Interview Methods

• Instrument Development
• Interviewer Training
• Mode of Administration
• Language of Administration

Instrument Development

The Child Survey. The original proposal for the NLSY79 Child data collection included plans for instruments that tapped several dimensions of child cognition, health, socio-emotional attributes, behavior, and home environment. Development of the first round of NLSY79 Child Assessment instruments began in the summer of 1985. CHRR assumes overall responsibility for selection, design, and adaptation of the Child Assessments. Field training and data collection were the primary responsibility of NORC. Close collaboration between NORC and CHRR began prior to each fielding period on such issues as placement and formatting of questions, survey timing, and special data collection considerations such as confidentiality, interviewer training, sample location, and testing conditions. On a continuing basis, NICHD also provided input to this process. Advice on question inclusion as well as review of each draft survey instrument was sought by CHRR from the various funding agencies, notably the Bureau of Labor Statistics, which has overall authority regarding survey content, NICHD, a technical advisory board, and designers of the original questions and scales.

In the first several survey rounds, pretests of the Child instruments were conducted several months prior to the beginning of the fielding period. Conducted at two sites with experienced NLSY79 interviewers, the pretest served to test questionnaire wording, to pinpoint items that may pose problems for the respondents or the interviewers, and to time the various sections of the instruments. This process also served to identify problems with questions, skip patterns, transitions between sections, questionnaire length, and other overt flaws. Following each pretest, staff from CHRR, NORC personnel, and representatives of the various funding agencies met to review the NLSY79 child survey instruments, analyze the response frequencies for selected questions, and discuss problems encountered by both the respondents and interviewers.

The Young Adult Survey. When the Young Adult survey was first designed for the 1994 survey round, many of the CAPI sections exactly paralleled those administered to the main NLSY79 respondents. Other sections of the questionnaire were tailored for this cohort. One important part of the design process was to review the 1979 NLSY79 questionnaire, to consider where Young Adults could be asked questions that were essentially the same as those that had been asked of their mothers in their first survey round. Other years of the NLSY79 were also reviewed for questions to include. The Young Adult questionnaire remained fairly stable as an instrument through the 1998 fielding, with changes in parallel sections mirroring those in the NLSY79 Youth.

The Young Adult survey instrument underwent a major redesign in 2000 and differs in a variety of important ways not only from the main Youth questionnaire, but also from the previous Young Adult instruments. The questionnaire was streamlined and adjusted for telephone administration, so that most interviews could be undertaken in less than one hour. Additionally, more pre-existing information was incorporated into information sheets to determine branching for each respondent's path through the questionnaire. Branching also occurred throughout the questionnaire based on answers provided by the respondent. Many of the items that are comparable across the main Youth and the Young Adult were retained. Although not an exhaustive listing of these items, Table 1 in Life Cycle Profiles for the NLSY79 Children provides users with a listing of attitudinal and behavioral sequences in the Young Adult and indicates where comparable data can be found for the mothers. 

The 2002 survey instrument remained very similar to the 2000 questionnaire while incorporating questions directed to the younger Young Adults that paralleled questions they had been asked in the Child Survey when they were 10 to 14 years old. Questions concerning weapons in school were also added. 

Prior to the 2004 fielding, we assessed the viability of changes made in 2000 and 2002 and questioned if there were additional needs. The redesign for 2004 focused on improving data collected on fertility and relationships, which have become increasingly important as this cohort ages, as well as enlarging the scope of health-related data.

For the 2006 fielding, we again assessed the changes we had made in the questionnaire for 2004 as well as incorporating additional questions to expand the range of data we collect. New questions were included to ascertain biological relatedness among the young adult's siblings in the Child database. We also added questions about catastrophic events, assets and debts, and the Ten-Item Personality Inventory (TIPI). The 2006 questionnaire also included a series of political questions, funded through the American National Election Survey (ANES) with a grant from the National Institute of Health. 

Prior to the 2008 fielding, the Young Adult survey instrument once again underwent a major re-evaluation and redesign. This redesign focused on social psychological issues, job characteristics, and military service, as well as on parenting for both residential and nonresidential children, maternity leave, and relationships. 

The redesign for the YA2010 survey round included bringing forward information on household members as well as current jobs from the date of last interview, as well as an expansion of questions concerning financial difficulties, perceived fairness in relationships, and gender role items. Questions about the height and weight of biological children were added. Additionally, a health module for older Young Adults was started in 2010. For 2010 and 2012, this module will be asked of all respondents age 29 or older.

The redesign for the YA2012 focused on evaluating the HOME items that have been used in the Young Adult and included adding additional HOME items as well as the Behavioral Problems Index, also from the Mother Supplement. Other additions in 2012 included a series of questions about day-to-day discrimination and another about major instances of discrimination. The mini-IPIP (a shortened version of the International Personality Item Pool) was also included, providing users the opportunity to compare it to the TIPI. Another series on childhood adversity, which has also been asked of the NLSY79 respondents, was added.

As part of the redesign of the YA2014 questionnaire, the answer categories for the questions concerning grade currently attending/highest grade attended and highest grade completed were changed. This change was made to alleviate a long-standing uncertainty on the part of both respondents and interviewers about how to code years of college, especially when a respondent is either taking classes part-time and/or has transferred universities. Additionally, the migration sequence in the YA was made parallel to the migration sequence in the NLSY79 questionnaire. A new series of questions on self-employment, modeled after those in the NLSY79, was introduced into the Employer Supplements. The expanded heath section was redesigned to incorporate additional questions for respondents who are over 40 years of age. Respondents who are 29/30 or 41/42 as of December 31st of the target year will enter this extended health module, as will respondents who were not interviewed at those age points.

Much of the redesign of the YA2016 questionnaire focused on reducing questionnaire time. Some questions, such as the detailed questions about indebtedness and foreclosure, were eliminated completely while others, such as the questions about volunteerism, had the amount of verbiage reduced. For some other questions, such as the household division of labor questions, are now being asked of a more restricted universe. Previously these had been asked to all YAs with a current spouse/partner; beginning in 2016, these will be asked of YAs with a new spouse/partner, YAs with the same spouse/partner but now with a first child, and at specific ages (25, 35, 45, etc.) if they still have the same spouse/partner. Throughout the questionnaire, 'other-specify' answers were reviewed and pick lists were adjusted based on this review. The category of “spouse” in the list of relationships of household members was spilt into “husband” and “wife,” and the wording of questions about dating and cohabiting were made more gender-neutral. New questions were added about the use of e-cigarettes, synthetic marijuana, and narcotics/opioids.

The YA2016 questionnaire was also adjusted for the inclusion of 12-14 year olds who in past rounds would have had the Child Supplement and Child Self-Administered Supplement administered to them. While 14 year olds, most of whom would have turned 15 by their interview date, would follow the same path as 15 to 16 year olds, several path modifications were made to branch 12 and 13 year olds around questions that were inapplicable. 12 and 13 year olds were also not asked some of our questions for first-time YAs but will be asked those in the next round, so that these questions are asked of comparable age groups across time. Additionally, certain items from the CS/CSAS, such as the Self-Perception Profile for Children (SPPC) and the moods scale, were included in the YA2016 questionnaire solely for this age group.

The YA2018 survey was streamlined, with some questions about volunteerism, drug use, asthma, self-employment, high school experiences, maternal employment, child support and co-parenting either reduced or curtailed. This redesign allowed the introduction of new questions without increasing survey time.

The YA2018 questionnaire included an expanded number of questions about loneliness and social isolation, as well questions about sexual orientation and gender identity. The YA2018 also included the General Anxiety Disorder Scale (GAD-7). Developed by Spitzer, Kroenke, Williams and Lowe, the GAD-7 asks respondents how often they have experienced certain feelings associated with anxiety in the past two weeks. The survey also included questions about the health and well-being of their parents as perceived by the Young Adult respondents.

The discrimination questions, which include major and day-to-day discrimination, first asked in the YA2014 questionnaire, were re-asked of all YAs in 2018 to allow researchers to look for changes over time as these respondents age.

The initial redesign of YA2020 questionnaire included the addition of the six-item Brief Resilience Scale asked of everyone and a series of questions about perimenopause and menopause asked of women over age 30. The advent of the Coronavirus pandemic, however, led to the development and inclusion of a variety of questions to assess the impact of the pandemic on the lives of the YA respondents. Questions were added to assess this impact on employment, income and assets, health, family, and children’s schooling. Additionally, all respondents were re-asked the General Anxiety Disorder Scale (GAD-7), so that there is a pre-Covid and post-Covid anxiety measure for respondents interviewed in both 2018 and 2020.

Interviewer Training

NLSY79 Child interviewers underwent a special training in preparation for administering the Child interviews and assessments. Their curriculum included at-home self-study, in person training (through 2010), and both written and observer evaluation. In some survey years this training was held at decentralized sites, but in later rounds training was been conducted at a centralized location, with materials distributed ahead of time. Detailed instructions on the administration of each assessment were reviewed in each training session, regardless of the experience of the interviewer. Videotaped footage was used to demonstrate the procedures and to allow interviewers to view the actual administration of the assessments to children of different ages. Trainees were given exercises to help them master the testing procedures. They participated in a series of scripted mock interviews, which were observed by supervisors, and then completed a final examination that was scored. Interviewers also recorded their initial child interviews, which were then reviewed by field management staff before the interviewer was allowed to complete a full case load. Details of the NLSY79 Child training in the early survey rounds can be found in the NLSY79 Child 1986 and 1990 handbooks (available on the Research/Technical Reports page).

NLSY79 Youth and Young Adult interviewers undergo basic survey and case management training as well as instrument-specific training on the two instruments.  Their curriculum includes at-home self-study, remote training, and both written and observer evaluation.

Mode of Administration

From 1986-1992, interviews with the NLSY79 children were conducted primarily in person using paper and pencil questionnaires. Beginning in 1994, the primary Young Adult and younger child instruments and assessments were administered using computer assisted personal interviewing (CAPI). By 2000, all survey instruments except the Child Self-Administered Supplement were fully computerized; the CSAS was computerized in 2002. Researchers are encouraged to review the questionnaires in conjunction with this user's guide and the actual data.

Child Interview Mode. Interviews with the NLSY79 younger children were typically conducted in the respondent's home by specially trained field staff. Reports were obtained from the children, their mothers and by interviewers trained to assess each child and to provide evaluations of each family's home environment. All interviewer-administered sections used CAPI; for the Child Self-Administered Supplement, the child used the laptop to complete the section. [The CSAS was also available on personal data assistant (PDA) in 2002 and 2004.] In rare instances, where in-person administration was not possible, as much information as possible was collected over the telephone. Table 1 outlines the interview mode and type used in each survey round. Note: Child Supplement and Child Self-Administered Supplement were not administered after 2014 survey year.

Child interview mode variables. While there is no single constructed variable in each Child survey year that indicates the mode of the interview, there are items that indicate the type of administration for each field instrument (see Table 2).

Prior to 1992, the Child Supplement (CS) was not administered by telephone. Starting in 1992, "telephone interview" was one of the response categories from which interviewers could choose to describe the mode and location of the Child Supplement interview, as outlined for each survey year in Table 3.

*Child Supplement was not administered after 2014 survey year.

The items that indicate the mode of administration used for the Child Self-Administered Supplement (CSAS) are listed in Table 4.

 *Child Self-Administered Supplement was not administered after 2014.

Young Adult Interview Mode. From 1994 to 1998, the primary mode of data collection for the Young Adults was an in-person interview, with the interviewer using a laptop computer for most questions and the respondent filling out a paper booklet for the self-report section. Beginning in 2000, the primary interview mode for the Young Adults shifted to telephone rather than in-home visits. The items in Table 5 indicate the mode of administration of the young adult questionnaire in each survey year.

Language of Administration

Spanish translations of several child assessment instruments were made available to respondents with limited proficiency in English. In 1986, a total of 354 children age eight months or older were assigned to bilingual interviewers. Of these cases, slightly more than 100 children were actually assessed in Spanish. More than 100 children were assigned to bilingual interviewers in 1988. By 1990, 52 children were assigned to bilingual interviewers, but of this number, only 17 were actually assessed in Spanish. In 1998 approximately 50 children were interviewed in Spanish but most of them were assessed in English. In 2000 the number of children assessed in Spanish declined to fewer than 10. Part of this recent decline results from the higher minimum age of children eligible to be assessed. By the 2000 survey round most of the Spanish-language parents would have resided in the U.S. for more than two decades.

From 1986 to 2000, the following questionnaire sections and child assessment instruments were translated for administration into Spanish:

1. Mother Supplement and MS assessments:
   • The HOME
   • Behavior Problems Index (BPI)
   • Temperament or What My Child is Like
   • Motor and Social Development (MSD)
2. Mother-report sections of the Child Supplement (background, school, health)
3. Interviewer administered assessments in the Child Supplement:
   • Parts of the Body
   • Memory for Location
   • What I am Like (SPPC)
   • Memory for Digit Span
   • PPVT-R (Test de Vocabulario en Imagenes Peabody or TVIP used in 1988-2000)

Starting in 2002 no Child instruments were translated into Spanish, although bilingual interviewers continued to be assigned to households in which Spanish was a principal language. Interviewers were instructed to make comments in the assessments or interviewer remarks section if other languages were used in the interview to facilitate understanding.