*5.2. Data Collection*

This study focuses on the timeframe of 1 January 2020-1 November 2020.This timeframe is related to our specific focus on the impact of COVID-19 on discourses about migration. Although COVID-19 spread before 2020, the implications of this for The Netherlands were first discussed around January 1st. In the context of this article, under 'political discourse' we understand the text and talk of professional politicians or political institutions, such as presidents and prime ministers and other members of government, parliament or political parties, both at the local, national and international levels to include both the speaker and the audience. We chose to combine an analysis of political discourse with an analysis of media discourse, as political discourse is increasingly mediatized and media discourse increasingly politicized. To grasp public and political discourse, we have respectively focused on national and local media as well as parliamentary and governmental debates in The Netherlands. The choice of parliamentary debates and reports, and of news articles, is not based on a reductive understanding of political and public discourse; in contrast, it is inherent in discourse analysis to work with sources that, while indicative of the bigger picture, are not fully representative. In addition, it should be noted that although we now seem to sugges<sup>t</sup> a strict separation between media and political discourse, as already mentioned, in reality these discourses are intertwined.

All the sources were open access available; for the parliamentary and governmental debates in The Netherlands the national database www.officielebekendmakingen.nl was consulted, a database that gives access to the transcripts of all governmental, parliamentary and other debates by political institutional actors and committees. Within the scope of our timeframe, we have collected and stored every document that mentions COVID-19 (or any variation thereof, like pandemic or corona). This led to a selection of 564 documents. These documents have been checked manually to see if they contained one or more of the following search terms: (labor) (im)migration, (labor) (im)migrant(s), alien(s), border(s), asylum(seeker; process; application), refugee(s), healthcare, crime and integration. This led to a final selection of 137 documents for the political discourse (all of which thus discuss both COVID-19 and one or more of the aforementioned search terms).

The sources for the media analysis were obtained through the NexisUni news database that is freely accessible through our University Library. In the selection of national newspapers and magazines we have paid attention to political diversity, and we have prioritized larger media platforms; for local newspapers, we have included all available newspapers within our timeframe7. Obviously, this approach leaves out an important domain where discourse happens nowadays, that is, social media. However, for this article we chose to limit ourselves to official media resources. For the media discourse the initial collection of media outings counted 408 and, after a similar check as described for the political discourse, we included a total of 84 articles.
