I am trying to see particular CPS-ASEC questions at the local level (county or city), which it is not designed to do. So I am combining single CPS years into a multiple-year sample, hoping to make it sufficiently large to use at local level. But how do I handle the weighting? Each year carries a distinctive weight that cannot be applied to other years. So do I simply run each year individually with those weights, then combine the resulting weighted figures together, then take an annual average? And other than weights, any other warnings or pointers about pooling CPS across time? Thanks.
There is really not one particular method for pooling and weighting CPS data across samples. The method you propose sounds reasonable. Another possibility is to divide your sample weight by the number of samples you are combining. So, for example, if you are pooling 10 samples together you’d divide the sample weight by 10 in your pooled sample. This is undoubtably imperfect, so you may want to try a couple different methods and see how the results compare.
I hope this is helpful.
Thanks.