Titre alternatif

MAFE Ghana

Titre parallèle

Migrations entre l'Afrique et l'Europe - MAFE Ghana (2009-2010)

Pays

Nom	Abréviation
Ghana	GH
Netherlands	NL
United Kingdom	GB

Couverture géographique

• Six European countries and three African countries participated in the MAFE surveys. Data collection was carried out in both sending countries in Africa and destination countries in Europe, in order to constitute transnational samples. For MAFE Ghana, data was collected in Ghana (African part), and the Netherlands and United Kingdom (European part).

Univers

GHANA Household: Households selected randomly from the updated list of households in the selected primary sampling units. Three strata were distinguished: households with return migrants, with migrants abroad, and without migrants. Individual: People aged 25-75, born in Ghana. This lower age limit was set in order to obtain informative life histories. By not including respondents younger than 25, the resources were used more effectively. The place of birth criterion was used to exclude people who were born out of their country of origin in order to exclude second generation migrants in Europe and to increase the homogeneity of sample. All the return migrants and partners of migrants, and one randomly selected other eligible person. Return migrants were eligible if their first departure was above at 18 or over. EUROPE In all the European countries, the surveys were conducted among males and females who were aged 25 and over at the time of the surveys, and who were 18 or over when they had left Africa for the first time for at least one year. Migrants from only Ghana were interviewed.

Type de données

survey data

Méthodologie

Dimension temporelle

Cross-section

Dates de collecte

Période	Libellé
2009-04 - 2009-07	Netherlands survey - First phase
2009-05 - 2009-05	Ghana pilot survey
2009-06 - 2009-11	United Kingdom survey
2009-07 - 2010-01	Ghana (Accra & Kumasi) survey
2009-11 - 2010-04	Netherlands survey - Second phase

Fréquence de collecte

GHANA In Ghana, the preparation of data collection started in February 2009, and a pilot survey was organized in May 2009. The selection of survey sites and the listing of households in the sites were carried out in May 2009. Fieldwork started in July 2009 and lasted approximately 6 months (from July 2009 to January 2010). Several interviewers dropped out during data collection and could not be replaced, which contributed to lengthening the fieldwork. Editing and data coding were done in parallel with data collection, and were over by the end of January 2010. Data entry and data cleaning started in February 2010, and ended in March 2010. EUROPE In the Netherlands and United Kingdom, data collection was conducted in 2009-2010. Data collection lasted about five months in the UK and three months in the Netherlands. Editing was done along data collection. Data entry was done between October and December 2009 in the UK, and between June 2009 and September 2009 in the Netherlands. The Netherlands started a second phase of data collection in November 2009 (they had some funds left which allowed them to increase the sample size).

Procédure d'échantillonnage

• Probability: Stratified GHANA A three-stage stratified random sample was used. At the first stage, primary sampling units (census district) were selected randomly with varying probabilities. At the second stage, households were selected randomly in each of the selected primary sampling units (PSUs). At the third stage, individuals were selected within the households. a) Selection of primary sampling units (first stage) For Ghana, the target areas were the cities of Accra and Kumasi. In each of the cities, a sampling frame of primary sampling units was prepared. In Senegal and Ghana, recent censuses were available and served as sampling frames at the first stage. At the first stage, census enumeration areas were randomly selected. In Ghana, 80 enumeration areas were selected with a probability proportional to size. As no information was available for stratifying, the sample was not stratified in the first stage in Ghana. b) Selection of households (second stage) A listing operation was carried out in each of the selected survey sites to prepare the sampling frame of households. The listing consisted in enumerating all the households in the selected sites, and in identifying whether these households included migrants of not. In DR Congo and Ghana, three categories of households were distinguished (households with return migrants, with migrants abroad, and without migrants). 8 households were selected in each of the 3 strata (if less than 8 households were available in one or several strata, the remaining households were selected in the other stratum). The sampling rate was higher in strata of households with migrants, in order to get a sufficient sample of such households. c) Selection of individuals (third stage) In each of the selected households, one or several respondents were selected among the eligible people (people aged between 25 and 75, and born in the origin country). In DR Congo and Ghana, all the return migrants and partners of migrants currently abroad were selected. In addition, one other eligible member was randomly selected. A special tool had been designed so that the interviewers could randomly select the people during the fieldwork. Two types of questionnaires were used in the departure countries: the household questionnaire and the individual life history questionnaire. - The first questionnaire was used among a representative sample of households in the target region. - The second questionnaire was used among a sample of individuals in the selected households, targeting both return migrants and non-migrants. The household questionnaire was thus used as the sampling frame for the selection of individual respondents. EUROPE The objective of the survey was to obtain a sample 'as representative as possible' of the African populations (Congolese, Ghanaian, Senegalese) in the destination countries (150 individuals per origin and destination country). The way the sample was constituted may vary across countries, but some common principles were respected: - The composition of the sample should be as close as possible to the population of (Congolese, Ghanaian, Senegalese) migrants in the country in terms of gender, geographic distribution, age, socio-economic category or occupation. - One exception: the sample should be gender balanced. Males and females should be equally represented in order to allow gender analyses. - Samples in origin and destination may be linked, but migrants with weak or no relationships at origin should not be excluded from the sample. - Both documented and undocumented migrants should be represented in the sample. As no suitable frame was available to select randomly individual respondents in five of the six European countries (Spain being the exception), it was decided to use quota sampling. In all the countries, the quotas were set by age and gender at least. In the UK, the place of residence was also used in the quotas. In the Netherlands, sub-regions concentrating the majority of migrants were selected. In the United Kingdom, the surveys were concentrated in the London area and in the places where Ghanaian migrants were living. Randomness was also included in the samples in different ways. For instance, in Belgium, a random sample of places was selected according to the number of people of Congolese origin living in these places. Respondents were selected in these places. The combination of different recruitment methods also ensured that different types of persons had a non zero probability of being included in the sample. For instance, some respondents were recruited in public spaces (street, metro station, hairdresser...), others were randomly selected from list of volunteers identified in churches... *The Netherlands* - Target areas: 3 cities (in 3 different provinces): Amsterdam, The Hague and Almere - Sample size: 272 - Quotas: By age, gender - Recruitment methods: Public spaces, churches, snowballing, interviewers' contacts *United Kingdom* - Target areas: Whole country - Sample size: 149 - Quotas: By age, gender and place of residence - Recruitment methods: Public spaces, churches, snowballing, interviewers' contacts

Mode de collecte

Face-to-face interview: Paper-and-pencil (PAPI) The general strategy was the following one: 1. A household survey was conducted among a sample of households in the capital cities in Africa (household questionnaire in origin countries); 2. A life history survey among a sample of individual respondents was conducted in the departure countries (non migrants, return migrants and spouses of migrants). The individual respondents were selected from the households in the origin countries (individual questionnaire in origin countries); 3. A life history survey was carried out among migrants in destination countries (individual questionnaire in destination countries). All the surveys were done using paper questionnaires through face-to-face interviews. GHANA In Ghana, both household and individual surveys were carried out at the same time. It necessitated drawing individuals within the households through the Kish selection method. The average duration of interviews for the household questionnaire was about 45 minutes-1 hour in Senegal, and a little shorter in DR Congo and Ghana. The average duration of the biographic questionnaire was also around 45 minutes, but it varied greatly depending on the age and migration status of the respondents. EUROPE In Europe, the moment for data collection depended on the selection method. In MAFE Senegal, the fieldwork could start only after completion of the household survey. As a consequence, the fieldwork for the biographic questionnaires was done almost at the same time in all countries. The work of the interviewers included three stages: o The interviewers first had to set up an appointment with respondents by using the phone contacts or another source of recruitment (except in cases persons were directly available). Interviewers needed to confirm the appointment. The choice of the place and time of the interview were left to the respondents. o The interview was then done. The average duration of interviews was between 1 and 1.5 hours. In most cases, interviews were carried out at the house of the respondents, but it also took places in various places (pubs, street, office…). The interviews were done during weekdays or week-ends, at various times. o Finally interviewers were asked to read quickly the completed questionnaire as soon as possible after the interview, in order to detect any missing parts or inconsistencies, and correct them on the spot. Approximately one additional hour was necessary for this. In Belgium, the Netherlands and the UK, interviewers also had to transcribe the information from the ageven grid to the questionnaire directly after the interview (this was done by the editors in the other countries). In all the countries, the respondents were offered a small gift at the end of the interview. In most countries, this was a calling card. In the UK, money was given to the respondents and in Belgium, respondents were given the choice between a calling card and a voucher in a supermarket. The value of the gift varied between 5 € (Italy) and £15 (UK). In all the countries, the gift was very much appreciated. Although the gift was offered after the interview, some participants knew in advance they would receive it. For instance, a few surveys were conducted in a center for asylum seekers in Belgium, and the information about the gift quickly spread among the Congolese migrants. Recruitment by snowballing also meant that respondents were sometimes aware they would receive a gift. Although this may have facilitated the recruitments of some persons, it may have affected negatively the composition of the sample.

Type d'instrument de collecte

Structured questionnaire The MAFE surveys rely on two different questionnaires: household and biographic. The questionnaires are almost entirely identical from one country to another. The few differences consist in: - Cultural variables (religion, ethnic groups, matrimonial status etc.) - New questions introduced on fostered children in the MAFE DR Congo and MAFE Ghana biographic questionnaires (Module on Children) - The order of the questions relating to migration in the household questionnaire (Module A) The Household Questionnaire: - Used only in African countries - Answered by a unique respondent who is usually the household head - Contains information on the members of the household (age, sex, education...) and also on people who live outside the household and who are related to it (head's children, partners living abroad, other relatives of the head or his/her spouse who live abroad and with whom the household has been in touch within the last 12 months). In addition, it contains specific modules on short migration histories, on relationships between each migrant and the household, and on housing conditions and owned assets. - Topics: socio-demographic variables of each individual, short migration histories, remittances, household assets, housing history - Available in French (MAFE-Senegal, MAFE-DR Congo) and English (MAFE-Ghana) The Biographic Questionnaire: - Used in all African and European countries - Answered by the migrant him/herself - Contains life histories of all the surveyed individuals, whatever their migratory status at the time of the survey (non-migrant, return migrant, current migrant). The questionnaire contains retrospective information on the following topics: dwelling, family, work, international migration of the interviewee (including attempts to migrate, return trips to the country of origin, transit migration and legal status in foreign countries), migration history of the migrant's relatives (list of their stays abroad, including dates and country names), goods and assets, and remittances and contributions to associations in the country of origin. - A grid was used, jointly to the questionnaire, to help the interviewee to recall important dates of his/her history - Topics: family formation, education and employment, housing, migration, investments (housing, business, community amenities or infrastructure)... - Available in French (MAFE-Senegal, MAFE-DR Congo), English (MAFE-Ghana), Italian and Spanish (MAFE-Senegal)

Caractéristiques de la collecte

The MAFE surveys collect information on potentially vulnerable populations (undocumented migrants) and on sensitive subjects (remittances, legal status…). In order to facilitate the fieldwork and increase the quality of the data, it was important to carefully inform the people who were to be interviewed. The legal pre-requisites changed according to the country. In France (only), a legal authorization had to be obtained before starting the fieldwork. The CNIL (Commission nationale informatique et libertés) was concerned by the way the contacts were going to be obtained in Senegal and, most of all, by the sensitivity of certain variables contained in the questionnaires (ethnic group, religion). We obtained the authorization to ask these questions, but in order to keep them in our files, we had to ask to the interviewees to sign a written informed consent. According to legal prescriptions, in all European countries, a letter was designed to explain their rights to the interviewees. In most countries, a leaflet was designed and used to sensitize respondents and authorities about the MAFE project. In advance of the survey, several communication actions have been undertaken: - In Africa, inform neighbourhood heads / municipalities of survey by an official letter or by a visit - Use local radio / migrants radio and chat show to present the survey - Inform an organisation of migrants who can support the survey - Visit the key places of the community (churches…) Because of the complexity of the questionnaires, only interviewers with a good experience in complex surveys were recruited. In African countries, it was highly recommended to hire the same interviewers to conduct both household surveys and individual surveys. This approach proved to be very efficient in all the surveys. In Europe, interviewers had to be able both to recruit the migrants and to fill correctly the questionnaire. As a general rule, it was preferable to have a relatively small number of well-trained interviewers than a large number of interviewers. Overall, around 20 to 25 interviewers and supervisors were involved in data collection in each country. The number of the interviewers per survey varied between 8 (survey among Ghanaians in the UK) and 17 (Netherlands). In all the countries, both male and females interviewers were hired; most of them had higher education and some experience with data collection. In some countries (e.g. France), some of them were professional interviewers. The selected interviewers were not necessarily from the same country as the respondents, but most of them also had foreign origins. For instance, 7 of the 12 interviewers in Belgium were of foreign origin, 5 of them from DR Congo. In the Netherlands, most interviewers were from Ghanaian origin. The fact that many of the interviewers were themselves of foreign origin seemed to have positively influenced the willingness of interviewees to participate in the survey. Directly after being filled, questionnaires were checked by the interviewers and supervisors. They were then sent to a small team of editors for an in-depth reading. The editors consisted of 9 people in Senegal, 6 in Ghana and 5 in DR Congo. The team had followed the same training as the interviewers, and also received a specific training for editing the questionnaires. Data entry was performed using MS Access programs prepared by Ined.

Traitements

Taux de réponse

GHANA For the household questionnaire, 1920 households were selected (1440 in Accra and 480 in Kumasi), only 1246 were successfully interviewed, including: - Non-migrant household: 449 - Household with at least 1 returnee: 346 - Household with at least 1 current migrant: 675 - Household with returnee(s) and current migrant(s): 224 This represents a response rate of 64.9%. For the biographic questionnaire, 1 490 individuals were selected, only 1 243 were interviewed, including: - Returnees: 319 - Partners left behind: 84 - Other non-migrants: 840 This represents a response rate of 83.4%. The overall response rate in Ghana is 54.1%. EUROPE 417 Ghanaians migrants were successfully interviewed: 279 in the Netherlands and 138 in the United Kingdom.

Autres formes d'évaluation des données

A methodological note entitled "Sampling international migrants with origin-based snowballing method: New evidence on biases and limitations", written by Cris Beauchemin and Amparo González-Ferrer, can be found in the study's related materials, as well as another methodological note in french "Biais de non-réponse dans l'enquête Migrations entre l'Afrique et l'Europe (MAFE-Sénégal)" written by Nicolas Razafindratsima, Stéphane Legleye and Cris Beauchemin.

Pondération

GHANA The computation of sampling weights relies on computing sampling probabilities at each stage. The product of sampling probabilities at each stage gives the overall sampling probability. Taking the inverse of the sampling probability gives the inflation factor. These factors are adjusted (trimming, adjusting for population size). They are normalized, so that their sum is equal to the sample size. EUROPE In the European countries, similar sample sizes were selected for males and females, resulting in an overrepresentation or underrepresentation in the MAFE samples. Similarly, older people were usually oversampled. For these reasons, post-stratification weights are computed to give each observation its proper weight and to match the samples as closely as possible to selected population characteristics. In the MAFE data, all survey weights have been rescaled (normalized) so that the sum of weights corresponds to the sample sizes of households and individuals respectively while the mean of the weight variables equals one. For further details about weights, please read the MAFE methodological note 6 entitled "Sampling and Computation Weights in the MAFE Surveys" (see related materials).

---