Fieldwork

According to the ESPAD handbook, data collection should be performed during spring. With the exception of three countries, data were collected between February and June 2015, with a majority of data-collection exercises conducted in the March-May period (Table G). For pragmatic reasons, the Belgian survey was carried out half a year in advance, in October-December 2014, while Georgia and the Netherlands collected data during the late autumn of 2015.

The ESPAD guidelines contain no rules as to whether teachers or research assistants should be responsible for data collection in the classrooms. Instead, the recommendation was to use the category of survey leaders whom the students trusted the most. In about half of the countries, teachers or other school staff administered the data collection, while research assistants did so in the other half (Table G).

Table G. Characteristics of the data collection. ESPAD 2015

To stress the anonymity and confidentiality of the survey, the handbook recommended the use of individual envelopes for each student to put his/her questionnaire in and then seal. Individual envelopes were used in about three quarters of the countries (Table G). In the remaining countries, other measures were taken which were judged to fulfil the same purpose. Examples include the use of large class envelopes, which were sealed in front of the students, or a closed box into which the students put their forms.

In Austria, Latvia, Liechtenstein and the Netherlands no such precautionary measures were needed since the traditional data-collection mode of pen and paper had been abandoned. Instead, web surveys were performed in those four countries. This differs in relation to prior ESPAD data collections in those countries, and also differs in relation to the other countries collecting data in the traditional way in 2015. There are of course several advantages to online surveys, such as cost-effectiveness and more rapid data collection. From the ESPAD perspective, one general disadvantage could however be that the changed datacollection mode may give rise to questions of comparability.

For Latvia, the online data collection seem to have not functioned so well. Technical problems resulted in a large proportion of uncompleted questionnaires. Since students were not assigned unique individual codes, it appears that some had entered the survey several times and there were also instances noted of students that had accessed the questionnaire outside school hours. Although efforts were made to identify and remove such invalid cases, this still leaves doubts regarding the implementation of the Latvian fieldwork.