Noninterviews and Item Nonresponse

Missing Data: Noninterviews and Item Nonresponse

Missing values are indicated in the data and on the codebook page for each individual question. Following general NLS convention, a response of "don't know" to an individual question is coded "-2," a refusal to answer an individual question is coded "-1," and an invalid skip is coded "-3." (Invalid skip means that the respondent should have answered the question but didn't; this was more common in the paper-and-pencil interviews when the respondent or interviewer might make an error following the skip pattern on a paper instrument. The incidence of invalid skips has been significantly reduced in the computer-assisted interviews.) In the NLSY79 Child and Young Adult, noninterviews and valid skips (when a respondent was purposely not eligible for a question) have been collapsed into a "-7." Thus, users must know how to distinguish between noninterviews and valid skips for some types of research.

Starting with the 1994 survey round for the Young Adult and the 1998 survey round for the Child, users can rely on the Interview/Assessment flags in the CHILD BACKGROUND area of interest to determine child or young adult interview status. For Child survey rounds prior to 1998, users can identify interviewed respondents as those with sampling weight values greater than "0." The Child sampling weights (values greater than "0") can be used to accurately identify children assessed in any survey year. See the discussion below on the specific procedures for the creation of the Young Adult sample weights that distinguish sample eligibility from interview status.

Variables such as age of child at mother's interview date occasionally show missing values for child cases in which the mother was not interviewed, even though the children themselves were interviewed. Since the AGECH and AGEMOM variables are based on inputs from the mother's record, users may want to default to the child's age at the date of assessment (e.g., CSAGE2010 in 2010). Users should note that Young Adult children are often interviewed in a given survey round, regardless of the interview status of their mothers. Contact with children under age 15 is largely predicated on completion of the mother's own interview and her authorization to assess her younger children.

Interview Status Flags

A set of variables has been constructed that indicates the interview status for children, young adults and their mothers. These interview status flags are cross-sectional and thus only reflect valid values for cases from a particular survey round. 

Created variables

This list includes child interview status variables, which are available starting in 1998. Also listed starting in 2002 is a flag that indicates the sequence in which the Child Supplement and Mother Supplement interviews took place. The "MINTRV" variables in this series indicate if the mother was interviewed as part of the main Youth survey. Finally, the Young Adult interview status flags available for 1994-present are also in the CHILD BACKGROUND area of interest.

1998

CINTRV98. INTERVIEW STATUS OF CHILD
CASSMT98. ASSESSMENT STATUS OF CHILD
INCSUP98. DOES CHILD HAVE A CHILD SUPPLEMENT?
INMSUP98. DOES CHILD HAVE A MOTHER SUPPLEMENT?
INCSAS98. DOES CHILD 10-14 YEARS OLD HAVE A CSAS RECORD?
MINTRV98. INTERVIEW STATUS OF MOTHER

2000

CINTRV2000. INTERVIEW STATUS OF CHILD
INCSUP2000. DOES CHILD HAVE A CHILD SUPPLEMENT?
INMSUP2000. DOES CHILD HAVE A MOTHER SUPPLEMENT?
INCSAS2000. DOES CHILD 10-14 YEARS OLD HAVE A CSAS RECORD?
MINTRV2000. INTERVIEW STATUS OF MOTHER

2002-2014

INCSUPyyyy. DOES CHILD HAVE A CHILD SUPPLEMENT?
CSCOMPyyyy. COMPLETION STATUS OF CHILD SUPPLEMENT 
INCSASyyyy. DOES CHILD HAVE A CHILD SELF-ADMINISTERED SUPPLEMENT? (2002 and 2004 only; after 2004 the CSAS was part of the Child Supplement)
CSASCOMPyyyy. COMPLETION STATUS OF CHILD SELF-ADMINISTERED SUPPLEMENT
CSMSORDERyyyy. SEQUENCE OF CS & MS INTERVIEWS
INMSUPyyyy. DOES CHILD HAVE A MOTHER SUPPLEMENT?

2002-2016

CINTRVyyyy. INTERVIEW STATUS OF CHILD 
MSCOMPyyyy. COMPLETION STATUS OF MOTHER SUPPLEMENT 
MINTRVyyyy. INTERVIEW STATUS OF MOTHER

1994-present

YASAMPyyyy. IS CHILD ELIGIBLE FOR YOUNG ADULT INTERVIEW?
YAINTVyyyy. WAS CHILD INTERVIEWED AS YOUNG ADULT?

More details on mothers' interview status are provided by a series of variables titled "REASON FOR NONINTERVIEW" in the NLSY79 main Youth area of interest called COMMON VARIABLES.

Sampling Weights to Identify Interviews

Users can also employ the Child and Young Adult sampling weight variables to delineate their analysis sample and to identify respondents interviewed in each survey round. More information on the sampling weights can be found in the Sample Weights section.

The Child sampling weights are assigned to the ASSESSMENT and SAMPLING WEIGHTS areas of interest. Children who have been assessed or interviewed in a given year have values greater than 0 on their sampling weight for that year.

The Young Adult sampling weights for each year are assigned to the YA COMMON KEYVARS and SAMPLING WEIGHTS areas of interest. These YA sampling weight variables are specific to young adults interviewed in that year who were eligible to be interviewed, so that any young adult not eligible to be interviewed or any child who is not a young adult in that year is assigned a value of "0." With various sample restrictions in place over different rounds, occasionally a Young Adult who was not fielded in a given round is interviewed to ensure the participation of other family members. Such respondents will have an interview status flag indicating that they were interviewed but have a sampling weight of zero. For example, in 2012, 26 young adult respondents who were part of the older ages not being fielded were nonetheless interviewed. Therefore, researchers must use caution in their choice between using the interview status flag (YAINTVyyyy) versus the sampling weight (YAyyWEIGHT) to determine the cases interviewed in a given year.

Missing Data in the Geocode Files

Following the same convention as the NLSY79 Child and Young Adult public release data, missing data values on the geocode data files are coded -7 which indicates either a) a non-interview for a given year or b) respondents who have a missing value in the data for any variables from the County and City Data Book for the following reasons:

  1. Respondents who were in the military or who had an APO address
  2. Respondents who were residing outside of the United States
  3. Respondents whose state or county codes could not be determined
  4. Respondents who reside in a county or SMSA/MSA for which there is missing data for that geographic location from the County and City Data Book for that specific item
  5. Respondents who do not reside in an SMSA for any survey year 1994-2004 who will be missing SMSA level environmental variables for that year
  6. Respondents whose state, county, and ZIP codes for any survey year 1994-2004 do not lead to an unambiguous SMSA designation. This generally applies only to a small number of respondents living in New England.

In the 1994-2002 geocode data file, for 1988 and 1994 metropolitan statistical area variables with NECMA codes, respondents living in the New England states of Connecticut, Maine, Massachusetts, New Hampshire, Rhode Island, and Vermont were not treated any differently than those residing elsewhere. The addition of the "Record Type" variable in the 1988 and the 1994 County and City Data Book data files allows the user to designate these cases as missing and remove them from the analysis, without having to conduct a county-by-county or state-by-state determination of NECMA/non-NECMA status. Starting in 2004, these data are not available as part of the geocode data release. The Young Adult geocode files are described in more detail in Geographic Residence & Geocode Data.