The National Survey of Family Growth, Cycle 5, 1995 (NSFG) is the fifth in a series of periodic surveys of women 15-44 years of age. Previous surveys were conducted in 1973, 1976, 1982, and 1988 with a telephone reinterview of the 1988 respondents in 1990. Topics covered in previous interviews included the month and year of first intercourse; pregnancy, contraceptive, marital, and cohabitation histories, employment and occupation, child care, fecundity and sterility, prenatal medical care, family planning services, birth expectations, ethnicity, education, religion, and income. For Cycle 5, event histories of education, living arrangements during childhood, and work have been added, along with complete marital and cohabitation histories, and sexual partner histories for 5 years prior to the interview. The survey also included, for the first time in 1995, characteristics of male partners, new items on consistency of contraceptive use, new questions on pregnancy wantedness, and a computer-assisted self- administered section containing questions on sensitive topics such as abortion and forced intercourse.

The overall objective of the NSFG is to supplement the vital statistics of fertility and of family formation and dissolution, with more detailed data on the "intermediate variables" which shape these trends and on the health and socioeconomic contexts in which they occur. The uses of the data gathered in the NSFG are broad, though mostly in the health and demographic fields. The 1995 NSFG obtained detailed information on factors affecting childbearing from a national probability sample of 10,847 women 15 to 44 years of age.

The 1995 NSFG data are divided into two files. The respondent file (Data Set N8-O4) contains most of the information from the Cycle 5 instrument (see below) using the individual respondent as the unit of analysis. This data set contains 5,711 variables for 10,847 cases. The Interval file (Data Set O5) uses the pregnancy as the unit of analysis and contains data for each pregnancy (regardless of the result) for all of the respondents. Several related variables from the respondent file are included as well. This data set includes 21,332 cases.