 % of bisexual women. Lesbians become parents through a more diverse set of pathways than other women, including adoption and parenting a spouse or partner’s child. Consistent with patterns in the broader population, but at odds with media portrayals, lesbian parents are more likely than lesbian nonparents to be women of color and foreign-born, and most appear to have become parents in prior heterosexual relationships. We found evidence, however, of a convergence in the pathways women follow to parenthood, with lesbians’ probability of biological parenthood increasing and their probability of adoptive or social parenthood decreasing between the two surveys. Recent changes in the legal and social context and improvements in medical technology provide grounds for speculating about this convergence. We recognize, however, that these speculations cannot be tested without population-based data collection efforts aimed at providing richer information on the diversity of family experiences in the contemporary United States.
    Population Research and Policy Review 08/2014; 33(4). DOI:10.1007/s11113-013-9296-3 · 0.76 Impact Factor
Show more