| In general, a substantial amount of detailed data is collected for the Reference Person (‘Head’ prior to 2017) and Spouse/Partner, if present. Considerably less detail is collected for Other Family Unit Members (OFUMs). |
| The PSID collects many data elements about housing, including housing type, characteristics, ownership, tax, insurance, etc. A list of such items collected in each wave is available here. |
| File release information is available through the News section of our website. |
Up through the 2001 interviewing year, the PSID distinguished between Main and Extra jobs. Someone could not have an Extra job unless they held a Main job during the same time period. The extra job must be held simultaneously with the main job. We made this distinction between main and extra jobs throughout. If two (or more) employers overlapped, the interviewer was supposed to ask which was the main one during that time and note in an open ended question the overlap and the hours and earnings of both jobs. Then this overlap period was to be included in the extra job sequences (BD82-BD106/CE74-CE98). Those who are only temporarily laid off are still employed at a main job and, therefore, could have an extra job during that time period. However, those who are unemployed, whether looking or not, have no main job employer during the time in question. Hence, any small job they may have is considered a main job--since it's the ONLY job. Use the month strings and dates of beginning and ending employment in the work history to tell whether time at B/D72-74a or C/E64-66a is temporary layoff or unemployment.
Beginning with the 2003 interviewing year, the PSID dropped the main vs. extra job distinction as defined above. Jobs are now classified as "current main job", "most recent main job" or "other" job. If someone reports 2 or more current jobs, or 2 or more recent jobs that ended at the same time, the interviewer asks which job they considers their main job. That one is listed as the current (or most recent) main job. Any other job is listed as an "other" job. A job can be an "other" job even when it does not overlap with a current or most recent main job. This situation could arise, for instance, when someone reports two jobs, with the current main job beginning before the old ("other") job ended. |
| The PSID contains a wealth of information that can be used to study the health of Americans and their family members, including measures of
health status and behavior
and personality and psychological functioning.
Measure of health,
personality and psychological functioning
are also available in the CDS and TAS. Information that has been collected specific to the Covid-19 pandemic is available for both the main
PSID
and CDS/TAS supplements. |
| The PSID used a one-digit, and later a two-digit, occupation code until 1981 when the three-digit 1970 Census code became standard for the main jobs of employed Reference Persons (‘Head' prior to 2017) and Spouses/Partners. It was also used for the most recent jobs held by Reference Persons and Spouses/Partners who were currently unemployed and looking for work and for any job held in 1980 by a Reference Person or Spouse/Partner who was currently retired or no longer in the labor force. From 2003, all occupation-industry data was coded using the three-digit 2000 Census code. A retrospective coding project used the 2000 Census to code first occupation and industry of all Reference Persons and Spouses as of 2003 and that of their fathers and mothers. Starting in 2017, the study is using the 4-digit 2010 occupation and 2012 industry codes. |
| Ages of individuals are asked and reported in each wave of the study. But interviews are seldom taken exactly twelve months apart for the same family from wave to wave. In fact, a family responding early in the interviewing period one year might respond late in the next year’s interviewing period, with 18 or more months between interviews for annual interviews (from 1968-1997). Conversely, a late responder in one wave could be an early responder in the next wave. Since the PSID transitioned to biennial interviewing (1999 through the present), the age gap can widen even further. Because of interview dates, there is a good possibility that an individual appears to have aged excessively or not at all. Also, individuals’ ages or birthdates can be misreported. Consistency checks for age discrepancies have always been done internally, but they are not altered if it cannot be determined which age is correct. |
| A modest upward distortion in the weighted estimates of Black individuals with children has been identified in PSID, beginning in the late 1990s, for selected cohorts. We recommend analysts use caution in estimating with the PSID beginning in 1997 the percentages with children in the household (and related statistics such as childlessness and fertility) for Black women born in the late 1960s and in the 1970s, and for Black men born in the late 1960s, 1970s, and early 1980s. This distortion may also affect estimates of multigenerational Black families in later years. Analysts who are not addressing questions of this type can reasonably ignore this concern. When estimating multivariate models with fertility-related outcomes for any year, we recommend that analysts include as a control variable the CDS eligibility indicator available on the 1997 Individual File (ER33418). Some analysts may wish to post-stratify the assigned PSID weights for Black individuals to the Current Population Survey (CPS) totals by presence of a child under the age of 13. PSID Technical Series Paper 16-01 provides additional details. |
|