Sampling Design for the GUIDE Study

To initiate the longitudinal study GUIDE (Growing Up in Digital Europe: EuroCohort), which aims to provide a reliable foundation for analysis, the sample design must be implemented to the highest possible standards. For each country and cohort, a sampling method should be selected that ensures the best possible coverage of the target population. However, procedures vary across countries, and other factors that may influence sample design, such as population density and the homogeneity of schools or local areas, also differ. Consequently, it will not be possible to apply the same sample design in every country.

The first wave of fieldwork in cohort 1 will begin in September 2028. This will be the eight-year-old cohort: individuals born between 1 September 2019 and 31 August 2020, regardless of ethnicity, citizenship or language, residing in the country. This includes children whose families were outside the country for less than six months (e.g., on holiday or working abroad). It excludes children whose families have been or will be outside the country for six months or more, as well as short-term visitors staying in the country for less than six months.

Attrition rates will inevitably vary across cohorts and countries over the 25-year duration of the study. Therefore, when determining the initial target sample size for each cohort, anticipated attrition must be taken into account to ensure that a sufficient number of respondents remain in later waves to support statistical analysis. GUIDE recommends that the minimum effective sample size in the first wave should reflect the country’s population size, using the number of births in 2019 as a reference. For the first wave of the age-8 cohort, the recommended effective sample size is 8,000 for large countries (more than 80,000 births in 2019), 4,000 for medium-sized countries (more than 10,000 births in 2019), and a full population survey for small countries. In Latvia, 18,786 children were born in 2019; therefore, the target sample size for the first wave is 4,000.

Sample design may vary across countries and cohorts to reflect national circumstances while adhering to common principles and parameters. Some countries may choose to use clustered sampling, in which sampled children are grouped within geographical areas identified in the sampling frame or within schools. This approach can reduce the costs of data collection, particularly in the first wave when face-to-face interviewing is expected to predominate. However, clustering generally reduces the precision of estimates. As a result, a larger sample size will be required, compared with a non-clustered sample, to meet the GUIDE precision requirements.

Taking into account the strategy used by the Survey of Health, Ageing and Retirement in Europe (SHARE) and the European Social Survey (ESS), GUIDE is developing written guidelines for national research teams. These guidelines are prepared by members of the GUIDE Central Survey Team, who provide support and advice to national teams in the development of their sample designs. Each national sample design must be approved by the GUIDE Central Survey Team before the survey is launched.

This initiative is implemented within the framework of ERDF project No 1.1.1.5/3/24/I/003 "Support for the Participation of Daugavpils University in the Horizon Europe programme".