Methodological Approach

Definition and analytical categories

The survey aims to shed light on the various factors shaping returnees’ patterns of reintegration in the Maghreb countries. It sets out to highlight the heterogeneity characterising the categories of returnees while analysing their respective needs in their country of origin.

The definition of the returnee, used by the team of the MIREM project, is the following:

Any person returning to his/her country of origin, in the course of the last ten years, after having been an international migrant (whether short-term or long-term) in another country. Return may be permanent or temporary. It may be independently decided by the migrant or forced by unexpected circumstances.

This definition partially draws on the one recommended by the United Nations. It refers specifically to migrants who returned to their country of origin in the course of the last ten years, for this time limit allows the impact of the experience of migration on the interviewee’s pattern of reintegration to be assessed. It also allows the respondents to recount their migratory experiences more precisely.

Two broad categories of returnees were taken into consideration:

• Migrants who decided to return independently to their country of origin;
• Migrants who were forced to do so owing to unfavourable circumstances.

As the collected data were gradually processed and analysed, it turned out that the distinction between the migrants who decided individually to return to their country of origin and those were forced to return was key to understanding how and why returnees’ patterns of reintegration differ from one another.

The interviewees belong to various occupational categories, namely:

• Employees;
• Entrepreneurs/employers;
• Unemployed people/job-seekers ;
• Students ;
• Retired people;
• Housewives.

Various profiles have been identified. They differ from one another in terms of:

• Patterns of reintegration;
• Return motivations;
• Human and financial capital and patterns of resource mobilisation;
• Return context;
• Experience of migration and duration.

Filling the deficit of information and sampling method

The information to be collected was identified following a thorough inventory of the existing statistical and documentary data related to return migration in Algeria, Morocco and Tunisia. The aim was to understand:

• The factors motivating the departure of the interviewees from their country of origin;
• The impact of the experience of migration lived abroad on the interviewee’s pre- and post-return conditions;
• The various post-return conditions of the interviewees and their perspectives of reintegration.

The statistical and documentary inventory allowed the sample and the sex distribution as well as the geographical stratification to be carried out in each Maghreb country. This was collectively done during the working sessions organised at the European University Institute and gathering all the institutional partners involved in the MIREM project:

• The European University Institute (EUI, Florence, Italy) ;
• The Centre de Recherche en Economie Appliquée pour le Développement (CREAD, Algiers, Algeria) ;
• The Association Marocaine d’Etudes et de Recherches sur les Migrations (AMERM, Rabat, Morocco) ;
• The Institut National de Statistique et d’Economie Appliquée (INSEA, Rabat, Morocco) ;
• The Office des Tunisiens à l’Etranger (OTE, Tunis, Tunisia).

Various versions of the questionnaire were circulated and exchanged among the partners, as a result of these working sessions, with a view to gradually heading towards a final draft. The questionnaire results from a collective compromise between all the partners of the project as well as from the desire to optimise its management on the field. The last version of the questionnaire comprises close-ended questions. However, open-ended questions have been included in the questionnaire, particularly regarding the degrees and occupations of the interviewees. The modality “Other” has been inserted in the questionnaire to gather further information if necessary.

Multiple-choice entries have been included in various questions. Often, their structure is dichotomic (Yes/No answers). This configuration was chosen in order to facilitate the ensuing data processing. Also, it allowed the complexity of certain issues, such as the family composition, the occupational class and sectors and the types of investments to be properly reported. On various occasions the interviewees were asked to classify by order of priority their replies, particularly regarding the return motivations.

Filter questions have been necessarily used in the questionnaire to highlight the variety of the migratory experiences and the manifold patterns of reintegration.

The Eurostat nomenclature of country codes has been used and the occupational ISCO codes have been simplified for the purposes of the survey.

A three-stage questionnaire

The questionnaire is structured in three distinct migratory stages:

These three stages allow the factors inherent in the returnees’ migratory experience, as well as those that are external to it to be identified, while viewing return as a changing process, whether it is permanent or temporary. In other words, thanks to this approach, it possible:

• To understand the extent to which the experience of migration, as well as the social and institutional context at home, have impacted on patterns of reintegration;
• To analyse why and how the human social and financial capital of the interviewee has changed over time;
• To compare diachronically the various factors having motivated and shaped the migratory stages.

Preparation of the survey
Once the variables have been consensually agreed, a pilot survey was organised and carried out in each Maghreb country. The questionnaire was administered directly with the respondents.

The pilot survey was a prerequisite to optimising the administration of the questionnaire and to maximising the response rate. Around ten pilot interviews were carried out. Then, the tested field data were processed on a common template which was prepared using the SPSS software. The pilot survey was critical in enhancing the wording of the questionnaire and in correcting its shortcomings. The online questionnaire results from a series of amendments.

The fact that each partner institution in Algeria, Morocco and Tunisia has a proven knowledge of the field and several contacts with migrant-aid associations and networks was essential to meet the respondents. Interviews were carried out in public and private places, sometimes at home, and more often than not in Arabic, sometimes in French.

Each partner was in charge of recruiting interviewers in the selected regions of inquiry. Training sessions addressed to interviewers were organised in each country with a view to ensuring that:

• The objectives of the survey were clearly understood and that the interviewers would administer the questionnaire properly without influencing the respondent;
• The rules of confidentiality and anonymity were respected;
• The duration of each interview did not exceed 45 minutes, as far as this was possible;
• The administration of the filter questions was optimal;
• The regional distribution of the teams of interviewers was properly done;
• The procedures for collecting the field data were respected and controlled, if need be, by the partner in charge of controlling the implementation of the survey operations.

The geographical stratification as well as the distribution by sex of the sample were verified and compared with the official statistical data with a view to having the possibility to weigh the collected data.

992 interviews were collected as a whole: 332 in Algeria, 330 in Morocco and 330 in Tunisia. The geographical stratification was as follows:

In Algeria, the wilayas of Algiers, Bejaia in Kabylie and Setif eastwards of the capital and Tlemcen westwards of the country were covered.

In Morocco, the region of Tadla-Azilal and the coastal regions of Casablanca, Chaouia-Ourdigha and Rabat-Salé-Zemmour-Zaër were privileged.

In Tunisia, the northern governorates of Tunis, Ariana, La Manouba, and Nabeul were covered as well as the governorates of Soussa and Sfax in the centre of the country, and of Medenin in the south.

Duration of the survey
The surveys started at the same time, in the three countries, in September 2006 and ended in January 2007.

Data processing
From the preparation of the survey, all the partners agreed to use a common template which necessarily drew on the structure of the questionnaire. Moreover, new variables were included in order to foster the exploitation of the processed data as well as their analysis.

In order to avoid any delay, the processing of the field data started as these were collected, validated and checked. The simultaneous collection and processing of the data allowed the geographical stratification and sex distribution of the sample to be controlled on a regular basis.

Moreover, each partner forwarded the processed data to the coordinating unit so as to ensure the harmonised codification and treatment of the field data.

• Reasons and factors motivating or determining the departure for abroad, and the post-return conditions;

• The type and length of the experience of migration;

• The impact of the experience of migration on the patterns of professional reintegration of the returnees in their country of origin and on the welfare of their households;

• The returnees’ projects before and after return;

• The skills acquired abroad and in the country of origin;

• The financial resources of the returnees and their patterns of reintegration at home;

• The returnees’ links with their former country(ies) of immigration;

• The facilitation from which the interviewees may have benefited when returning to their country of origin;

• The types of investments made by the interviewees in their former country(ies) of immigration and their country of origin;

• The returnees’ perception of their institutional environment.

• Descriptive. The analyses will be based on a series of cross-tabulations highlighting the evolution of some variables compared with others, while referring to the abovementioned three-stage structure of the questionnaire. Various profiles of returnees will be identified differing from one another in terms of patterns of reintegration;

• Exploratory, by using factor analyses in order to explain the variability of observed and unobserved variables (simple correspondence analyses, multiple correspondence analyses) thanks to a software allowing various sociodemographic and economic variables to be crossed;

• Interpretative. Regression models will allow the functional dependence of some elements to be analysed with reference to a series of explanatory variables or predictors. Various models will be tested with a view to leading to a model highlighting the most significant independent variables.