Our history

The first collections are carried out

In 1959, school director Mats Hultin and first actuary Klas Wallberg were commissioned by the National Board of Education (SÖ, known today as the National Agency for Education) to investigate the scope of school statistics when they were moved to Statistics Sweden (SCB). The purpose of this investigation was to collect data from one-tenth of pupils who attended 6th grade in elementary or unit school and compile nationwide individual statistics. The collection would take place again every five years and data from other central authorities was planned to be collected annually for follow-up using the individuals' social security numbers. The students in the sample would be born on 5, 15 and 25 of any month in 1948.

Prior to this investigation, Kjell Härnqvist, professor at the University of Gothenburg, emphasized that the collection of individual statistics could be even more valuable if basic information such as grades, school choice, cognitive tests and students' own answers to questions could be collected at the same time. This would mean a lot to researchers who needed to spend a lot of time retroactively collecting that data.

Wallberg and Hultin were positive to the proposal, but lacked the resources to carry out the tests. Therefore, at the request of Härnqvist, the Swedish Council for Social Research appointed a committee consisting of Kjell Härnqvist and professors Gösta Carlsson and Torsten Husén to plan and carry out the tests. The Committee proposed a supplement containing the following information: 

1. Standard test results in Swedish, mathematics and English.
2. Tests inspired by intelligence tests and containing tasks of a verbal, inductive and spatial nature. These would act as a complement to the standard tests, as all students would have completed the tests at the time of collection. 
3. Answers to questions about the student's hobbies, school motivation, home background, education plans, etc. 

The supplement was approved by the Swedish Council for Social Research, which financed the collection with SEK 25,000. Kjell Härnqvist, together with student Allan Svensson, designed the tests and organized the implementation. The work, which became known as the "Individual Statistics Project", took place at the University of Gothenburg and the first collection was carried out in May 1961.

Test booklets containing survey questions, information for teachers and principals, test booklets and documentation for compiling the standard test results were sent to approx. 1250 principal districts and included more than 12,000 students. 89% of the students (equivalent to 10,700) responded to the questionnaires and it was clear that the first collection had been a great success.

The processing of the collected material, where entire test booklets were registered manually and with careful checks, took approx. 1,600 hours and the data was transferred on punch cards. This success was made possible largely thanks to SÖ's support, which, together with the teachers' and school leaders' unions and Statistics Sweden, saw the importance of a longitudinal project for research in the school system. The funding from the Swedish Council for Social Research was certainly limited, but enabled Härnqvist and Svensson to set aside a significant part of their services to process the collections. With the help of a number of archive workers and students, the costs of this extensive collection were kept down.  

In 1966, the second collection was carried out with a cohort in 1953, which was equally extensive and with an equally high response rate. Prior to that collection, the individual statistics project was granted SEK 31,000 from the Swedish Council for Social Research. 

Collaboration with Stockholm University

The third collection was expected to be carried out in 1971. Due to limited resources, this was not possible, but also issues of the ethical maangement of individual data became increasingly relevant in Sweden which affected the work with the Individual Statistics Project. Laws for handling personal data needed to be introduced to protect the privacy of individuals, and longitudinal databases that required large amounts of data with regular updates were met with restrictive conditions to continue operating.

In the mid-1970s, the question of the project's continuation was raised by Statistics Sweden. Ingemar Emanuelsson, laboratory technician at the Educational Centre in Stockholm, was appointed as responsible for providing a design that would enable the project to proceed. Emanuelsson proposed that the project be transformed into a research project with collection of data from upcoming cohorts. The project was named "Evaluation Through Follow-up" (UGU) and thanks to an established, close collaboration with Stockholm University and the need for a coherent research project with individual data, we were able to continue our collections. 

SÖ granted funds for the collection and UGU was able to collect data on its third cohort in 1980.  

The financing of UGU over time

Over time, interest in longitudinal research increased among authorities and researchers in various fields. This resulted in more databases with a longitudinal design being added, and the Research Council Board (FRN - which is part of the Swedish Research Council today) was commissioned to finance the collection of material for the databases. UGU was granted annual grants between 1986 and 1990, which enabled improved organization and long-term planning of the work. Cohort 1977 could be carried out thanks to these grants, although to some extent as the grant was too limited to carry out collections in year 9.

In 1996, a committee (Committee for Longitudinal Research - KLF) was appointed by FRN that would be responsible for managing SEK 10 million / year among longitudinal research projects. The number of applicants was high and already five years later the budget had increased by another SEK 5 million per year. When FRN merged with a number of research councils to form the Swedish Research Council in 2001, the committee was closed down and responsibility for funding longitudinal research projects was assigned to the Committee for Infrastructure (KFI). This reorganization was due to VR's increased focus on databases other than the longitudinal ones. Longitudinal research databases were thus not prioritized to the same extent, and this was reflected in the increased difficulty of getting funds granted for our upcoming fundraising during the years 2005-2010, as competition rose. 

A longitudinal research project requires continuity with a consistent collection of material. The collection also consists of costly multi-step procedures, where external parties such as SCB, UGU's staff and logistical factors need to collaborate within fixed time frames. In order to collect data for a longitudinal project, we in the UGU group have been dependent on funding from the Swedish Research Council (VR) and the University of Gothenburg. In the years where funds were not granted or granted to some extent, we who worked with the fundraising had to eliminate certain surveys. This explains the gap in data collection between the 1958 and 1967 cohorts, as well as the variations in collections between the cohorts. 

How we work today

Today, UGU's working group consists of ten people who work to varying degrees with the management, planning and development of the study. About a dozen researchers in education, political science, psychology, economics and public health apply annually for our data to investigate developments in Swedish schools, economics and the labor market. You can read more about UGU publications in the ribbon to the left for inspiration.

Our primary focus is to work on improving the process for the collections that take place every three years. Through solid preparatory work, we can increase the response rate and knowledge of our activities. As part of increasing knowledge about us, we have created an Instagram and LinkedIn account where we share our work and have an opportunity to reach researchers, students and teachers who can benefit from the data we collect. 

We have also updated our codebook and increased the availability of our data for interested researchers. The codebook contains metadata that describes the properties and content of the data, which increases the searchability and reuseability of the data, in line with the FAIR-principles.