The number of people born outside of Sweden as a share of the Swedish population increased since 2010. That year, 1.38 million of the country's inhabitants were born outside of Sweden, whereas this number had increased to 2.17 million by 2023. In other words, foreign-born citizens made up around 20 percent of the population in Sweden in 2023. Of the 2.17 million people born outside of Sweden, the highest number came from Syria.

The city of Amsterdam is divided up into 7 city districts, 22 city areas, 99 neighbourhoods and 481 smaller units. The neighbourhood level (which the municipality calls ‘wijken’) was used in the sampling process. Out of the 99 neighbourhoods, 12 have less than 2.000 inhabitants and were therefore excluded from the sampling. Out of the remaining 87 neighbourhoods, 42 are majority minority neighbourhoods (MMN), which were all included in the sampling. The municipal register (Basisregistratie Personen, BRP) was used to conduct the sampling. This was done by the research department of Amsterdam (OIS) on the basis of the composition of the population as of March/ April 2019. In total a sample of 2.000 inhabitants was selected with an equal number of persons per majority minority neighbourhood. It is important to note that only one person per independent domicile could be selected. To ensure that only individuals, who fit the BaM project definition of ‘no migration background’ participate in the study, at the beginning of the survey in all cities all respondents were screened by asking about their birthplace and the birthplace of their parents. Those who did not fit the study’s target group were excluded from the remainder of the questionnaire. 1.2. Rotterdam Rotterdam, the other Dutch city in the BaM survey, is divided into 15 areas and 86 neighbourhoods. Here too the neighbourhood level was used in the sampling process. Out of the 86 neighbourhoods, 23 have less than 2.000 inhabitants and were therefore excluded from the sampling. Out of the remaining 63 neighbourhoods, 35 are sampled in the BAM survey as MMN. The municipal register (Basisregistratie Personen, BRP) was used to conduct the sampling. This was done by the research department of Rotterdam (OBI) on the basis of the composition of the population as of March/ April 2019. In total a sample of 2.000 inhabitants was selected with an equal number of persons per neighbourhood. Again, one person per independent domicile could be selected. 1.3. Antwerp Antwerp is divided into 9 districts, 70 neighbourhoods and 298 smaller units. The neighbourhood level was used in the sampling process. Out of the 70 neighbourhoods, 14 have less than 2.000 inhabitants and were therefore excluded from the sampling. Out of the remaining 56 neighbourhoods, 27 are sampled as MMN. The city of Antwerp does not provide data on the country of birth. The city however does provide figures on the nationality of the person, or on their first nationality and the nationality or the parents. This is done as in the following order: 1. First nationality of the father; 2. First nationality of the mother; 3. First nationality of the person; 4. Current nationality of the person. If all four nationalities are Belgian, the person is labelled as having no migration background (in Flemish ‘autochtoon’). As this definition might lead to an underestimation of the population with migration background3, we considered neighbourhoods where more than 45% (rather than 50%) of the population is defined as ‘allochtoon’ as MMN in Antwerp and also included them in the sampling. Data from the Belgian postal company Bpost was used to conduct the sampling. This was done by the fieldwork bureau DESAN. In total a sample of 3.000 inhabitants was selected. DESAN acquired lists of names, addresses and age of people living in the selected neighbourhoods from the Belgian postal company Bpost. It then applied onomastic sampling (i.e. they selected Belgian sounding names) to the file obtained to isolate persons without migration background. DESAN has extensive experience with this and estimates that onomastic method applied is valid for 90% of the records. The selected respondents were first screened if they belonged to our target group. 1.4. Malmö Malmö is divided into 5 city areas, 10 city parts, and 135 subareas (neighbourhoods). The subarea (neighbourhood) level was used in the sampling process since it was more comparable to the area levels in the other cities. Out of the 135 neighbourhoods, 73 had less than 2000 inhabitants and were therefore excluded from the sampling. Out of the remaining 62 neighbourhoods, 33 are MMN and are sampled. It is important to note that the Swedish definition of migration background deviates from the BaM one, i.e. in Sweden a person has a migration background if they were born abroad or had two parents that were born abroad (compared to the definition used in the BaM project, in which a person is considered to have a migration background if they were born abroad or at least one of their parents were born abroad). We based the neighbourhood selection on the Swedish definition, as that was the available data. As this definition might lead to an underestimation of the population with migration background, we also considered neighbourhoods where more than 40% (rather than 50%) of the population is defined as having a migration background as MMN in Malmö. Data collection provider Norstat used data provided by Itesco to conduct the sampling. In total a sample of 8500 inhabitants was selected in two sampling rounds. The sample file contained the following information: name, address, age and neighbourhood. Initially Norstat purchased a sample from a database provided by Itesco that covers addresses and phone numbers of about 80% of the Swedish population. Using this database, Norstat first conducted onomastic sampling (i.e. they selected Swedish sounding names) and then selected individuals between 25 and 45 years of age. The first sampling round only considered individuals living in neighbourhoods that in 2017 (i) had more than 2000 inhabitants, and (ii) had more than 50% of inhabitants with a migration background (according to the official Swedish definition). This step resulted in the selection of 5000 names in 24 neighbourhoods. Given that Sweden uses a less strict definition of migration background than we do in the BaM project (see neighborhood selection) which might result in an underestimation of the population with a migration background, we also initiated a second sampling round in September 2019. We did this in neighborhoods in which 40-49% of the inhabitants had a migration background according to the official Swedish definition (an additional 9 neighborhoods). The same sampling procedure as during the first round was used, and approximately 3500 people were selected. To ensure that only individuals, who fit the BaM project definition of ‘no migration background’ participate in the study, Norstat screened the respondents at the beginning of the survey by asking about their birthplace and the birthplace of their parents. Those who did not fit the study’s target group were excluded from the remainder of the questionnaire. This strategy is in fact applied to all the fieldworks we have conducted in other cities.