Jobsuppwt in Job Tenure and Occupational Mobility Supplement


I am having trouble interpreting the variable JOBSUPPWT in the 1983 and 1987 Job Tenure Supplements. In particular, if I tally the count of workers by year by summing all of the weights, I get wild changes in the working population between 1983 and 1987. Is there something wrong here?


Thank you, your inquiry has helped us uncover errors in our calculation of the job tenure supplement weight JTSUPPWT in the 1983 and 1987 samples. For the 1983 sample, JTSUPPWT was inadvertently truncated and for the 1983 and 1987 samples, JTSUPPWT needed to be multiplied out to four decimals (as opposed to two). These errors have been fixed and will show up on the website with the next IPUMS CPS data release. Note that while these corrections put totals for each sample in the millions, their values still fluctuate quite a bit (1983 = 109,483,242.2912 and 1987 = 188,892, 499.1488) which is consistent with the original CPS data. We look forward to sending you an IPUMS mug as a token of our appreciation for helping us improve IPUMS CPS.

1 Like