The Young Adult Survey:

1
The Young Adult Survey

• Following the Children of the NLSY79 Mothers into
  the Future

2
Beginning of the YA Survey

• Began in 1994
• Designed to follow the NLSY79 children as they
  approached adulthood
• Children age up when they will turn 15 by the
  end of the current survey year
• Age cap of 20 at interview in 1998
• Portion of over-sample cases not fielded in 2000
• No age or sample restrictions since 2002

3
Survey Content

• Questions from the earliest years of the Youth
• Knowledge of the World of Work
• Womens Roles
• Ethnic Identification

4
Survey Content

• Questions parallel to the current Youth Survey
• Household Record
• CPS, Employer Supplements (some
• questions eliminated), Military Service and
  Gaps
• Fertility
• Income (less assets)
• Training
• Attitudes such as Perlin Mastery, Rosenberg
  Self-Esteem, and CESD

5
Survey Content

• Continued or Parallel questions from the CS/CSAS
• Health (routine checks, accidents and injuries)
• Dating
• Relationship with parents
• Some parent/child interaction
• Risk Taking
• School Experience and Satisfaction
• Computer Use

6
Survey Content

• Tailored questions for this age group (1)
• Information about fathers, especially
  non-residential ones
• Timing of and Reasons for leaving home
• Contact with parents as YAs leave home
• Marriage and Cohabitation History
• High School and College experiences

7
Survey Content

• Tailored questions for this age group (2)
• Transition into the world of work
• Current childcare usage and cost
• Greater detail on non-normative activity such as
  substance use and criminal behavior
• Greater detail on sexual activity

8
Survey Content

• Major redesign for the 2000 survey (1)
• Switch to telephone as primary mode
• CPS and Gaps eliminated
• Some sections shortened (Employer Supplement,
  fertility, training)
• Questions with complex answer choices broken down
  into multiple questions
• Key questions from Self-Report Booklet converted
  into CATI/CAPI instrument

9
Survey Content

• Major redesign for the 2000 survey (2)
• New areas added
• Closeness to Parents
• Greater migration sequence
• Movement into and out of parents home
• Ever homeless
• Monetary contributions to noncoresidentional
  children added to fertility
• Parenting Attitudes added to fertility

10
Survey Content

• Major redesign for the 2000 survey (3)
• New areas added
• Modified HOME questions asked about youngest
  child of respondent in the household
• Total family income and financial strain
• Addiction to and workplace use of marijuana
• Amphetamine usage as a separate series for the
  first time

11
Survey Content

• Redesign for the 2002 survey
• New questions in Military to find out
  whether/when reservists had been called into
  active duty
• Questions about part-time, irregular employment
  added for younger YAs parallel to those asked in
  CSAS

12
Survey Content

• Redesign for the 2004 survey (1)
• Household Record redesigned to reduce interviewer
  error
• Marriage history of spouse/partners asked
• Children from other relationships of
  spouse/partners asked
• Monetary contributions of nonresidential parents
  asked for each biological child

13
Survey Content

• Redesign for the 2004 survey (2)
• Specific Child Care arrangements dropped
• Home Items revised and asked of all children in
  the HH tied to the YA
• Questions about household division of labor and
  economic enmeshment added for YAs with
  spouse/partners
• Questions about serious but not cohabiting
  relationships added

14
Survey Content

• Redesign for the 2004 survey (3)
• Asthma series added
• Healthy behavior series added
• Gender role items from CSAS added for 14-16 year
  old YAs

15
Survey Content

• Redesign for the 2006 survey (1)
• Asthma series tailored to update appropriately
• Catastrophic events series added
• Questions on assets and debts expanded
• Series on financial help with living expenses
  added

16
Survey Content

• Redesign for the 2006 survey (2)
• Ten-Item Personality Inventory added
• Family conflict series added
• Series of political questions added
• New series to ascertain biological relatedness
  among YA and siblings

17
YA 2006 Preliminary Release

• Contains an overview with links into
  questionnaire sections
• Has an HTML version of the 2006 questionnaire
  that is hyper-linked within and across sections
• Has raw, unedited data for all interviewed YAs,
  along with a small number of constructed
  variables
• Can be merged with either the Child/YA 2004
  release or the NLSY79 2004 release

18
Navigating in the YA Survey

• Understanding survey flow and universes

19
YA 2006 Survey Flow Chart

• The following screens show a flow chart of the YA
  2006 survey
• Use this flow chart to
• Understand the general paths through the survey
• Understand the universes for survey sections (but
  not individual items)
• For greater detail, refer to the HTML survey

20
(No Transcript)
21
(No Transcript)
22
(No Transcript)
23
(No Transcript)
24
(No Transcript)
25
(No Transcript)
26
Survey Flow for Individual Items

• The path a respondent takes is determined by
  three major sources of information
• Preloaded information from past survey rounds
  and/or mothers data
• Answers to specific questions in the survey
• Values constructed from preloads, answers or a
  combination thereof

27

• The path a typical respondent might take can be
  explored through following the hyperlinks in the
  HTML questionnaire
• Decide on an age and characteristics (living with
  mom, in college, etc).
• Click on links appropriate to these
  characteristics and answers you choose
• Path can be retraced backwards by using Lead-In
  links

28

• Differences between the HTML questionnaire and
  the codebook
• Codebook is also hyperlinked and can be used to
  trace survey flow however,
• The HTML questionnaire includes questions that
  are not released, and
• The codebook includes constructed variables that
  will not be in the questionnaire

29

• Examples of questions in questionnaire but not
  released
• Locating or other private data
• Verbatims, which may be coded and released as
  numeric variables such as occupation and industry
• Questions that are used as inputs for summary
  rosters, such as in the household record or the
  biological child roster these data are released
  in their summary form

30

• Examples of items in codebook but not in
  questionnaire
• Coded items such as occupation and industry
• Constructed variables such as age at first birth
  or month and year of fist marriage
• Cleaned versions of existing variables such as
  residence

31

• Caveats for understanding survey flow and
  universes (1)
• Some questions are asked only the first time a
  respondent goes through the survey
• For example, fathers race is asked only once,
  during the first YA interview
• Usually these questions are directed at 14-16
  year olds, but occasionally older YAs are
  interviewed for their first time
• Some items may be deliberately restricted to new
  YAs under a certain age e.g., questions about
  teenage drinking patterns (see YASR-10 for skip
  into these)
• Machine checks prior to substantive questions
  allow users to understand universes

32

• Caveats for understanding survey flow and
  universes (2)
• Some questions are asked only if a positive
  response has not been given in the past
• For example, there are preloaded flags for ever
  dated, ever been pregnant, ever smoked, ever
  drank, etc.
• Machine checks prior to substantive questions
  allow users to check universes (e.g., YASR-47)
• Occasionally, either respondents change their
  minds or flags reflect previous interviewer
  error, so flags may be inconsistent over survey
  rounds (see YASR-49 for example of an escape
  clause)
• For self-report items, survey checks were not
  introduced until the 2000 redesign

33

• Caveats for understanding survey flow and
  universes (3)
• Some questions have a substantive question needed
  to understand universes
• These substantive questions either allow for a
  nonnumeric response or branch according to the
  type of response
• For example, Q2-4B1 (When did you most recently
  begin living in this city/town?) offers the
  answer choices Select to enter date and Lived
  here all my life
• Select to enter date responses are branched to
  Q2-4B to enter the date, then into the rest of
  the migration sequence
• Lived here all my life responses are branched
  out of the migration questions

34

• Caveats for understanding survey flow and
  universes (4)
• Some questions have both a machine check and a
  substantive question needed to understand
  universes
• For example, Q3-0E is a machine check to see if R
  has reported dating in the past, and codes of
  zero branch to Q3-1
• Q3-1 asks age R began dating and offers the
  choices of Select to enter age, Never
  started, and Volunteered same-sex dating
• Select to enter age and Volunteered same-sex
  dating are branched to Q3-1A to enter age
• Never started responses are branched to the
  next machine check, Q3-1C

35

• Caveats for understanding survey flow and
  universes (5)
• Some questions are only updated since the date of
  last interview for previously interviewed YAs but
  ever for new YAs
• Generally there are not machine checks available
  prior to these questions, as text fills determine
  the question text
• The question text in the HTML questionnaire and
  codebook alerts users to these questions
• Bracketed text such as Since DLI have you/Have
  you ever will appear in such questions (see
  Q3-4)
• Machine checks that everyone hits, such as Q2-1
  (Y1451900 in 2006), allow users to identify new
  vs. old YAs

36
Understanding Missing Data

• Although data extracts with the SAS or SPSS
  statements option automatically include language
  to set negative values to special system missing
  values, users should be aware of what the various
  missing codes in our data mean so that they can
  better understand the universes that are missing
  from various data points

37

• Codes of -7 in the Young Adult Data mean a valid
  skip, indicating that
• The respondent may not have been interviewed in
  that round, or
• The respondent may not have been eligible for a
  given instrument, or
• The respondent may not have entered a given
  section of the survey, or
• The respondent may have been skipped out of a
  specific path

38

• Codes of -7 can be differentiated by
• Using the eligibility and interview status flags
  for the YA in the Child Background area of
  interest
• Using machine checks within each survey
  instruments to determine survey paths

39

• Codes of -3 in the Young Adult Data mean an
  invalid skip, indicating that
• The respondent did not answer questions that
  should have been answered (YASRB in 1996 and
  1998), or
• The one or more of the inputs needed to construct
  a variable are missing, or
• There was a problem with preload data or the
  system during an interview
• Users should note that in the CAPI/CATI data an
  invalid skip will show up in the machine checks
  that branch the respondents into questions, not
  in the questions themselves

40

• Codes of -2 and -1 in the Young Adult Data are
  respondent answers
• Codes of -2 indicates a response of Dont Know
• Codes of -1 indicates a refusal to answer

41
(No Transcript)
42
(No Transcript)
43
(No Transcript)
44
(No Transcript)
45
(No Transcript)
46
Cross-Generational Research Possibilities
Later Childhood
Early and Middle Childhood
Adulthood
Maternal Background Experience
47
Potential areas to explore

• Family Transitions
• Parenting Attitudes and Behavior
• Educational Attainment
• School to Work Transition
• Labor Force Participation
• Occupational/Status Attainment
• Transmission of Attitudes and Beliefs
• Deviance across generations

48
(No Transcript)
49
Content Crosswalks

• The following tables help you
• Examine some of the attitudinal and behavioral
  items asked of the mothers, the children and the
  Young Adults
• See the frequency with which some of these
  questions have been asked
• Locate key variables in the Young Adult data files

50
(No Transcript)
51
(No Transcript)
52
(No Transcript)
53
(No Transcript)
54
The constructed Fertility and Relationship
Variables also includes the cleaned date of
birth, gender, usual residence for each survey
round, and date of death (if applicable) for all
children born to YA respondents.
55
Sample Size

• The following tables show you
• The number of Mothers, Children and Young Adults
  interviewed through 2004
• The number of survey points available for all YA
  respondents across years
• Potential sample sizes for pooled age groups
• Details about YA respondents interviewed in 2006

56
(No Transcript)
57
(No Transcript)
58
(No Transcript)
59
(No Transcript)
60
(No Transcript)
61
(No Transcript)
62
(No Transcript)
63
(No Transcript)
64
(No Transcript)