Text: Kaisa Järvelä
THL Wants to Allow Efficient Use of its Data
The National Institute for Health and Welfare (THL) has recently drafted a data policy and is now building a one-stop service for processing access authorisations and the delivery of their data. THL also aims to utilise the Finnish Social Science Data Archive more actively than before.
THL wants to bring efficiency to the use of its hundreds of datasets. This objective has been included in the institute's data policy, drawn up for the years of 2016–2020. The institute has already begun active work to achieve the objective.
For one thing, the efficient use of data means that THL strives to move away from the collection of overlapping data. However, this is not enough – THL wants both national and international use of its datasets to be as extensive as possible.
Increasing reuse requires sufficient metadata
THL has begun to produce descriptions of its existing datasets, which will make it possible for users to find them.
'December will see the publication of a catalogue that covers one hundred datasets', Head of Unit Risto Kaikkonen from THL's Information Resource Services states.
THL, Statistics Finland and the FSD launched a project in the autumn aimed at developing the descriptive metadata through national cooperation. Funding for the project came from Sitra, the Finnish Innovation Fund. The project also includes the publication of a shared data catalogue that will be made publicly available online.
The systematic description of datasets and metadata publication are to be developed into an established practice at THL in the future.
Different types of datasets into one-stop service
THL data include register data, survey data and biological samples archived at the institute's own biobank. The processing of different types of data is governed by different legislation.
'Processing access authorisations and data delivery has been fragmentary thus far', Kaikkonen admits.
To remedy the situation, the intention is to make all data available for reuse through the same service. This solution would significantly facilitate the work of researchers who need to use both survey and register data.
Anonymised datasets would be available at Aila
At present, the FSD's Data Service Portal Aila features 38 datasets from THL. Despite the one-stop service model currently under development, the number of datasets archived at Aila is expected to increase in the future.
'Aila is an excellent service for those who use THL datasets for activities such as teaching, studying or writing their thesis', Kaikkonen praises.
The idea is that users for whom anonymised survey data are sufficient could download at least the most popular datasets directly from Aila. However, if researchers need to use more extensive datasets or want to combine survey data with, for instance, register data, they could turn directly to THL's own service.
'The anonymised data archived at the FSD are sufficient for a certain user group, but other users may require more detailed data. Cooperation will allow us to easily serve both user groups', Kaikkonen envisions.
It is likely that some kind of an environment for ensuring the secure processing of data will be developed to complement THL's service.
'I do not want to restrict the discussion to the currently available remote access systems because a future solution could be an entirely novel innovation.'
The upcoming service model can be expected to be operative in around a year. The plans concerning the one-stop service should enter the implementation stage by the end of 2017.
