For the 1993 wave of data collection, the PSID began using electronic interviewing for the first time. Due to system crashes, a number of interviews were conducted using paper and pencil and were entered by coding staff into the interviewing application, and hence the Survey Research Center sample management systems logged the entry date as the interview completion date.

PSID staff failed to notice this for some time, and when the error was discovered, mounting an effort to identify and locate the paper materials and to hand code the dates was not practicable with staff resources then available, although we fully intended to correct the data.

However, the date on which the interview actually occurred is very important for some users, and a timely request by Dr. Joseph Altonji at Yale University and Dr. Nicolas Williams at the University of Cincinnati combined with a staffing window of opportunity precipitated our completion of this project.

This file contains corrections for the month-day combination variable (V22403) for date of interview for the PSID 1993 Family data. The dates are in error for 857 of the 9,977 families; this file contains only those cases requiring corrections.

The 1993 Date of Interview Correction file consists of one data file with 857 cases and two variables, and each record has a length of 9. In addition to the data file above and this document, we also include SAS, SPSS and STATA data definition statements.

The data in the .zip package are in raw ASCII form, and the file is named IWDATE93.DAT. Refer to the corresponding .SAS, .SPS or .DO file for record format layout information.

Two variables are present on the file: V21602, "1993 INTERVIEW NUMBER" and V22403, "3. DATE OF 1993 INTERVIEW". The contents correspond to variables with the same names in the 1993 Family codebook. To use the corrections, match to the family file using the family identifier, V21602.

The corrections file is not available through the PSID Web Site Data Center, since the main family file will be corrected at the time of release of our next wave of data.