this report vary depending on the nature of the analysis. To indicate educational
level before enrolment in AE, SUN1990 is employed. Most of the individuals
in the cohort concluded their upper secondary education in 1988, but SUN1990
also includes correct levels for those that for some reason delayed completion
of upper secondary school. In the second part of the analysis, SUN2001 and
SUN2003 are used to indicate the attainments of individuals registered in
higher education.
The population of the study consists of the full cohort born in 1970 and re-
siding in Sweden from 1988 to 2001 (in total 126,604 individuals). In 1988, the
year from which there is uninterrupted register data on studies at komvux, they
were 18 years old. The educational achievements of foreign born persons in
Sweden have been shown to differ if immigration occurs at the age of 12 or 13
(see e.g. Bohlmark 2005), so those who immigrated to Sweden after 1982 are
excluded (leaving 106,727 remaining individuals).5
Naturally, komvux is most interesting for individuals with a short education.
Table 1 presents frequencies of the educational level reported in SUN1990. The
individuals were then 20 years old, which explains the low fractions with
higher education. From 1988 to 2001 there were 36,380 individuals, or 34.1 per
cent, who at some point were registered at upper secondary komvux. The par-
ticipation rates across gender differ quite sharply, 25.6 per cent among men and
43.3 per cent among women.
5 Schooling becomes more individualized from the age of 13. Pupils no longer remain in the
same classroom and instead of having one teacher to cover most subjects there are several spe-
cialized teachers.
IFAU - Does adult education at upper secondary level influence annual wage earnings?
11