All time references are in CEST
Sampling migrants in general population surveys |
|
| Session Organisers | Dr Francesco Molteni (University of Milan) Dr Riccardo Ladini (University of Milan) Professor Ferruccio Biolcati (University of Milan) |
| Time | Tuesday 18 July, 09:00 - 10:30 |
| Room | |
Sampling migrant groups is one of the biggest challenges in survey research. This challenge has been addressed by utilizing both probabilistic and non-probabilistic research designs that focus on specific sub-populations, often with a local emphasis. With very few notable exceptions (e.g., the BHPS and the SOEP), a significant gap in studying migrant populations with survey methods is the lack of comparison with the native population. This comparison is crucial for understanding the differences and similarities between the groups.
To address this gap, one main approach has been to refer to subsamples of people of foreign origin within general population survey samples. While this strategy has several advantages, it also presents several drawbacks. These include the low number of such populations, potential selection bias leading to the under-representation of specific migrant groups, and the variety of sampling frames, which often depend on the different availability and composition of population lists across contexts. This issue is particularly relevant in comparative research.
Building on these considerations, this session welcomes papers that focus on strategies for including subsamples of migrants in general population samples, as well as strategies designed for studying migrant populations that can be easily transferred to designs targeting the general population. We invite contributions that address specific techniques (e.g., oversampling techniques, onomastic procedures, focused enumeration), those that offer methodological insights concerning issues related to population coverage, lists, different response rates between natives and migrants, and spatial segregation, and those aimed at assessing and discussing the quality of subsamples of migrants in international and national population surveys.
Keywords: migrants, natives, general population surveys
Professor Antje Roeder (Philipps-Universität Marburg) - Presenting Author
Dr Stephanie Müssig (FAU EZIRE, Friedrich-Alexander-Universität Erlangen-Nürnberg)
Using general population surveys for research on immigrants and their children is getting increasingly popular, a development driven by a growing public and political demand for knowledge on persons with migration history. Some surveys, such as the European Social Survey, a high quality biannual survey conducted in 28+ countries since 2001/2, have responded to this by including more detailed measurements of migration background over time. By pooling rounds and countries, ESS data enable researchers to obtaining sufficient cases of persons with migration history to tackle research questions with demanding analytical strategies, such as multiple-origin and destination comparisons.
However, an initial review of more than hundred studies that have used the ESS for this purpose illustrates that researchers rarely address that the ESS has not been not designed as an immigrant survey and reflect the consequences this may have on their results. Of the studies we have reviewed to date, only very few compare share and composition of the immigrants in the ESS with administrative data, such as census data. What follows from this is that it is unclear for which migrant groups survey data, such as those of the ESS, speaks for and who may be excluded or under-represented.
To address this gap in knowledge, we compare data from general population surveys used for immigrant research, with data from immigrant-specific surveys and census data for several European countries where relevant data sets are available (e.g. Germany, UK). Additionally, in our contribution we discuss reasons for and possible consequences of bias across data sources.
Professor Ferruccio Biolcati (University of Milan) - Presenting Author
Dr Francesco Molteni (University of Milan)
Dr Riccardo Ladini (University of Milan)
The use of population lists is the golden standard for obtaining representative probabilistic samples of immigrant subpopulations. However, those lists are not always available to survey researchers. This is the case of Italy, where the use of population lists is not allowed for institutions (e.g. universities) and survey programs (e.g. EVS, ISSP) not adhering to the official statistical system. In these situations, the suboptimal strategy consists in the use of electoral lists, which do not include non-Italian citizens.
To overcome this issue and minimize the resulting bias, we propose a strategy for the inclusion of non-Italian citizens, first employed in the CoValues survey carried out in Lombardy in late 2022. Aimed at analysing the native-migrant divide in values, the survey employs the same probabilistic sampling design of the Italian edition of the European Values Study to collect interviews among the Italian citizens. To sample non-Italian citizens, it employed a focused enumeration strategy, namely, a spatial sampling design driven by respondents themselves. Each respondent was asked to name the three nearest addresses where he/she thoughts people with a foreign origin were living. This strategy allows a first screening of the immigrant population, by providing to interviewers addresses where to plausibly find non-Italian citizens.
Starting from this experience, this contribution aims at analysing the pros and cons of this strategy, by evaluating its effectiveness and comparing the distribution of the immigrant sample with the distribution of the immigrant populations according to the more relevant characteristics.
Ms Tanja Stojadinović (Ipsos UK) - Presenting Author
In the recent years Ipsos have implemented various random probability surveys across Europe targeting immigrants and descendants of immigrants (from North Africa, Sub-Saharan Africa, Syria and Turkey), refugees (from Ukraine) and ethnic minority populations (Roma and Travellers). In these surveys the choice of sampling method predominantly depended on available sampling sources.
In this presentation, we discuss the available sampling sources for targeting these hard-to-reach populations. In most cases these are the same sources that are used for general population surveys. We also explain how the sampling sources can be constructed through thorough mapping of areas where the target population live, when reliable sources are not available. And we talk about approximations that were used to estimate concentration levels when available data did not align with the definition of the target groups, or the sources were not fully reliable, and how these compare to the eligibility rates recorded during data collection.
We provide details on how the availability of sampling sources directed choices of the random-probability sampling approaches and modes of data collection:
• Unclustered single-stage samples, with direct sampling of the target population, for an online push-to-web survey;
• Multi-stage clustered samples, with an onomastic procedure used to identify the target group members, for an online push-to-web survey;
• Multi-stage clustered samples, with screening for eligibility, for a face-to-face survey. These included methods for improving efficiency: oversampling areas of higher concentrations of the target populations, focused enumeration and adaptive cluster sampling;
• Location sampling, with screening for eligibility, for a face-to-face survey.
We discuss the sample designs, the target populations coverage they provide, the challenges of implementing them, and the final sample efficiency.
Mr Carsten Broich (Sample Solutions BV)
Ms Nadica Stankovikj (Sample Solutions BV) - Presenting Author
The study focuses on a survey studying everyday life experiences of the Muslim population, which although represents one of the largest minority religions in the Netherlands, is defined by only 6% of the total population. Given this context, traditional sampling methods often struggle to effectively reach and represent religious minority groups such as the Muslim community, which is diverse in origin and background. To address these challenges, an onomastic approach was used, where probability based sample was enriched with auxiliary data from social media platforms. By analyzing names and surnames gathered from publicly available sources, the country of origin has been inferred, focusing on countries with higher percentages of Muslim populations. The main sample consisted of panelists from Lifepanel, a probability panel, where browser language data has been analysed to identify potential Muslim respondents, for example whether the language is spoken in a predominantly Muslim country.
The survey was conducted using two contact modes: Computer-Assisted Telephone Interviewing (CATI) and Computer-Assisted Web Interviewing (CAWI), allowing for a direct comparison of response and completion rates between the two contact modes. The main objective of this study is to assess the effectiveness of these methods in engaging the target population, while also evaluating the onomastic and digital approaches’ potential for identifying Muslim individuals within a predominantly non-religious population.
Data were collected over a two-month period, with a sample size designed to reflect the demographic diversity of the Muslim population in the Netherlands. Various challenges encountered during the fieldwork are presented, such as high ineligibility rate and overrepresentation of highly educated population. Early findings indicate notable differences in completion rates between CATI and CAWI, with distinct engagement patterns observed across the two methods.
Dr Andrea Turković (University of Milan) - Presenting Author
Dr Francesco Molteni (University of Milan)
Dr Livia Elisa Ortensi (University of Bologna)
Dr Giorgia Papavero (ISMU Foundation )
A significant gap in studying migrant populations through survey methods lies in the lack of direct comparisons between native and migrant populations, despite efforts by a few notable exceptions like the BHPS and SOEP. Such comparisons are crucial for understanding cultural, economic, and social differences and similarities between these groups, as well as for highlighting the potential dynamics of integration and assimilation. However, conducting such comparative research involves several methodological challenges, including constructing comparable population lists, navigating language barriers during interviews, addressing differing response rates, and ensuring that survey questions meet cross-cultural validity criteria.
In this paper, the sampling and administration strategy, along with preliminary results from the "Between Origin and Destination" (BOnD) survey, a collaborative project by the Universities of Milan, Bologna and the ISMU Foundation, are outlined. The aim of the survey is to explore the relationship between cultural assimilation and the socioeconomic integration of migrants in Lombardy (Italy), with a particular focus on women. The study targets three specific yet comparable samples: native Italians, first-generation migrants, and second-generation migrants. To effectively capture a comprehensive and representative sample of these groups, the Centre Sampling Method, an effective technique for reaching hard-to-survey populations, was employed. This approach allowed the inclusion of both registered and unregistered migrants by selecting participants from "aggregation centres" such as commercial, public, and religious spaces, as well as administrative offices. A total of 2,800 face-to-face interviews, evenly distributed among the three groups, are being conducted by trained interviewers, often serving as cultural mediators, to ensure comprehensive coverage of both urban and peripheral areas.