Humboldt-Universität zu Berlin - TRANSMIT

Humboldt-Universität zu Berlin | Projects | TRANSMIT | Data | TRANSMIT Data Explorer

TRANSMIT Data Explorer


The TRANSMIT Data Explorer explores our standardized survey data collections in Lebanon, Turkey, Nigeria, Senegal and the Gambia. For the Middle East data collections, the data shown are drawn from the first (Turkey) and second (Lebanon) wave of the “TRANSMIT Surveys of Syrian Nationals and Their Neighbors in Lebanon and Turkey (2019-2023)” collected in face-to-face interviews between November 2020 and April 2021. Sampling for all surveys was conducted via stratified random sampling (for details, see below). [1]

The Middle East data collections aim to provide the empirical foundation for a transnational perspective on a key migration route to Europe via large-scale surveys among Syrians and their host communities in two major countries of mixed-migration contexts, namely Lebanon and Turkey. Both countries, in addition to their own history of emigration, host large refugee populations. The surveys cover both migration biographies and aspects of the socioeconomic participation of Syrian nationals in Lebanon and Turkey. Surveying the host populations in equal terms allows researchers to examine local contexts and group differences. The survey data are available as SUF. For details, please see below [2] [3].  

For the West Africa route, the data shown here are drawn from the first and second waves of face-to-face surveys conducted in Senegal, the Gambia, and Nigeria. Data collection for the first wave occurred between March and April 2019 in the Gambia, between October and December 2019 in Senegal, and in April 2021 in Nigeria. Data for the second wave in Senegal and the Gambia were collected between November 2021 and January 2022 and in July 2023 in Nigeria.

The surveys in Senegal and the Gambia targeted the potential migrant population and are representative of the population aged 15 to 35 for the whole country for the case of Gambia and the regions of Dakar and Casamance for the case of Senegal.  For the case of Nigeria, the focus of the surveys was both potential migrants as well as the communities they are embedded in, such that the samples were not restricted by age. The two surveys in Nigeria are representative of the populations of the states of Edo and Kaduna in the first wave and of the state of Edo in the second wave.

The West Africa route surveys yield information about family structures, demographic characteristics, migration experiences and migration aspirations, and economic and social well-being as well as information about experiences of climate change, discrimination, gender inequality, and violence.


Below you can see an overview of the data obtained in our web tool [switch to full screen version].

 

Use ☰ to choose the diffrent outcome variables available.

 

IFrame: https://transmit-dezim.github.io/transmit-data-explorer/

 

Baseline Sampling


Middle East

The baseline survey in Lebanon took place in 2019 and in Turkey in 2020. The target population for the TRANSMIT Surveys included individuals aged 15 years or older who were residing in either Lebanon or Turkey at the time of the surveys and who were residing in private housing (excluding individuals residing in formal refugee camps, detention centers and barracks). The survey was stratified according to the country of birth of the head of household, such that 50 percent of respondents were from Syrian households and 50 percent were from non-Syrian households. Due to a lack of publicly accessible population registers of the target populations, respondents were selected using area sampling combined with the random walk method. In Lebanon, five hierarchical lists were created on five geographical levels (governorates, zones within the governorates, sub-areas within the zones, clusters within the sub-areas, and neighborhoods within the clusters). In order to achieve the target size of the Syrian sample while remaining economically viable, only neighborhoods with an above-average proportion of Syrians are included in this selection.  In Turkey, the districts (cf. counties) with the highest and second highest Syrian population shares were selected in each region (cf. states) of Turkey. Within each city, blocks and sampling points were drawn at random. For details, see Gundacker et al. (2024). [2] 

West Africa

In Senegal, we used census discricts (DR) as the sampling base for selecting the primary sampling units. There are 17,165 census districts, and the survey covers 366 of them, with 84 in Dakar and 282 in the Casamance region. A systematic selection of DRs was conducted within each zone, with probabilities proportional to the number of households in each district. At the second stage, a random sample of twelve (12) households was selected from each DR that was chosen.

The sampling strategy for the survey in The Gambia utilized a nationwide random stratified sampling, which involved dividing the country into 14 substrata. These substrata were further divided into a total of 4,098 districts. From these districts, 164 districts were randomly selected to ensure representativeness across the entire country. To enhance the precision of the sample, the selection was weighted according to the relative population shares of individuals aged 10 to 30 years as recorded in the 2013 national census. In Nigeria, due to the lack of recent census data, the sampling relied on raster population data provided by Facebook, in collaboration with the Center for International Earth Science Information Network (CIESIN). Interview locations were determined using a multi-stage clustered random sampling method that was generated using GIS software.

For Edo, we implemented a two-stage simple random sampling procedure. First, we randomly selected 125 clusters, each measuring 5x5 km, with probabilities proportional to the population sizes within those clusters. This selection was made with replacement, allowing the same cluster to be chosen multiple times. Next, within these selected larger clusters, we randomly chose the corresponding number of smaller 500x500 meter clusters without replacement, again using probabilities proportional to the population size. Within each of these smaller clusters, 12 individuals were recruited using a random walk method, combined with the random selection of household members.

In Kaduna, four local government areas (LGAs) (i.e., Giwa, Birnin Gwari, Kauru, and Zangon Kataf) were excluded from the sampling frame for security reasons. To avoid skewing the sample towards certain parts of the state, we stratified the population before applying the sampling procedure. Specifically, we created three strata corresponding to the senatorial districts: Kaduna North, Kaduna Central, and Kaduna South, which contain 35%, 38%, and 27% of the state’s population, respectively. Sampling within these districts was conducted based on their relative population shares. In Kaduna, only 1,200 respondents were to be selected through random sampling, so we chose 100 clusters instead of 125.

The remaining 300 observations in Kaduna came from mobile herdsmen/pastoralists. Since these individuals are semi-nomadic, a sampling frame based on place of residence is not possible, as their exact location at any given time is unpredictable. Therefore, we adopted a flexible sampling scheme. At each of the 100 randomly chosen interview locations, the interview team inquired about nearby camps of mobile populations and recruited between 4 and 12 members for interviews.

Follow-up


Middle East

Follow-up surveys were conducted on an annual basis (Lebanon wave 2: 2020/21, wave 3: 2021/22, wave 4: 2022, wave 5: 2023; Turkey wave 2: 2020/21, wave 3: 2022/23). In up to three attempts, the participants from the baseline surveys were contacted again on the basis of the addresses where the previous interview took place. If the respondent could not be located, the agents tried via phone to arrange an in-person interview appointment (given a telephone number was provided in the previous wave). To compensate participant dropouts, a refreshment sample was recruited. If the sampling target could not be met within the neighborhoods of the dropped-out participants, additional neighborhoods were selected for the recruitment of participants, following the area sampling strategy of the baseline surveys.

West Africa

Follow-up surveys were conducted between November 2021 and January 2022 in Senegal and the Gambia and July 2023 in Nigeria. In all three countries, the initial plan was to first contact and ask participants from the first wave to participate in the survey again and, due to attrition, to select additional participants as a refresher sample according to the sampling process described above.

In practice, we obtained a very low number of recontacts of same respondents. Due to this, the second waves in Senegal, the Gambia, and Nigeria are effectively repeated cross-sections instead of a panel.

 

Interview Process


The surveys were conducted through Computer Assisted Personal Interviews (CAPI), based on questionnaires designed by the project team for each survey context. In the case of Lebanon, the interviews were conducted in Arabic by native speakers. In Turkey, the interviews were conducted in Arabic or Turkish according to the preference of the participants. In both cases, a native speaker conducted the interviews. In Senegal and The Gambia, survey interviews were conducted in French (Senegal), English (The Gambia) or a local language (Wolof, Pulaar, Mandinka, or Diola) according to the preference of the participants. In Nigeria, the interviews were conducted in English, Nigerian Pidgin or Hausa, according to the preference of the participants.

After explaining the purpose and content of the study and obtaining the consent of the participants, the interview was conducted in an ideally secluded place in the premises of the interviewee, which was determined by the interviewee him-/herself. Interviews lasted on average between 45-60 min. Participants were free to pause the interview at any time, skip individual questions, or stop the interview.

 

Data Processing and Data Availability


The data provided in the Data Explorer have been cleaned of outlier observations based on standardized criteria and are fully processed. Prospectively, follow-up Waves of the survey will be added to the Data Explorer as they become available.

The Middle East survey data (“Transnational Perspectives on Migration and Integration – TRANSMIT Surveys of Syrian Nationals and Their Neighbors in Lebanon and Turkey (2019–2023”) are available as SUF through the DeZIM.fdz [3]. Please also check [2] for additional information. For remaining questions, please contact Irem Tuncer-Ebetürk (ayse.irem.ebetuerk(at)hu-berlin.de). If you are interested in the West African datasets, please contact Irene Pañeda-Fernández (irene.paneda(at)wzb.eu).

 

[1] The web application was developed by Long Nguyen (source code). The content is licensed under a Creative Commons Attribution 4.0 International license.

[2] Gundacker, Lidwina, Hertner, Laura, Kühnert, Nora, Rischke, Ramona, Ruhnke, Simon, Talebi, Nader, Trocka, Caroline and Brücker, Herbert. 2024. Transnational Perspectives on Migration and Integration – TRANSMIT Surveys of Syrian Nationals and Their Neighbors in Lebanon and Turkey (2019–2023), Journal of Economics and Statistics. https://doi.org/10.1515/jbnst-2024-0091

 

[3] https://datasearch.fdz.dezim-institut.de/item/de.dezim/0179b282-f8b1-4bc9-accd-4fe9aecfc02e