Design and Measurement Challenges in Researching Migrants 3 |
|
| Session Organisers |
Dr Agata Górny (University of Warsaw) Dr Joanna Napierała (University of Warsaw) Dr Justyna Salamońska (University of Warsaw) |
| Time | Thursday 18th July, 14:00 - 15:30 |
| Room | D09 |
Growth in international mobility has been one of the reasons for the proliferation of research on the migrants. Yet, migrants are populations that pose particular challenges when being surveyed and administrative information about this group is scarce. Difficulties in studying mobile populations can occur at various stages of research design and fieldwork. In our session we would like to focus on the difficulties related to the design and measurement problems. In particular, we propose to map these challenges and provide a forum for a discussion on the ways to tackled them in practice. We would like to focus on challenges related e.g. in designing a longitudinal survey on migrants, to measuring some important migration-specific variables, such as migration events (e.g. trips), ethnic identification indicators, sensitive issues relating to migration (incomes, remittances, legal status), and others. These measurement problems augment in cross-national surveys, essential approach in studies on migrants, that are conducted in differentiated cultural, linguistic and institutional contexts. Additionally, survey traditions and specific answer scales that differ across countries make the harmonisation of data difficult. The measurement challenges intersect with modes of data collection. While majority of surveys on migrant groups are still conducted with the help of face to face interviews (PAPI or CAPI), web-based surveys are reaching in importance in these studies. An opportunity to reach mobile persons via internet opens up a new chapter in migration studies and creates an environment in which surveys can play a more important role in creation of data on migrants in cross-national perspective. Debate on the role of social media and usage of big data in this process paves the way for the future of surveying migrant groups. We invite migration scholars to contribute to the discussion about the future of the migration survey research by sharing their experiences in surveying those groups during our session.
Keywords: migrants, mobile populations, web-based surveys, longitudinal surveys, measurement in surveys
Professor Jochen Mayerl (Chemnitz University of Technology)
Mr Manuel Holz (Chemnitz University of Technology) - Presenting Author
Measurement problems in terms of different understanding of the meaning, concepts and implications of survey questions as well as tendencies to different response-sets (e.g. acquiescence or social desirability) are reoccurring problems in comparative migration research. Cultural factors, time-varying effects and survey design lead to response biases. It could be shown that depending on the region of origin, migrants tend to extreme or centred responding behaviour. Macroscale changes in attitudes towards disclosure of personal information influence acquiescent biases. Therefore, the question arises whether the presence of measurement equivalence can be assumed, especially within instruments regarding subjective feelings, when comparing migrant and native populations.
The aim of the study is to test for intercultural and longitudinal equivalence of the Health Short Form 12 - items scale (SF -12), trying to answer the question whether this instrument is useful to compare health of migrants and native Germans over time. The study contributes to the ongoing discussion on measurement equivalence of the SF -12 by including variables concerning integration and assimilation, employment, family status and varying types of migration and years since migration to the analysis.
Using a Multiple Group Structural Equation Modelling approach analyzing longitudinal data of the German Socioeconomic Panel (SOEP), we test the validity and measurement equivalence (configural, metric and scalar invariance) of the latent construct ‘health’ over time and between (non-)migration groups. In addition, different item functioning can be detected for a series of covariates. The study concludes in recommending a reduced version of the SF-12 to improve the avoidance of measurement artefacts when comparing the health of migrants and native Germans.
Dr Jean Philippe Pierre Décieux (University of Duisburg-Essen) - Presenting Author
Dr Nils Witte (Federal Institute for Population Research (BiB))
Dr Andreas Ette (Federal Institute for Population Research (BiB))
Professor Marcel Erlinghagen (University of Duisburg-Essen)
Mr Jean Guedes Auditor (Federal Institute for Population Research (BiB))
Dr Nikola Sander (Federal Institute for Population Research (BiB))
Professor Norbert F. Schneider (Federal Institute for Population Research (BiB))
With the German Emigration and Remigration Panel Study (GERPS) we have recently started to establish a new and unique longitudinal data set to investigate consequences of international migration from a life course perspective. This task is challenging, as internationally mobile individuals are hard to survey for different reasons (e.g. sampling design and approach, contact strategy, panel maintenance). In our presentation, we would like to report experiences we have made and how we handled with methodical problems.
GERPS is funded by the German Research Foundation (DFG) and surveys international mobile German citizens (recently emigrated abroad or recently re-migrated to Germany) in four consecutive waves within a panel design. Based on a probability sample, GERPS elucidates the individual consequences of cross-border mobility and concentrates on representative longitudinal individual data. It bases mainly on online surveys and covers topics like changing living conditions or life events linked to the migration process.
Our first wave was conducted in November 2018 and the second wave will be started in April 2019. This enables us to present first hand results and experiences of the first two waves of this new and unique longitudinal dataset, which focuses on international migrants. We will mainly reflect the effectiveness of our register-based sampling focusing on traditional indicators of data quality such as response rates or attrition rates. Additionally, we will present our results concerning the success of different incentive approaches (lottery, pre-paid and post-paid incentives) and contact strategies, the used devices to run our online responsive online panel and field experiences we made (e.g. field time and fieldwork during this waves). These results can be very helpful to international researchers in the context of surveying mobile populations.
Dr Pilar Rioseco (Australian Institute of Family Studies)
Dr Galina Daraganova (Australian Institute of Family Studies) - Presenting Author
Mr John De Maio (Australian Institute of Family Studies)
Building a New Life in Australia (BNLA) is a ground-breaking longitudinal study that aims to understand the settlement experiences of humanitarian migrants in Australia. More than 1,500 individuals and their families (n=2,399 in total) who had been granted a permanent humanitarian visa to live in Australia – including onshore and offshore applications - participated in the first interview in 2013. Five annual waves of data have been collected between 2013 and 2018, including face-to-face (Waves 1,3 and 5) and telephone interviews (Waves 2 and 4). Interview materials were translated into multiple languages.
This paper describes some of the key learnings identified around the factors associated with survey participation and non-response for this population. First, we examine non-response by interview mode. In particular, we investigate whether participants who responded to telephone interviews only (after Wave 1) differ from those who responded to both telephone and face-to-face data collection methods in their socio-demographic characteristic (pre-migration and at the last completed wave), using logistic regression models. We also examine the role of interviewer characteristics. Second, the paper presents a descriptive analysis of item non-response and differences in rates of item non-response by country of birth.
Results show that, after controlling for relevant covariates, socio-demographic characteristics are significantly associated with mode non-response, including pre-migration factors (migration pathway, paid work and pre-migration English language proficiency) and characteristics at the last completed interview (paid work, improvements in English, location and having a new interviewer). Regarding item non-response, significant differences were observed by country of birth in mental health items (e.g., post-traumatic stress scale and sources of stress), as well as in financial items (e.g., government payments and main source of income), trust (in community, police, etc) and relationship with partner.