Created variables
- CV_ENROLLSTAT. Informs the user about whether the respondent is enrolled; whether enrolled respondents are in grades K-12 or in a 2- or 4-year college or graduate program; and what level of degree or diploma (or GED certificate) respondents have obtained.
- CV_SCHOOL_TYPE. Lists whether the school currently or most recently attended by the respondent (if primary or secondary) is public, private, parochial, or of some other type.
- CVC_GRADES_REPEAT_EVER and CVC_GRADES_REPEAT_YR. The total number of grades, through high school, that the youth respondent has ever repeated as of the survey date; and the total number of grades that the respondent has ever repeated as of June 30 of the survey year. Created for all respondents regardless of interview status.
- CVC_GRADE_SKIPPED_EVER The total number of grades, through high school, that the youth respondent has ever skipped. Created for all respondents regardless of interview status.
- CVC_HGC_EVER_EDT. Highest grade completed by respondent, created for all respondents regardless of the round's interview status.
- CVC_HIGHEST_DEGREE_EVER_EDT. Highest degree completed by respondent, created for all respondents regardless of the round's interview status.
- CV_HIGHEST_DEGREE_1314 (for round 16). Highest degree received prior to the start of the 2013/2014 academic school year. Similar variables available for earlier rounds (suffix denotes the year).
- CV_HGC_1314 (for round 16). Highest grade completed prior to the start of the 2013/2014 academic school year. Similar variables available for earlier rounds (suffix denotes the year).
- CVC_GED. Using a continuous month format (which labels January 1980 as month 1, February 1980 as month 2, and so on), this lists the month GED was received. This is a cumulative variable created for all respondents regardless of interview status for the round.
- CVC_HS_DIPLOMA. Using a continuous month format (which labels January 1980 as month 1, February 1980 as month 2, and so on), lists the month the high school diploma was received. This is a cumulative variable created for all respondents regardless of interview status for the round. Similar CVC variables include AA_DEGREE, BA_DEGREE, MA_DEGREE, PROF_DEGREE, And PHD_DEGREE.
- CVC_HS_LEFT_DATE. Date respondent left high school, using a continuous month format (January 1980 is month 1, February 1980 is month 2, etc.). The variable is created for all respondents regardless of interview status for a particular round.
- CVC_HS_LEFT_HGC. Highest grade completed at the time the respondent left high school. The variable is created for all respondents regardless of interview status for a particular round.
- CVC_SCH_ATTEND_EVER. Number of regular elementary/middle/high schools ever attended. This variable is created for all respondents regardless of interview status for the round.
Important information: Grade progression
There are a number of apparent inconsistencies in the raw survey data with respect to grade progression. Through a data quality review after round 6, survey staff determined that the complexity of the survey questions, coupled with problems in the way the data were interpreted during the programming of the event history arrays, led to a significant number of spurious repeated and skipped grades. For example, because of errors in reporting or programming, it may appear that a respondent completed 10th grade twice and then jumped ahead to 12th grade when in fact the respondent had a normal progression through the grades. The following paragraphs detail the six main problems found in the data and the steps taken to correct them.
- Survey staff reviewed the grade reported in the initial 1997 survey and the date of high school graduation. While the detailed school enrollment loops ask for information that individuals may not always report correctly, the date of graduation from high school is a salient event that respondents should report correctly with a high degree of accuracy. Using this information, survey staff identified all respondents who moved from the grade reported in 1997 to high school graduation in the expected amount of time. If a respondent's graduation date indicates that the respondent should have a normal school progression--completed one grade per school year--the event history program flagged the respondent and imposed a normal progression on the event history variables.
- A number of respondents enroll in college courses while they are still in high school. Event history arrays only contain a single grade attended for a given time period, and the original event history program was written so that college courses were given precedence over high school. For example, if an 11th-grader also took a freshman-level college class during first semester, the program assigned a grade of "13" (first year in college) for that semester. If the student then finished 11th grade but did not take any college classes during second semester, it would appear in the data that the student jumped ahead to year 13 of schooling and then back to 11th grade during the course of a single year. This resulted in a number of extra promotions and regressions. Consequently, the event history program has been rewritten to prioritize high school over college, removing these spurious grade changes.
- Some respondents provided a high school graduation date but then reported additional secondary school enrollment after that date. Survey staff decided to exclude post-graduation secondary school enrollment from the event histories, although this information is preserved in the raw data for researchers who might be interested in the additional training received by respondents after graduation.
- While answering the schooling questions, some respondents reported initial enrollment at a school but apparently did not understand that they should report each grade attended at that school in a separate loop within the schooling section. This resulted in some respondents appearing to remain in one grade for a long period of time, particularly if they had missed one or more interviews, and then apparently jumping ahead several grades. If, for example, a respondent appeared to be in 9th grade for 3 years and then jump ahead to 12th grade, the most likely reason is that he or she did not understand the schooling questions and actually did progress normally through 10th and 11th grade. The event history program now flags these respondents and adjusts their schooling history to follow a normal grade progression.
- In a number of cases, respondents appear to jump backward and then forward across multiple grades. For example, some respondents were listed as attending 9th grade, then 1st grade, then 11th grade. The most likely explanation for this pattern is a data entry error where the interviewer accidentally dropped the zero from 10th grade. Jumps in a normal school progression which appear to be caused by a missing digit in a two-digit grade were corrected.
- Finally, data review of individual cases indicates that, when asked what grade they had first attended at a given school, some respondents reported instead the first grade offered at that school. As with the problem in the previous paragraph, this causes respondents to appear to jump backwards across a number of grades and then jump forward again the next year. Hand edits were made to adjust the event histories for these respondents to a normal grade progression. The six changes described above significantly reduced the number of abnormal grade progressions found in the event history SCH_GRADE_PROGRESS variables. About 3/4 of the promotions and demotions found in the raw survey data for rounds 1-6 appear to be the result of reporting or programming errors. After the corrections were implemented, about 100 demotions and 570 promotions remained. Although it is possible that errors remain, based on inspection of the data survey staff feel that the vast majority of these grade changes reflect actual atypical progressions.
Event History variables
The schooling event history variables comprise three distinct groups of information designed to summarize and simplify the raw data NLSY97 respondents provide on their schooling experience. The first group of event history variables provides yearly summaries of the respondents' experience during primary and high school. The second set of variables provides a month-by-month view of primary school and high school experience. The third set provides a month-by-month view of any college and graduate school experience. The event history variables are created using a complex set of programs that draws information from the edited schooling rosters, not-edited questions, the parent questionnaire, created education variables, and special procedures that handle cases with inconsistent data. Here is a summary of the three types of event history variables:
Yearly schooling variables
The yearly schooling variables provide a summary of the respondents' experiences during primary school and high school. The yearly information tracks each respondent's educational experiences beginning in the mid 1980s, when the first information is available in the survey, through the round 11 interview. In general, these variables refer to the school year rather than the calendar year. Because almost all respondents have completed their primary and high school education by round 11, the information in these arrays stops in 2007.
- SCH_YEAR_TO_GRADE, SCH_GRADE_TO_YEAR. These two sets of variables track the grade the respondent attended during the school year and the school year during which the respondent attended a certain grade. For example, SCH_YEAR_TO_GRADE_1990 refers to the grade attended by the respondent during the school year that starts in fall 1990 and ends in spring 1991. Similarly, if the respondent attended fourth grade in 1992-93, then SCH_GRADE_TO_YEAR.04 would have the value 1992.
- SCH_CHANGES. This variable counts the number of times that the school the respondent attended changed during the school year. For example, SCH_CHANGES_1990 shows how many different schools the respondent attended during the school year that started in fall 1990 and ended in spring 1991.
- SCH_MNTHS_MISSED. This variable tracks the number of months during the school year that the respondent did not attend school. This variable does not include summer vacation.
- SCH_SUMMER_SCHOOL. This variable shows whether the respondent attended extra school classes, such as summer school, during an educational break in a given school year.
- SCH_SUSPENSIONS. This variable counts the number of days during the school year the respondent was suspended from school. For example, if SCH_SUSPENSIONS_1990 has a value of 3, then the respondent was suspended from school for three days during the 1990-91 school year.
- SCH_GRADE_PROGRESS, SCH_YEAR_PROGRESS. This variable reports whether the respondent was skipped ahead or demoted during a given grade in school or during a given school year.
College monthly schooling variables
These monthly variables report the respondent's college experience starting from 1997 through the current interview date.
- SCH_COLLEGE_STATUS. This variable reports the respondent's enrollment status during each month from the round 1 interview date through the current interview date. Coding categories include not enrolled in college, enrolled in a 2-year college or enrolled in a 4-year college.
- SCH_COLLEGE_TERM. This is a two-part variable. The last two digits refer to the term number in college. The term number is a value from 1 to 99. The first two digits show the type of school: public, private, religious or unknown.
- SCH_COLLEGE_ID. This variable shows the ID of the school the respondent attended. It permits users to link array information from the college schooling roster in the main data file to other information about the college.
- SCH_COLLEGE_DEGREE. This variable tells researchers if the respondent was going to school full time or part time. It also tells researchers the type of degree the respondent is attempting to complete. Degrees include a two-year degree, four-year degree, master's degree, doctoral degree, professional degree, and a joint BA/MA.
Primary and High School Monthly Schooling Variables
These monthly variables report the respondent's primary and high school experience.
- SCH_STATUS. This variable reports the respondent's enrollment status during each month from the round 1 interview date through round 13. Coding categories include unknown, not enrolled, in grades K-12, on vacation, expelled, and other.
- SCH_TERM. This variable reports the respondent's school type (public, private, or religious) and grade for each month in the time period.
- SCH_ID. This variable permits users to link array information to the school roster in the main data file and access other information about the school. For each month that the respondent was enrolled in the SCH_STATUS array, the corresponding monthly variable in this array contains a school identification code. Note that this ID does not provide users the ability to determine if multiple respondents were in the same school.
- SCH_DUA. This variable flags the small number of NLSY97 respondents who went to two different schools in the same month. Because only the first school can be reported in the other arrays, this variable flags these special cases. There is only one variable for each school for the period between each interview; the exact month when the overlap occurred is not indicated, and overlap may have occurred in more than one month.
- Deny Variables. Deny variables in the schooling section identify respondents who deny ever attending a school reported in a previous interview round.
Overview of how Schooling Event History variables are created
The schooling event histories are created in a multi-step process: First, grade school and college data are extracted separately for each round. Second, separate programs create a simple event history for just that round. One monthly event history for each round is created for the respondent's grade school experience and one event history for just that round is created for college experience. The third step combines all the individual rounds of event history together and produces the master event history. After combining all the rounds, the program checks for problems and cleans up inconsistent answers (see Grade School Cleanup below). Finally, for just the grade school data, yearly summary variables are produced for researchers who want information at a higher level.
Grade school cleanup
The grade school event history programs have a large amount of code designed to handle missing data and inconsistencies. Here are the most important checks:
First, the program checks each three-month group of data to see if there are spurious grade transitions. An example of a three-month group is September, October, and November. A spurious transition is when the two outer months match but the middle month does not. For example, if the respondent reports attending 10th grade in September, 9th grade in October, and 10th grade in November, the program changes the 9th grade into 10th grade since few respondents jump down and then up a grade within a 30-day time frame.
Second, the program checks on the overall grade progression. In the overall grade progression the program assumes that all respondents know when they graduated from high school. The program looks for respondents who reported the same grade twice but who graduated on time. For example, if a respondent reports being in 9th grade, 10th grade, repeats 10th grade, 12th grade and then graduates, this suggests that they actually attended 11th grade and did not repeat 10th grade. The program would change their progression to be 9th grade, 10th grade, 11th grade and 12th grade. If the respondent did not report the type of school (private, public or religious) in a previous round, the program searches the event history arrays to see if the school was ever reported. If it was previously reported, the information is then propagated through the event histories.
College cleanup
The program checks each three-month group of data to see if there are spurious semester transitions. An example of a three-month group is September, October, and November. Since it is possible for a student to take a college course that lasts just one month, the program handles transitions differently than the grade school program. If the respondent reports attending college in September, October, and November, but reports that the October college experience was different than September's and November's, then each month is counted as a separate semester. If the respondent did not report in a particular round the status (full or part time) or the type of school, the program searches the event history arrays to see if the school was ever reported. If it was previously reported the information is then propagated through the event histories.
Information on NLSY97 respondents' educational experiences—school enrollment, attendance, and attainment—is collected in each survey year.
Enrollment
The round 1 Youth Questionnaire gathered information on the respondent's enrollment status. Because the fielding period included summer vacation, the survey determined enrollment status through a number of questions. Each respondent was first asked about his or her current enrollment status. Those on summer vacation reported whether they were (1) currently enrolled in summer school, (2) on break from a regular school, or (3) not attending summer school. Respondents who were enrolled at the time of the survey, those on break who reported being enrolled in the spring of 1997, and those who were in summer school and reported being enrolled during the 1996-97 academic school year were classified as "enrolled."All other respondents were classified as "non-enrolled" (e.g., interviewed during the school year and not enrolled, on break and not enrolled in the spring of 1997, in summer school and not enrolled during the 1996-97 academic year).
In later rounds, the schooling section begins by confirming and correcting the enrollment status data collected during the previous round. The survey then proceeds through questions about enrollment status in an event history format.
In each survey, those who report a gap in enrollment are questioned about their reason for leaving school and the date at which this separation occurred. All respondents are asked for the name of their current or most recent school and the type of school (e.g., public, parochial, private). The respondent then verifies the location of the primary or secondary school from a list of school names and addresses called the School Finder. The School Finder is a record of information about primary and secondary schools located in the United States. This information is taken from the "National Education Database," provided under copyright by Quality Education Data, Inc., and is not released to the public.
Attendance
Youth Questionnaire. If enrolled since the date of last interview, respondents are surveyed about gaps within enrollment. These questions ask for gaps since the last interview and the reason respondents left school when each enrollment period ended.
Respondents are also asked whether they have ever been suspended from school and the grade level(s) in which this occurred. For each reported incident, the surveys collect information on the duration of the suspension. Starting in round 2, the survey also asked for retrospective information about any suspensions during the period before the previous interview if the information had not been reported by the respondent in that survey.
In round 1, respondents in the 12th grade or lower were also questioned about the number of days they were absent from school during the fall 1996 term.
Parent Questionnaire (round 1). In the round 1 survey, the responding parent answered questions on whether the youth ever missed a month or more of school since the 7th grade (not including summer vacation). For each absence, follow-up questions collected information on the length of the gap, the grade level(s) in which this occurred, and the reason for that absence.
Additional information was gathered on whether an NLSY97 youth ever skipped or repeated a grade level. If the responding parent indicated that a youth had either skipped or repeated a grade level, he or she was asked to state the grade level before and after the change.
If the parent reported a youth in the 9th grade or higher, he or she was asked whether that youth had ever taken an academic class during a school break. If so, the responding parent was surveyed on the reason for that class.
Attainment
The Youth Questionnaire asks all respondents to state the highest grade level attended and the highest grade level completed. Beginning in round 2, data from the previous round are confirmed or corrected and information since date of last interview is then collected.
If a respondent is currently enrolled in high school, information is collected on the date he or she expects to graduate.
If round 1 respondents were either enrolled in college or they were not enrolled but reported their highest grade level attended as 8th grade or higher, were asked whether they had received a high school diploma. In subsequent rounds, this question was addressed to respondents who listed their highest grade completed as 12th or more or who had attended college. After establishing that a respondent received a high school diploma, all the surveys asked for the date the degree was received and the name of the high school where the degree was earned. The respondent then verified the school's location as provided in the School Finder, which is a list that provides information about K-12 schools in the United States.
Respondents who reported earning a GED were surveyed on the date and the state where the GED was earned. A follow-up question established the type of program (e.g., job training, adult education, child care program) used to earn the GED. If the respondent reported college enrollment, college-specific information was collected on the type of diploma or degree sought (e.g., associate's, bachelor's). Because some respondents report their GED in the youth training section of the survey, users are encouraged to use the CV_GED variable to obtain more complete information on GED receipt and the date of receipt. Additional questions on training programs that helped the respondent earn a GED are described in Training. In rounds 7 and up, respondents reported the highest degree they received.
| Comparison to Other NLS Surveys | Enrollment status in the past year, current enrollment status, highest grade attended and completed, and degrees and diplomas received have been collected for all cohorts except the Older Men. Respondents in the Older Men cohort provided information on their highest grade completed and highest college degree received. For more details about specific survey years in which these data were gathered, refer to the appropriate cohort's User's Guide. |
|---|---|
| Survey Instruments | Questions on educational status and attainment are found in the schooling section (question names begin with YSCH) of the Youth Questionnaire and section PC8 of the round 1 Parent Questionnaire. |
| Related User's Guide Sections | College Experience School Experience School & Transcript Surveys Training |
| Main Area of Interest | Educational Status & Attainment |
| Supplemental Areas of Interest | Education: College Experience Education: School Characteristics Education: School Experience |