A Tale of Two Fathers
Appendix I. Methodology
NATIONAL SURVEY OF FAMILY GROWTH ANALYSIS
Description of Survey and Analysis
Since its inception in 1973, the National Survey of Family Growth (NSFG) has been conducted by the National Center for Health Statistics (NCHS) in seven cycles. While originally focused on the fertility and family formation patterns of women of childbearing age, since 2002 the NSFG has expanded to interview men as well.
Cycle 7 of the NSFG includes interviews conducted from 2006 to 2010 with a national sample of men and women ages 15-44 living in households in the United States. In each year, a nationally representative sample of men and women in 33 areas (Primary Sampling Units, or PSUs) was interviewed. By the end of four years of interviewing, in June 2010, more than 22,600 interviews had been completed in 110 areas.
The first Cycle 7 public use data files were released in May 2010 and consist of 13,495 interviews conducted between 2006 and 2008 (7,356 female and 6,139 male). These interviews are the basis for all NSFG analyses included in this report.
All noted subgroup differences were tested at the 95% confidence interval. An absence of comment on any subgroup difference does not necessarily mean that the difference was non-significant. Statistics for cell sizes with unweighted counts smaller than 100 are excluded.
Pew Research Center Survey
Results for this survey are based on telephone interviews conducted with a nationally representative sample of 2,006 adults living in the continental United States. A combination of landline and cellular random digit dial (RDD) samples was used to represent all adults in the United States who have access to either a landline or cellular telephone. A total of 1,344 interviews were completed with respondents contacted by landline phone and 662 interviews were completed with respondents contacted on their cell phone. The data are weighted to produce a final sample that is representative of the general population of adults in the continental United States.
- Interviews conducted May 26-29 and June 2-5, 2011
- 2,006 interviews
- Margin of sampling error is plus or minus 2.6 percentage points for results based on the total sample at the 95% confidence level.
Interviews were conducted under the direction of Princeton Survey Research Associates International. Respondents in the landline sample were selected by randomly asking for the youngest adult male or female who is now at home. Interviews in the cell sample were conducted with the person who answered the phone, if that person was an adult 18 years of age or older.