University of Gothenburg

Cohort 7: Pupils born 1987

Cohort 7 is a stratified municipality and principal area selection where the majority of the students were born in 1987. The first collection was made in 9th grade in 2003 and consisted of questionnaires.


Unfortunately, it was impossible to conduct a study in 6th grade for this cohort. It is thus the only cohort where the participants did not complete the test.

The purpose of the comprehensive data surveys is to enable both longitudinal and cross-sectional surveys of large and representative student selections. The follow-up survey of this current sample began in the spring of 1997 when the pupils were in 3rd grade of compulsory school. Until 8th grade only school administrative data was collected through Statistics Sweden. In the spring of 2003, when the majority of the students attended 9th grade, the school administrative information was supplemented with data obtained directly from the students.

The survey population consists of the pupils who attended 3rd grade of compulsory school in the spring term of 1997. The population is by definition delimited with regard to grade affiliation, which means that it includes pupils with different birth years. The majority, or just under 97 per cent, were born in 1987. The sample eventually included 9,927 students. The number of students who remained in the survey in 2003 when the first questionnaire was sent out was 9464. The dropout is mainly due to moving abroad. The number of students who answered all or parts of the questionnaire in 9th grade is 6747 or 71 percent.

No parent or teacher questionnaires have been sent to this cohort.

The results from the data collection in 9th grade is described in a report through descriptive tables and diagrams. Several of the questions contain sub-questions that may represent indicators of more elusive phenomena. These scored sub-questions may form the basis for scales, which provide a more reliable means of measuring a phenomenon than individual questions can do. The reliability of the scales has been estimated using Cronbach's alpha and in general all scales show an acceptable reliability.

The next questionnaire was sent to the students in the spring of 2006 when the majority of the students attended 3rd grade of high school. The number of respondents was 5361 or just over 55 percent of the original sample.

Basic information

Basic information

Register data that Statistics Sweden collects in connection with the selection of classes. These mainly consist of:

  • year of birth
  • gender
  • migration background
  • socio-economic data

School administrative data

The school administrative information is obtained from 3rd grade and thereafter throughout the compulsory school period. A form is sent out to the school the student attends and the school fills in information for each student included in the survey. Information is collected in 3rd grade, initially sampling entire classes, but as students move over time, the form is sent to more and more schools which are requested to complete information for individual students. This is done so that the students who were included from the beginning can be followed over the years up to adulthood. This information consists of:

  • municipality
  • school
  • school grade
  • age integrated or age diverse class
  • mother tongue
  • Swedish for immigrants
  • date when moving to another school
  • new municipality
  • new school
  • grades when given
  • national exams in Swedish, English and mathematics from year 9

The collection in 9th grade

The original number of students in the sample was 9727 and the number of students who participated in the data collection in 9th grade amounted to 9464. The response rate was 71 percent and the internal dropout rate was generally low. Distortions in data caused by the sample design and omissions are handled using weighting variables.

The scale constructions that refer to self-assessments of general and specific school performance as well as the general interest in learning more in school largely correspond to the scales reported in the previous collection for 9th grade regarding cohort 6 born in 1982.

Conduct of the survey

The selection is based on entire principal areas in compulsory 3rd grade. Due to classes being divided and students changing classes since the selection was made, the students were so scattered that the only reasonable way to distribute the questionnaire was to send it to each student's home address. This raised the question of whether the questionnaire should be addressed directly to the student or to the student's guardian. After the issue was discussed with lawyers at Statistics Sweden, it was clear that it was in accordance with current rules that the questionnaire be addressed directly to the student.

This entailed some difficulties as it was assumed that not all or even the majority of the students were aware that they were included in the survey and that permission for the student's participation was obtained from the parents when the sample was drawn when the student was in 3rd grade. For that reason, the wording of the introductory text informing about the survey is particularly important (see first page of the questionnaire, appendix).

The final version of the student survey and cover letter was sent to Statistics Sweden (in mid-January), which printed and distributed the survey as a postal survey. The student questionnaire was sent out in week 7 in 2003. A thank you and reminder card was sent in week 9. A reminder with a new questionnaire was sent partly in week 10 and partly in week 12. The collection ended in week 18 in 2003, after which the bound questionnaire responses were scanned by Statistics Sweden and sent to the University of Gothenburg in unidentifiable condition.

More about the implementation can be read in the Technical Report Statistics Sweden compiled in connection with the submission of data.

Questionnaire for students

The final questionnaire included 30 questions of which 19 questions concern primary school, 2 questions concern computer use, 5 questions concern the pupil's continued schooling after compulsory school, 2 questions concern the pupil's leisure activities and work, 1 question concerns where the pupil lives and 1 question about the number siblings. Several of the questions are formulated as main questions with a number of sub-questions followed by four or five answer alternatives. The questions concern the following areas.

  • Six questions (1, 2, 22, 23, 24 and 25) concern the choice of primary and secondary school, i.e., active choice of municipal or independent school and the main grounds for these choices.
  • Six questions (3, 4, 5, 6, 7, 8) are about the students' general perception of their ability or competence to acquire the subject matter that the school teaches in nine different areas of knowledge, interest in acquiring the subject matter in these areas, self-perceived degree of difficulty in studying during the last three school years in compulsory school, and perception of their ability or competence to cope with certain specific tasks in Swedish, English and mathematics.
  • Five questions (9, 16, 17, 18 and 19) are about working methods in the classroom and students' attitudes towards grades.
  • Six questions (10, 11, 12, 13, 14 and 15) concern parents' involvement in students' schoolwork, homework and access to and use of computers.
  • Two questions (20 and 21) concern students' experiences in certain specific school situations and general school well-being.
  • Two questions (27 and 28) concern leisure activities and work.
  • Two questions (29 and 30) are about the student's family relationship and number of siblings.

The exact wording of the questions can be found in the questionnaire for the students (pdf 2,5 MB in Swedish).

Collection in upper secondary school 3rd grade

The collection was handled by Statistics Sweden in Örebro and was designed as a paper questionnaire. The questionnaire together with a cover letter was sent to those who were still in the survey at this time. A total of 9493 people were invited to answer the survey.

The survey was conducted as a postal questionnaire with three reminders. Letters of thanks and reminders were sent to everyone after a week or two and two more reminder letters were sent out to those who did not answer the questionnaire at this time. In total, the survey was answered by 5,361 people (56.5% of the questionnaire sent out) and although the response rate is quite low, there are good opportunities to map the dropout rate because we have access to the school administrative data and also from the ones that answered the questionnaire in 9th grade. The questionnaires were registered via scan. A data file was prepared from the scanned material.

Details of the implementation and results from the data collection are described in reports that can be downloaded under the heading reports.

Questionnaire for students

The final survey included 51 questions and aimed to provide information on:

  • opinions about the education received after a couple of years in compulsory school
  • which upper secondary education has been chosen and which factors have guided this choice
  • if there have been study interruptions and the reasons for these
  • well-being in upper secondary school
  • views on upper secondary education
  • perceptions of one's own ability in various subjects
  • scope of gainful employment in addition to the studies
  • how much time was spent on different hobbies
  • plans for future education and profession

The exact wording of the questions can be found in the questionnaire for the students (pdf in Swedish).

Interactive variable list

The number of variables for each cohort varies from a few hundred to a couple of thousand. Some variables are present in all cohorts, others are only found in some. Some variables enable longitudinal studies, others only cross-sectional studies. ETF has developed an interactive list of variables to make it easier to search among the variables and see relationships between variables in different cohorts.

The variables are accessed through the links below. Questions about documentation are answered via e-mail to data controllers within ETF.

Selection principles

The sample design for cohort 1987 consisted of a stratified random two-step cluster selection based on the number of principal areas when the students attended 2nd grade in the 1995/1996 academic year. The municipalities were grouped according to population into six strata. Stratum one consists of the five largest municipalities and from the other five strata, six municipalities were drawn from each. A total of 35 municipalities were elected, which included 708 principal areas. The sample includes 222 principal areas. Selected principal areas were then completely examined with regard to the students in the intended year.