University of Gothenburg

Data collected

ETF currently has ten samples of students who have been followed through the school system. The samples are nationally representative and make up about 10 percent of the population in a birth cohort. With ETF's collections, it is possible to follow a cohort through the education system and compare different cohorts at the same time in the educational system.

The first cohort consists of individuals born in 1948. Five years later, another cohort of students born in 1953 was added. The next cohort includes students who were mainly born in 1967. Since then, new cohorts have been added at five- to six-year intervals. The youngest cohort consists of students who were in grade 3 in the year 2014. The majority were born in 2004.

Details about selection principles and collected information can be accessed via the menu on the left where you can read about each cohort. The table below provides an overall picture of what data is available for each cohort.

The data collected are mainly of two types

  1. School administrative data, which is collected annually by Statistics Sweden during the time the pupils are in the general school system (primary and secondary school). This information is provided by the school office and is available in various lists, e.g. school, class, special support, study choices and grades. Survey data from the pupils are collected, for most cohorts, for the first time in compulsory school year 6.
  2. The collection in school year 6 concerns e.g. questions about self-perception and interest in learning, school attitude, hobbies, school motivation and future plans. Furthermore, results from various aptitude tests and knowledge tests carried out in school year 6 are included. For several of the samples, questionnaire data has also been collected from the students' parents and teachers.

General information about collected data

In order for the collected data to be of the greatest possible value, the ability to study the development among students within the same year group, but also be able to make comparisons between different year groups is desirable. In order to fully meet the latter condition, the data should be acquired at the same age and be identical, or as similar as possible, so as to enable meaningful comparisons between cohorts.

For various reasons, this is not possible.  Hence, some of the instruments used may need to be revised and supplemented. The table below presents the data available for each cohort.

This is followed by general information on how the collections were carried out for pupils in grades 3, 6 and 9, and after compulsory school. This is followed by general information on selection principles and an interactive list of variables. For those who want to take a closer look at one or more cohorts, it is possible to read specifically about each cohort's selection, collections, tests, surveys, etc.

Overview data

The year of birth of the cohorts is stated in the table heading. For each cohort there is a X if the described data is collected and available. There are sometimes differences between questionnaires even if they are given in the same year, for more detailed information see the respective cohort.

Data 2004 1998 1992 1987 1982 1977 1972 1967 1953 1948
Basic data


School administrative data X X X X X X X X X X
Pupil questionnarie school year 3 (incl. tests)           X X      
Pupil questionnarie school year 6 (incl. tests) X X X   X X X X X X
Pupil questionnarie school year 9 X X X X            
Pupil questionnarie school year 10             X X    
Pupil questionnarie school year 12   X X X X X        
Parent questionnarie school year 3             X      
Parent questionnarie school year 6 X X X   X X X X    
Teacher questionnarie school year 6 X X X   X          

Collection school year 3

School administrative data is collected annually by Statistics Sweden, starting in year 3 of compulsory school for all cohorts (from sample 4, i.e. students born in 1972). For two of these cohorts (1972 and 1977) a survey was also conducted this year.

Collection school year 6

The information collected for the nine cohorts (1948, 1953, 1967, 1972, 1977, 1982, 1992, 1998 and 2004) when they were in school year 6 of compulsory school is as follows:

  • School administrative data collected by Statistics Sweden. This type of data has been collected since the fourth cohort, but the information obtained has varied somewhat, e.g. due to changes in curricula. A more detailed description of this data collection can be found in reports published by Statistics Sweden. For the two oldest cohorts, the report applies to Statistics Sweden 1976 and for the four later cohorts, refer to the respective reports by, Statistics Sweden 1983, 1986, 1991 and 1996.
  • Results on three aptitude tests. The aptitude tests used have been identical on all nine occasions. As the aptitude test is only offered in year 6, this type of information is not available for the seventh cohort, from which questionnaire data were first collected in year 9. The aptitude test consists of a verbal, a spatial and an inductive test, each containing 40 tasks and specially designed for ETF- the project. The verbal test is a vocabulary test of antonyms . The spatial test is a so-called sheet metal folding test and the inductive test is made up of series of numbers. The reliability of the test, intercorrelations and connection with school grades are reported by Svensson (1971). However, for reasons of confidentiality, there is no account of the test data included.
  • Results on knowledge tests. For the first three cohorts (1948, 1953 and 1967), the knowledge tests consist of the standard tests in Swedish, mathematics and English that were offered to all pupils in compulsory school year 6 up to and including the beginning of the 1980s. For the fourth cohort (1972) specially prepared tests in reading and mathematics were used. The reading test consists of 27 tasks and aimed to identify students with reading difficulties. The mathematics test, which was also offered for the fifth cohort, (1977) contains 19 tasks. After fairly extensive revision due to the fact that the previously used test was judged to be somewhat too simple, this test also been offered to the sixth cohort (1982). For further information, see Ek and Pettersson (1985) and Reuterberg et al. (1996). Results on the mathematics test are not available for the 1987 cohort. The mathematics test was also not offered to the pupils in the eighth cohort as the test did not seem to correspond fully with current curriculum intentions in mathematics.
  • Survey data. These deal with students' self-esteem and attitudes to school and education, specialization, study and career plans. However, they have had a different design on the nine collections that have been made. Detailed accounts of the questionnaires used on the various occasions can be found in Svensson (1964), Rovio-Johansson (1972), Emanuelsson (1981), Ek and Pettersson (1985), Reuterberg and others. (1991), Reuterberg et al. (1996) and Giota (2008).
  • Survey information from the guardians. Such data are available only for the later seven cohorts (1967, 1972, 1977, 1982, 1992, 1998 and 2004). The questionnaires contain questions about e.g. the child's upbringing conditions, ambitions and wishes regarding the child's education, views on the school's objectives and the parents' own educational and professional situation. The questionnaires used for the first two cohorts are described by Emanuelsson (1981) and the questionnaires used for the next two cohorts are described by Reuterberg et al. (1991, 1996).
  • Questionnaires from teachers. In order to get as complete a picture as possible of the students' school situation, a questionnaire was also offered to the teachers for the 1982 and 1992 cohorts. The teacher survey consists of questions, which concern certain facts about the teacher, the size and composition of the class, the teacher's assessments of the class' level of knowledge, etc., school resources, working methods and parental involvement and questions about the existence of evaluations (Reuterberg et al., 1996).

Collection school year 9

For the cohorts born in 1987, 1992, 1998 and 2004, questionnaires were conducted during the spring term in year 9.

Collection after primary school

The students are followed up even after they have left primary school. For instance. data collection is done during the time they are in high school. Then school administrative data such as e.g. choice of upper secondary school line / program and grades after completing studies. For seven of the cohorts, in addition to school administrative data, survey data were also collected from the students.

Those included in cohorts 3 and 4 (1967 and 1972 cohorts) received a survey one year after the end of compulsory school when the majority were in upper secondary school year 1 (spring term 1984 and 1989). The questions concerned i.e. views on and experiences from compulsory and upper secondary school, perceptions of one's own ability, employment during leisure time, study and career plans

The individuals in the cohorts thereafter (1977, 1982, 1987, 1992 and 1998 cohorts) did not receive a questionnaire one year after completing primary school. Instead, this was done three years after the end of compulsory school, in the spring term when the majority were in year 3 of upper secondary school (or 12th year in the school system). The main reason for the postponement was to obtain fuller information about the students' experiences of their studies in upper secondary school. In addition to the questions used for cohorts 3 and 4, one year after completing compulsory school, this questionnaire also contains corresponding questions about all their time in upper secondary school (Svensson, 2001).

Selection principles

ETF's samples are nationally representative and make up about 10 percent of the population in a birth cohort. The first cohort consists of individuals born in 1948. Five years later, another cohort was added with students born in 1953. The next cohort includes students who were, for convenience reasons, mainly born in 1967. Since then, new cohorts have been added at five / six year intervals.

ETF's selection process has been based on two different principles. For the first two cohorts, the selection was based on birthdate and for the remaining cohorts by school class. From the collection in 1980, for students who were mainly born in 1967, major changes were made to the management of statistics at Statistics Sweden and the concept of panel surveys was introduced. Cohort 1967 was named Student Panel 1, the subsequent selections Student Panel 2, and so on. In some contexts, therefore, both the term cohort and student panel appear.

Also note that the samples have been selected at different ages in different cohorts. The first three cohorts were formed when the majority of the students went to grade 6. The subsequent cohorts, on the other hand, were drawn when the students went to grade 3. ETF's longitudinal design means that no new students are added after the selection has been made. This means, among other things, that students who have moved to Sweden after the selection has been made will not be represented in ETF. For each cohort, there is a description of the selection principle in brief. A more detailed description of the project can be read in Utvärdering genom uppföljning Förberedande planering för individualstatistikens fortsättning en promemoria från SCB (pdf 16 MB ocr in Swedish) or Ett nytt individualstatistikprojekt (downloadable pdf from GUPEA in Swedish) and Longitudinell individforskning under ett halvsekel, (downloadable pdf from GUPEA in Swedish) .