on Facebook
on Twitter
on YouTube
Data Alert for RAND HRS Ver H (Feb-2009)
We released Version H of the RAND HRS dataset at the end of Feb-2008. Since then we have discovered two inconsistencies that have now been corrected. Many thanks to the users who brought them to our attention.
The RAND HRS data files have been updated on the HRS web site in the data download area.
You can downloaded the entire updated dataset from HRS web site. If you prefer a file with just the changed variables to merge with your current data, please contact RANDHRSHelp@rand.org. Note that these corrections will be incorporated in RAND HRS Ver I, which should be released within a month or so.
We are sorry for any inconvenience this may have caused. Please let us know if you have any questions by emailing RANDHRSHelp@rand.org.
The corrections are:
Corrections to SSI disability application status
An error was found in the disability section. We inadvertently missed updating a raw variable in an array. The result is that we did not see anyone in R8DSTAT receiving or applying for SSI disability. The information on Social Security disability application and receipt was correctly coded. The problem affected about 700 cases in wave 8 and 200 cases in the episode variables.
The following variables have changed values due to the correction:
Wave specific variables:
R7DSTAT, R8DSTAT — Respondent disability application status W7W8
S7DSTAT, S8DSTAT — Spouse disability application status W7W8
Episode variables:
RADNEPI Number of disability application episodes
RADTYPE1 to RADTYPE3 — type of episode (SSI or SSDI)
RADSTAT1 to RADSTAT4 — status of episode (e.g., applied or received)
Here is the replacement Social Security codebook pages for Section E: Social Security.
Corrections to Total Years Worked
A correction has been made to Version H that reflects a more complete early job history and an increase in the number of years worked for waves 7 and 8. The changes primarily affect the Early Baby Boomer (EBB) cohort. A special thank you is extended to Vipul Bhatt for bringing to our attention lower than expected job years found in the EBB cohort. It is just one more example of why we welcome and value feedback from our user community.
The affected measures include RwJYEARS, SwJYEARS (Years Worked/selfrpt+ jobh) and RwJYEARM, SwJYEARM (Years Worked/missing flag) for waves 7 and 8. Here are descriptive statistics of the corrected measures.
Top