Hello,
For the variable ts23201_g4 (Professional Title) in spEmp, SC6 cohort I find a value of -29 which corresponds to the label:" value from the last sub episode " the harmonized spells. can you kindly outline the procedure to fill these missing values , if possible from subspells. Thank you.
Dear Lisha,
in harmonized episodes, the missing value -29 (“value from the last sub episode”) generally should not occur. If it does, this is an indicator for an error in field work and/or data preparation: a subspell lists this value, but no other subspell corresponding to the reported episode lists a valid one.
However, in version 10.0.0 of the NEPS SC6 SUF, there was an error in executing the rules that are followed in order to the harmonized episodes being filled. This led to erroneously filling in -29 (and other, less suspicous values) in several variables (for instance all of the generated variables ts23201-g*). Thus, we published a minor update to this Scientific Use File (version 10.0.1, doi:10.5157/NEPS:SC6:10.0.1) last week (on October 24th) which fixes this issue:
. neps set study SC6 version 10.0.0
. neps : use spEmp , clear
. count if ts23201_g4==-29 & spgen==1
16,017
. neps set study SC6 version 10.0.1 , replace
. neps : use spEmp , clear
. count if ts23201_g4==-29 & spgen==1
0
Regarding the more generic part of your question: The missing value -29 indicates that during fieldwork, the interview system used a value that has been reported in a preceding interview for navigating through the interview instrument (e. g. for triggering filter conditions). For persons without a preceding interview, the information is directly assessed.
We are aware of the fact that in some occasions, these missing values can be confusing, and are currently evaluating on how the situation could be improved.
Kind regards
Bela