OECD (2000)

ANNEX 3
INDICATOR C1: Overall participation in education
General notes
Notes on methodology
• Reference dates
Statistics that relate participation data to population data are published for the reference date that was used
by national authorities for these statistics. The assumption is made that age references in the enrolment data refer
to 1 January of the reference year. For Australia 1 July is used as the reference date for both enrolments and population
data. For Japan 1 October is used as the reference date for ages.
The dates or periods at which students, educational staff and educational institutions were counted have not
been provided to the Secretariat by all countries. Some countries collect these statistics through surveys or administrative
records at the beginning of the school year while others collect them during the school year, and yet others
at the end of the school year or at multiple points during the school year. It should be noted that differences in the
reference dates between, for example, enrolment data and population data can lead to errors in the calculation (e.g.,
net enrolment rates exceeding 100 per cent) in cases where there is a significant decrease or increase over time in
any of the variables involved. If the reference date for students’ ages used in the enrolment data differs from the reference
date for the population data (usually 1 January of the reference year), this can be a further source of error in
enrolment rates.
• Expected years of schooling
School expectancy (in years) under current conditions (Table C1.1) excludes all education for children younger
than five years. It includes adult persons of all ages who are enrolled in formal education. School expectancy is calculated
by adding the net enrolment ratios for each single year of age. No data are available for the ages 30 and
above. For persons aged 30 to 39 enrolment rates were estimated on the basis of five-year age bands and for persons
40 and over enrolment rates were estimated on the basis of the cohort size of 39 year-olds.
© OECD 2000
340
• Expected hours of training over the life cycle and participation rates in continuing education and training
Data on continuing education and training are from the International Adult Literacy Survey (IALS), which was
undertaken by Statistics Canada and the OECD at the end of 1994 and in 1995. For details on IALS see Indicator A2.
Indicator C1 combines information on participation in formal education, based on the UOE data, and in continuing
education, based on IALS. When combining these two sources the problem of different coverage arises. IALS
asked for any education and training, which includes education under the coverage of the UOE. In order to complement
the participation statistics in formal education it is necessary to exclude in IALS education of students who are
most likely to be covered in the UOE. Therefore IALS data were not included for those students who took at least one
course that exceeded ten weeks and which lead to a university degree, a college diploma/certificate, a tradevocational
diploma/certificate or an apprenticeship certificate.
The expected number of hours of training is calculated as the sum of the average number of hours of training by
64
single year of age for the ages 15 to 64. SH = ∑ Ha where Ha
is the average number of hours of training for persons
a=
15
aged a. is calculated as,
∑ H
Ha
Ha = where H is the number of hours per participant and weight is the IALS
∑Weight
sampling weight for each participant. All standard errors were calculated using jack-knife estimates based on the
30 replicate weight. For details see IALS Microdata Package Guide Section 8.1.
IALS participants were asked in how many courses they were enrolled in the last 12 months. More details were
obtained for the three most recent courses. Therefore the number of hours was calculated on the basis of these three
courses. In order not to underestimate the total number of hours for those students participating in four or more
courses, information on these courses was imputed
Therefore all respondents are divided into classes by country (i) and number of courses taken (nc). For each group
the average duration of a single course was calculated as ∑(
hi,
nc /3)
Ci,
nc =
. For each person in country i who took
weighti,
nc
∑