Errata for NLSY97 Round 17 Release

Newest errata

Posted: 11/14/2019

Duplicate children and erroneous statuses assigned on CV_CHILD Roster

Some duplicate children were identified on the CV_CHILD roster in survey year 2015. These children have been deleted and other children moved up in position where necessary. In addition, a number of incorrect codes have been identified for the CV_CHILD_STATUS.## variables in survey years 2011, 2013 and 2015. These erroneous codes were the result of changes in questionnaire function for which adjustments were not fully made in the variable creation programs.

Other errata

Missing tax-advantaged account and credit card debt values for 2015 respondents

Posted: 5/22/2019

Calculations for the assets at age 30 and age 35 variables listed below inadvertently omitted values for tax-advantaged accounts for respondents reporting those assets in survey year 2015 (round 17). In addition, credit card balances for a small subset of respondents in survey year 2015 were not included in the debt or household net worth calculations at age 35. In connection with the missing credit card balances data, two lead-in questions were skipped in error for a small set of respondents. Variables affected for 2015 (round 17) are listed below.

Affected variables include:

Problem in pregnancy series for female respondents in Round 17 (2015)

Posted: 4/17/2018

In the round 17 data, due to an error in the preloaded data, the majority of female respondents answered a complete retrospective series on pregnancies, rather than on pregnancies since the date of last interview. This is the reason for the large frequency number in question YSAQ2-314 (2590 responses) and the low frequency number for question YSAQ2-314C (980 responses). This error also affects the pregnancy outcome frequencies in YSAQ2-326, YSAQ2-326D, and YSAQ2-326F.

Case data deletion

Posted: 4/9/2018

After the release of the 2013 data, the case data for respondent id 1061 were determined to be invalid for that survey year. The 2013 data for id 1061 have been removed from the round 17 release data, and the status of this case for survey year 2013 has been set to a non-interview.

NLSY97 Schooling Event History errata

Posted: 3/1/2018

Researchers comparing the latest round of the schooling event history variables with previous rounds of data will notice some historical data has changed. In general, some cases in the historical schooling data are modified each round since the NLSY97 first released the schooling event histories.

The histories get modified for a number of reasons. First, the overall schooling event histories are created by laying down an early round of information, then overlaying any changes on top. The program that creates the histories first puts down round 2’s information. Then the program layers round 3’s information on top. When this is done, round 4’s information is then layered over all the previous sets of data and the process repeats until the current round. Because respondents are allowed to provide information about schooling that is not constrained by the date of the last interview, some respondents provide changes and additions that modify earlier historic data that has been released to the public.

For example, a respondent could provide information that changes the values in the SCH_COLLEGE_TERM array between round 16 and 17. The SCH_COLLEGE_TERM array shows which schools and what semesters or terms the respondent is attending. This respondent in round 16 reported going to two different schools. For the second school they attended they reported going to six different semesters or terms. In round 17 they came back and reported different term dates for attending the second school. In the schooling event histories, newer information is assumed more complete and correct. Because they provided newer information, these new data would over-write the older information and changed the historical values.

Another reason why historical schooling data can change is because the NLSY97 does not ask about full- or part-time status until after a term is marked as completed. This means there are some respondents who start schooling but their full- or part-time status is unknown until later. These changes show up in the SCH_COLLEGE_DEGREE arrays.

For example, in round 16 the historical data for respondent could have a missing code (flagged as a 303) in 2001 to 2004. In round 17 the questionnaire determined that all the schooling in these years was done as part of a full-time program (coded as 103). The historical data from 2001 to 2004 would change from a code of 303 to 103 to reflect this new information, which was acquired in 2017.

Documentation correction to titles: "How Old When Incarcerated Relative Went to Jail"

Posted: 2/26/2018

The following NLSY97 variables were incorrectly titled. The correct title for each listed variable should be "HOW OLD WHEN INCARCERATED RELATIVE WENT TO JAIL." These corrections will be included in the next NLSY97 data release.

• T1066000 YHEA-2860.01 HOW OLD WAS R WHEN RELATIVE 01 DIED 2007
• T1066100 YHEA-2860.02 HOW OLD WAS R WHEN RELATIVE 02 DIED 2007
• T1066200 YHEA-2860.03 HOW OLD WAS R WHEN RELATIVE 03 DIED 2007
• T1066300 YHEA-2860.04 HOW OLD WAS R WHEN RELATIVE 04 DIED 2007
• T1066400 YHEA-2860.05 HOW OLD WAS R WHEN RELATIVE 05 DIED 2007
• T3159700 YHEA-2860.01 HOW OLD WAS R WHEN RELATIVE 01 DIED 2008
• T3159800 YHEA-2860.02 HOW OLD WAS R WHEN RELATIVE 02 DIED 2008
• T3159900 YHEA-2860.03 HOW OLD WAS R WHEN RELATIVE 03 DIED 2008
• T3160000 YHEA-2860.05 HOW OLD WAS R WHEN RELATIVE 05 DIED 2008
• T4577700 YHEA-2860.03 HOW OLD WAS R WHEN RELATIVE 03 DIED 2009
• T4577800 YHEA-2860.05 HOW OLD WAS R WHEN RELATIVE 05 DIED 2009
• T9108900 YHEA-2860.01 HOW OLD WAS R WHEN RELATIVE 01 DIED 2013
• T9109000 YHEA-2860.02 HOW OLD WAS R WHEN RELATIVE 02 DIED 2013
• T9109100 YHEA-2860.03 HOW OLD WAS R WHEN RELATIVE 03 DIED 2013
• T9109200 YHEA-2860.04 HOW OLD WAS R WHEN RELATIVE 04 DIED 2013
• T9109300 YHEA-2860.05 HOW OLD WAS R WHEN RELATIVE 05 DIED 2013

Marriage Event History Array data updated for small number of cases

Originally posted on 10/19/2017; updated with more cases on 12/12/2017

Due to inconsistences found between the marriage event history arrays, the following respondents and variables were affected. The updates will be included in the next release.

• PUBID=252:
  MAR_COHABITATION_2011.12 through MAR_COHABITATION_2014.03
• PUBID=440:
  MAR_STATUS_2004.01 through MAR_STATUS_2005.11 and MAR_PARTNER_LINK_2004.12 through MAR_PARTNER_LINK_2005.11
• PUBID=807:
  MAR_COHABITATION_2002.02 through MAR_COHABITATION_2010.06 and MAR_COHABITATION_2011.03 through MAR_COHABITATION_2013.12
• PUBID=1100:
  MAR_COHABITATION_2004.01 through MAR_COHABITATION_2011.06, MAR_COHABITATION_2011.08 through MAR_COHABITATION_2013.11, and MAR_PARTNER_LINK_2004.01 through MAR_PARTNER_LINK_2004.11
• PUBID=5811:
  MAR_COHABITATION_2004.02 through MAR_COHABITATION_2005.05
• PUBID=6437:
  MAR_COHABITATION_2011.10 through MAR_COHABITATION_2013.12

The following cases were added to the list on 12/12/2017:

• PUBID=1428:
  MAR_COHABITATION_2000.04 and MAR_COHABITATION_2000.05
• PUBID=2447:
  MAR_COHABITATION_2000.02 through MAR_COHABITATION_2000.09, MAR_COHABITATION_2003.10 through MAR_COHABITATION_2004.05, MAR_COHABITATION_2006.02 through MAR_COHABITATION_2015.11, and CVC_COHAB_TTL
• PUBID=2845:
  MAR_COHABITATION_2011.11 through MAR_COHABITATION_2014.01
• PUBID=3699:
  MAR_COHABITATION_2011.12 through MAR_COHABITATION_2012.05
• PUBID=4093:
  MAR_COHABITATION_2011.10 through MAR_COHABITATION_2014.06
• PUBID=4608:
  MAR_COHABITATION_2011.10 through MAR_COHABITATION_2013.11
• PUBID=5044:
  MAR_COHABITATION_2009.01 and MAR_PARTNER_LINK_2009.01
• PUBID=5467:
  MAR_STATUS_2002.08 through MAR_STATUS_2004.10, MAR_STATUS_2004.11 through MAR_STATUS_2006.06, MAR_COHABITATION_2002.08 through MAR_COHABITATION_2004.04, MAR_COHABITATION_2004.05 through MAR_COHABITATION_2004.09, MAR_COHABITATION_2004.11 through MAR_COHABITATION_2007.11, MAR_PARTNER_LINK_2002.08 through MAR_PARTNER_LINK_2004.04, MAR_PARTNER_LINK_2004.05 through MAR_PARTNER_LINK_2004.09, MAR_PARTNER_LINK_2004.10, MAR_PARTNER_LINK_2004.11 through MAR_PARTNER_LINK_2007.11, CVC_FIRST_MARRY_DATE~M, CVC_FIRST_MARRY_DATE~Y, and CVC_FIRST_MARRY_MONTH
• PUBID=6359:
  MAR_COHABITATION_2010.07 through MAR_COHABITATION_2010.10
• PUBID=7482:
  MAR_COHABITATION_2011.11 through MAR_COHABITATION_2012.11
• PUBID=8486:
  MAR_STATUS_2008.02 through MAR_STATUS_2009.01, MAR_STATUS_2009.02 through MAR_STATUS_2010.10, MAR_COHABITATION_2004.01 through MAR_COHABITATION_2006.11, MAR_COHABITATION_2008.02 through MAR_COHABITATION_2010.01, MAR_COHABITATION_2010.11 and MAR_COHABITATION_2010.12, MAR_PARTNER_LINK_2010.02 through MAR_PARTNER_LINK_2010.12, CVC_FIRST_MARRY_DATE~M, CVC_FIRST_MARRY_DATE~Y, and CVC_FIRST_MARRY_MONTH
• PUBID=3521:
  MAR_COHABITATION_2011.10 through MAR_COHABITATION_2013.12, MAR_PARTNER_LINK_2011.10 through MAR_PARTNER_LINK_2013.12
• PUBID=1587:
  MAR_COHABITATION_2004.10 through MAR_COHABITATION_2005.01, MAR_COHABITATION_2013.06 through MAR_COHABITATION_2014.10, and CVC_COHAB_TTL
• PUBID=2001:
  MAR_COHABITATION_2000.01 through MAR_COHABITATION_2002.05, MAR_COHABITATION_2004.01 through MAR_COHABITATION_2007.06, and CVC_COHAB_TTL
• PUBID=1566:
  MAR_COHABITATION_2004.02 through MAR_COHABITATION_2005.01, MAR_COHABITATION_2005.04 through MAR_COHABITATION_2007.05, and CVC_COHAB_TTL
• PUBID=1935:
  MAR_COHABITATION_2000.01 through MAR_COHABITATION_2000.12
• PUBID=2516:
  MAR_COHABITATION_2004.01 through MAR_COHABITATION_2004.09, MAR_COHABITATION_2007.03 through MAR_COHABITATION_2008.01, MAR_COHABITATION_2008.06 through MAR_COHABITATION_2010.03, and CVC_COHAB_TTL
• PUBID=871:
  MAR_STATUS_2002.04 through MAR_STATUS_2005.03 and MAR_COHABITATION_2002.04 through MAR_COHABITATION_2005.03

College major code variables missing in latest NLSY97 data release

Posted: posted 10/31/2017

The NLSY97 Round 17 college major codes were inadvertently omitted from the most recent public data release. The variables will be included in all future public releases of the NLSY97 data.