Competence Centres

In order to ensure the desired MEDem outcomes, it is envisioned that five centers will structure the work of MEDem. These are:

Center for Survey data (CSD)

Proposed location: FORS (lead), University of Lausanne and, at a later stage, University of Geneva

Centre for Textual Data (CTD)

Proposed location: Universities of Vienna (lead) and Amsterdam

Centre for Institutional Data (CID)

Proposed location: SciencesPo Paris

Centre for Methods and Standards (CMS)

Proposed location: University of Gothenburg

Centre for Data Archiving and Dissemination (CAD)

Proposed location: GESIS –Leibniz Institute for the Social Sciences

When building a consortium for the MEDem project, various aspects where taken into account, such as how to best run the project but also how to make sure that the MEDem work will have long-lasting positive effects for democracy research data collections by contributing to the establishment of a future MEDem decentralised infrastructure.

The institutional design of MEDem emerged from various meetings and discussions with projects and individual scholars over the course of the past three years. MEDem launched a call to potential hosts of competence centers in spring 2019. The applications were peer-reviewed by internationally prominent scholars and the most highly-rated institutions were proposed to host each one of the different centers. All these institutions have not just extensive research experience in the field of electoral democracies but also a large experience in running or contributing to social science research infrastructures.


The institutions hosting a competence centre will need to commit internal resources for the running of those centers. It is foreseen in the future that those institutions will continue to play a central role in building MEDem as an infrastructure.​


Three of the five competence centers are focused on specific data types: survey data (CSD), textual data (CTD) and institutional data (CID),[1] which are the data types used in research regarding electoral democracies. Building the competence centers around data types helps disentangling the specific challenges posed by each data type regarding methodology for data integration.

For survey data the key challenges involve measurement invariance and scale comparability as well as data representativeness. The typical challenges are whether questions and scales indeed capture the underlying concept, whether they do so in a comparable way across counties, and whether they are representative of the wider population.

For textual data, in addition to operationalisation and measurement issues, the challenges lie in what information to use, how to access that information and how to translate text into meaningful codes that can be used in comparative research across cultural and language boundaries.

For institutional data the challenges are to identify the relevant institutional and contextual dimensions and then to collect information that sufficiently specifies this context in terms that are theoretically relevant for democracy research.

Two further competence centers focus on two specific cross-cutting functions, one on methods and standards and one on data archiving and dissemination.

The competence center for methods and standards (CMS) will be charged with keeping an overview across data types and making sure that data can be linked across data types. The CMS will also steer the scientific process in developing standards and in ensuring that those standards are accepted by the scientific community. The development of agreed standards is critical for comparative research and lies at the heart of MEDem's harmonization ambitions.

The data archiving and dissemination competence center (CAD) will be in charge of creating a single point of open access to MEDem data. It will play a critical role in enforcing MEDem-agreed standards as well as in ensuring data protection and curating standards. It will be responsible for providing syntax files and documentation facilitating access to the complex data structure produced by MEDem.


[1] Institutional data is also referred to as "enumeration" data as it comprises mainly public record data enumerating numbers of votes, parties, candidates and their histories of formation, mergers and splits as well as more strictly institutional factors such as electoral systems and policy-making arrangements. "Contextual" is another adjective describing this data type.