Response to Comments

NLS Response to Comments on NLSY79 (OMB Control # 1220-0109) from Michael Lovenheim and Elizabeth Peters

August 30, 2016

Response to Michael Lovenheim:

1. Sources of college payments:  These questions refer to a retrospective module added in 2014 to the NLSY79 that asks about education costs for respondents’ children ages 15 or older.  Note that some of the children may be in their 20s, 30s, or even 40s at this point.  Professor Lovenheim suggests we get very detailed information about Federal aid of each type, private loans, grants, and such.  This is an interesting idea, but could be difficult for respondents’ whose children have been out of school for a long time.  Note that we do interview NLSYY79 female respondents’ biological children, and so if one looks in the NLSY79 Young Adult data set, one can find more detailed information on financial aid, college costs, and such for these same children.  The NLSY97 contains a great deal more information about how more recent youth cohorts have paid for college.

2. National Student Clearinghouse administrative records: Professor Lovenheim suggests we link Child/Young Adult NLSY79 data to administrative enrollment and attainment records.  In general, we think linking to administrative data would help expand the uses of the NLSY79.  BLS has recently made arrangement to have the NLSY79 data available in the Census RDCs in attempt to facilitate matching to other data sources.  On all issues of linking, the program will need to look into issues of consent and confidentiality.  These requirements will be determined in consultation with the support offices in BLS. 

3. Federal financial aid administrative records:  As noted above, we believe linking to administrative records could be extremely beneficial, but will need to explore the consent and confidentiality concerns.

4. Social Security and federal assistance administrative records:  The NLS has a planned meeting this fall with SSA regarding linking to administrative records.  In the past, issues of consent/confidentiality have stymied any agreement. 

Response to H. Elizabeth Peters:

1. Prime-age women’s increasing death rates:  Examining behavioral changes between cohorts in relation to drug abuse, mental health, obesity, and smoking, among others, is an important research topic.  As Dr. Peters points out, a major advantage of the NLS data sets is that respondents have been followed over time, and the NLS data sets all gather information across multiple domains.  We believe the information already available in the NLSY79 will allow for this type of research, and we will continue to ask questions pertinent to the study of this topic (for example, we will be introducing a ‘health at 60’ module).

2. Older women’s cohabitation behavior:  Cohabitation behavior among older women is another research project that one can pursue with NLSY79 data, and we continue to collect dates of cohabitation, partner’s characteristics, and dates of changes in marital status as the respondents age.

3. The role of the family as providing a social safety net:  Time and money transfers to and from family members is a topic we have pursued with the NLS women’s cohorts, although we found that doing it right is very time intensive.  We do collect some information relevant to this topic in the NLSY79 (for example, household structure and caregiving).  As the respondents age, we will consider transfers questions as well as others related to aging and retirement.

4. Declining women’s labor force participation:  Dr. Peters suggests studying differences between labor force participation over time for women in the NLSY79 and NLSY97 cohort.  This can certainly be done, as the surveys contain a complete work history, as well as marital and fertility histories, and other relevant information.

1