ETF - Cohort-sequential longitudinal databases Evaluation through follow-up

Summary

Many have emphasized the value of the longitudinal cohort-sequential design, the accessibility of a rich and varied set of variables, as well as the opportunities to add data from other sources by means of the Swedish National Identification Number. For some studies, changes in the measurement instruments between cohorts can make comparisons more difficult. Another weakness is that it is has not been possible to gather data before the 6th school year for some cohorts. Up until recently very high participation about 80 per cent.

Type of data

Data Source
Registry + Survey

Type of Study
Survey same
Survey different
Cohort study

Data gathering method
Registries
Self administered questionnaire

Access to data

Conditions of access
Data are available only for the scientific community. A CD is provided after agreement.

Type of available data (e.g. anonymised microdata, aggregated tables, etc.)
Anonymised microdata

Formats available
ASCII, SAS, SPSS, Text

Coverage

Coverage Years of collection, reference years, and sample sizes
Other sources can be added to ETF-data and therefore it is possible to research ETF-data from the start of a cohort to the current situation. The cohort born in 1948 from the 6th school class year, 1961, until 1968. The sample size consisted of 10,000 pupils. The cohort born in 1953 from the 6th school class year, 1974, until 1968 The sample size consisted of 10,000 pupils. The cohort born in 1967 from the 6th school class year, 1980, until they completed Upper Secondary School. The sample size consisted of 10,000 pupils. The cohort born in 1972 from the 3rd school class year, 1982, until they completed Upper Secondary School. The sample size consisted of 10,000 pupils. The cohort born in 1977 from the 3rd school class year, 1987, until they completed Upper Secondary School. The sample size consisted of 5,000 pupils. The cohort born in 1982 from the 3rd school class year, 1992, until they completed Upper Secondary School. The sample size consisted of 10,000 pupils. The cohort born in 1987 from the 3rd school class year, 1997, until they completed Upper Secondary School. The sample size consisted of 10,000 pupils. The cohort born in 1992 from the 3rd school class year, 2002, until they completed Upper Secondary School. The sample size consisted of 10,000 pupils. The cohort born in 1998 from the 3rd school class year, 2008, and will go on until they complete Upper Secondary School. The sample size consisted of 10,000 pupils. The project comprises a nationally representative sample of pupils consisting of those born on the 5th, 15th or 25th of any month in 1948. The material covers some 12 000 pupils. The collection of the data was commenced when the majority of the pupils were in Grade VI (aged about 12), that is to say in 1960/61.

First year of collection
1961 for the cohort born in 1948.

Stratification if applicable
No.

Base used for sampling

Geographical coverage and breakdowns
The data set covers the entire country and is a representative sample of the age cohort. The project comprises a nationally representative sample of pupils consisting of those born on the 5th, 15th or 25th of any month in 1948. The material covers around 12 000 pupils.

Age range
From 10-13 years of age until they are no longer living

Statistical representativeness
Population representative

Coverage of main and cross-cutting topics
There are almost limitless possibilities providing the opportunity to add registry data from other sources. The project comprises a nationally representative sample of pupils consisting of those born on the 5th, 15th or 25th of any month in 1948. The material covers some 12 000 pupils. The collection of the data was commenced when the majority of the pupils were in Grade VI (aged about 12), that is to say in 1960/61. The data kept in the register may be said to be of two types: I) Background data were collected in the spring of 1961. On this occasion the Department of Pedagogics at University of Gothenburg collected particulars of the results of certain intelligence tests and of national standardised school tests, as well as replies to questionnaires about the pupils' leisure-time interests, future plans, and about the attitude of the pupils and their families to school and school work. At the same time Statistics Sweden (SCB) collected particulars on the father's/mother's occupation/education, the number of their brothers and sisters and the distance travelled by them between home and school. 2) Annual data ("school-year data") on the pupils' later schooling have been collected since 1960/61 by the SCB, up to the end of 1968/69. These data relate to place of residence, type of school system, grade, terminal marks, and certain optional subjects. This collection of annual data related to pupils at public elementary schools, experimental schools, comprehensive schools, intermediate schools, girls' schools, continuation schools, vocational schools and upper sec-ondary schools or corresponding schools (i.e. the general upper secondary schools according to the 1954 curriculum, technical and commercial upper secondary schools, upper secondary schools according to the 1965 curriculum, and integrated upper secondary schools).

Linkage

Standardisation
Use national standard harmonised with international types

Possibility of linkage among databases
There is an ID variable which makes it possible to link registry databases to Statistics Sweden and all official registers.

Data quality

Entry errors if applicable
High quality.

Breaks
Please specify in no more than 1,200 characters with spaces (approx. twelve lines). Changes in the school system and in the community has led to some changes in the questionnaires. More information in the Anthology of the ETF-project: Svensson, A. (Red.). (2011). Utvärdering genom uppföljning. Longitudinell individforskning under ett halvsekel https://gupea.ub.gu.se/handle/2077/24366 (in Swedish) http://www.ips.gu.se/forskning/forskningsprojekt/ugu/ (in Swedish) https://gupea.ub.gu.se/handle/2077/24366 (Summery p.p. 267-277 in English) http://www.ips.gu.se/english/Research/research_projects/ETF/Document/ (in English)

Consistency of terminology or coding used during collection
During the 50 years of existing, there have been developments with the sampling technique.

Governance

Contact information
Jan-Eric Gustafsson, Department of Education and Special Education
Gothenburg University
Box 300
405 30 Gothenburg Sweden Phone: +46 31 786 0000
Email: michael.hansen(at)ped.gu.se
Url: http://www.ips.gu.se/forskning/forskningsprojekt/ugu/ http://www.ips.gu.se/english/Research/research_projects/ETF/

Timeliness, transparency
There are about 12 months between when data is collected and then published.