As part of the larger Eurostat project on Trusted Smart Statistics II, teams from Utrecht University (UU), Vrije Universiteit Brussel (VUB), and The National Institute of Statistics and Economic Studies (INSEE) are collaborating on the question of how data acquired from smart and traditional surveys can be integrated by official statistics institutions, and how we can analyze and deal with mode effects between smart, "clever", and traditional methodologies.

More explicitly, the goal is to establish whether there are large differences in time use and household spending depending on whether a smart survey or more traditional web or paper diary survey is being used. One key task is to separate sample selection differences from mode measurement effects and isolate the diffe rence between the datasets caused by the different measurement methods.

This task builds upon the general design of the field test that randomly assigns respondents to three modes (mobile app, web app, combination). Surveys using smart features will be compared with the data generated without involving smart features. The design differs between involved countries to allow for investigation of these differences for the purpose of estimation of pure mode effects (measurement effects), and for estimation of the size of a break in time series caused by a move from traditional to smart surveys.

This Github organization has been designed to facilitate collaboration on reports and scientific papers. Each deliverable will have an associated project and repository. A fourth repository for project management will contain agendas, plans and meeting notes that are non-project-specific.

Collaborators should be aware that documents and data shared in the repositories are publically available.

Shield:

This work is licensed under a Creative Commons Attribution 4.0 International License.