Frequently asked questions

You need to be a doctoral student or disputed researcher and have a research plan that is related to our data. You also need to be affiliated with a higher education institution that complies with the GDPR regulation. 

You apply by submitting a completed user agreement, research plan and a tailor-made list of variables to ugu@ped.gu.se. The user agreement and the variable list can be found here.

Delivery to doctoral students is free of charge. However, there is an entry fee of 5000 SEK for researchers. The cost covers the processing of the application and the compilation of data, and initial datasupport.

We always recommend that you fill out an ethics application for your project. However, we have a question "Think about how it has been both inside and outside of school!" which may include sensitive personal data. If you request this survey question, an approval from the Ethics Board will be required.

Unfortunately not. Neither surveys nor register data are distributed in their entirety. This is because our security classification means that data is handled in a safe and secure manner. In order to ensure this, we want you to carefully select your variables in our variable list. This selection should reflect the submitted research plan.

To facilitate the choice of survey questions, we have produced reports that thematically illustrate our survey questions. These can be read under "Reports" for each cohort. The list of variables can be found in our codebooks.

It is therefore not possible to access entire cohorts or surveys.

We mainly refer to Statistics Sweden's non-response analyses for information on calibration weights for each cohort. 

In general, the variables "gender, guardian's country of birth, principal, grades, test results and school size" are used for weighting at Statistics Sweden before data is submitted to us. We lack background variables that characterize the guardian's country of birth, but we do have categorised variables on the basis of continents.

We recommend that you enable the weight variables where available, but for some cohorts (as of 1992) they are crucial and should be used.

If you want to analyze the cognitive tests/survey responses for grade 6 (or grade 9, 10 and 12), you should use Normweight (which counts the number of respondents) or QS6weight (which counts the Swedish population).

If you want to analyze school administrative data or grades, use RS6weight (which counts the number of students remaining in the sample at grade 6).

When analyzing the parent survey, you should use Qp6weight, which counts up to the Swedish population.