This is a flag variable which identifies observations with
inconsistent changes in the information on highest educational
qualification compared to the previous observation or year.
Questions on highest educational attainment have been asked in
the first survey and were only updated in subsequent years if
the respondent reported a change. In the year 2000, every single SOEP participant was asked his highest level of
educational attainment which produced a number of
inconsistencies between the most recent information from 2000
and the generated information from previous years.
These inconsistencies include both
higher and lower educational attainment and are not just due to repeating the
question about educational attainment in 2000. They also occur
more generally, although to a lower degree, in the second
survey wave of new samples when respondents to individual and
life history questionnaires are asked to state their
educational attainment. In both situations, respondents are not
only asked annual questions about any changes in educational
attainment since the previous year, but are also asked to state
their highest level of educational attainment.
In our view there is no means of unequivocally correcting for
these inconsistencies. The flag variable helps researchers to
identify observations with inconsistent answers to educational
questions in the cross-sectional perspective.
Researchers need to decide how to deal with these on a case-bycase
basis depending on the research question at hand.
So far, we have not found evidence that respondents with a
change in the year 2000 differed systematically from other
respondents. One possible approach would be to exclude these
individuals from the analysis when sample size allows.
Alternatively, one could apply the information collected in
2000 to the prior years in which no changes were recorded
between two years and test whether the results differ from
those obtained when these individuals are left out.
File BIOEDU contains new data on consistent longitudinally tested
educational transitions.
This is a flag variable which identifies observations with
inconsistent changes in the information on highest educational
qualification compared to the previous observation or year.
Questions on highest educational attainment have been asked in
the first survey and were only updated in subsequent years if
the respondent reported a change. In the year 2000, every single SOEP participant was asked his highest level of
educational attainment which produced a number of
inconsistencies between the most recent information from 2000
and the generated information from previous years.
These inconsistencies include both
higher and lower educational attainment and are not just due to repeating the
question about educational attainment in 2000. They also occur
more generally, although to a lower degree, in the second
survey wave of new samples when respondents to individual and
life history questionnaires are asked to state their
educational attainment. In both situations, respondents are not
only asked annual questions about any changes in educational
attainment since the previous year, but are also asked to state
their highest level of educational attainment.
In our view there is no means of unequivocally correcting for
these inconsistencies. The flag variable helps researchers to
identify observations with inconsistent answers to educational
questions in the cross-sectional perspective.
Researchers need to decide how to deal with these on a case-bycase
basis depending on the research question at hand.
So far, we have not found evidence that respondents with a
change in the year 2000 differed systematically from other
respondents. One possible approach would be to exclude these
individuals from the analysis when sample size allows.
Alternatively, one could apply the information collected in
2000 to the prior years in which no changes were recorded
between two years and test whether the results differ from
those obtained when these individuals are left out.
File BIOEDU contains new data on consistent longitudinally tested
educational transitions.