Navigation – Plan du site

Volunteer Associations and Integration of Immigrants in France

William Berthomière, Mathilde Maurel et Yann Richard
Cet article est une traduction de :
Intégration des immigrés et associations en France. Un essai d’approche croisée par l’économie et la géographie


The issue of integration is key in the literature about migration. It refers to a set of public policies aiming at integrating foreign populations in a given society. This paper aims at investigating the relationships between the integration of four communities settled in France (Algerian, Portugese, Turkish, and Vietnamese) and the presence of associations. The methodology is based on two approaches, quantitative economics and geography. It uses a new database, extracted from the Journal Officiel and several surveys, noticeably TeO. In a first step we ask whether the regional distribution and the density of associations explain the degree of integration of the migrants stemming from the four communities. In a second step we test whether membership in an association increases or decreases the adoption of oppositional identities and if the latter influences the integration via the access to employment.
migration, econometrics, geography, social integration

Haut de page

Notes de l’auteur

This article and its translation have been produced with the support of labex DynamiTe.

Texte intégral

1The term “integration” was used for the first time by the public authorities in France in 1974 on the suspension of work-based immigration and the introduction of family reunification. The notion of integration refers to: “The process, of indeterminate length, by which one or more individuals living in a society, which is by definition foreign, demonstrate their willingness to participate in the construction of the national identity of said society, which takes a series of economic and social measures to help the individual(s) achieve this objective” (Khellil, 2005). When the process applies to the entire population of a country, it is called “socialisation”. It requires action on the part of two stakeholders: the individual or group that integrates by travelling over a certain distance; and the host country (through government policy, for example) since integration demands a set of policies implemented by the public authorities to enable populations (whether immigrants or not) to be integrated properly into a given society. The ultimate aim may be assimilation in the majority society, which marks the end of relations with the country of origin (a physical and cultural separation).

2Integration is a complex process. As defined by the Chicago School and Émile Durkheim, it also denotes the process by which an individual or group of individuals gradually take their place in a society on two registers. The first is economic, with access to employment, followed by the process of learning about the norms and standards that govern the way the society operates. This learning takes place largely within the family context or at school. Some authors emphasise the fact that integration is not a linear, monolithic process. Nor is it always translated into upward mobility, which sometimes does not go hand-in-hand with cultural integration (Safi, 2006). Furthermore, integration takes different forms in different countries (e.g. “integration contracts”, compulsory language courses for newcomers or tests on arriving in a country) and may even entertain cultural particularities. The United Kingdom, for instance, has had a policy of multi-culturalism for some time, and in the Netherlands there are denominational community institutions that are co-funded by the state.

3In this article, we analyse the situation in France, which has long been presented as a model of assimilation (Noiriel, 1998). The idea of the “French melting pot” – creuset français – has been the subject of debate for two key reasons. Some authors, on taking a close look at public policy, have demonstrated that governments have had a broader view of the issue of migration since the 2000s. This is reflected, for example, in the way in which the Muslim population has been taken into account in the fight against ethnic discrimination (Brubaker 2001, Simon and Amiraux 2006). Other authors (Beauchemin, Lagrange, Safi, 2011) argue that the Jacobin approach still prevails, with priority accorded to the principle of integration, characterised by severing ties with the country of origin and creating ties with the immigrant population belonging to the same community.

4We adopted an inter-disciplinary methodology that combined a geographical and economic approach. We were keen to study the potential relationship between the integration of individuals from four different communities and the existence of volunteer associations. Is there a conflict between integrating into French society and belonging to an association? In other words, does maintaining ties with one’s country of origin and community through community associations prevent integration? In an attempt to answer these questions, we studied possible relations between the integration of members of four communities (Algerian, Portuguese, Turkish and Vietnamese) resident in France and the density of the network of volunteer associations. We selected this sample because the communities are very different in terms of their numbers, how long they have been in France, the rate at which they arrived in the country, and so forth.

5The first part of this article reviews research into the connections between integration and volunteer associations. We then introduce the method we applied and the sources used. In the second section, we present the geography of the community associations linked to the four selected groups. The aim is to determine whether there is a relation between the geography of the associations and the geography of the performance of these populations in terms of integration by comparing the different regions in Metropolitan France. We look at the density of the provision of associations (the number of associations for each group in each region), taking into account only community associations, i.e. associations that are exclusively connected to Algerian, Portuguese, Turkish and Vietnamese first-generation immigrant communities. In the third part, our focus turns to the tendency of members of the four communities to belong to associations, whether they are community-based or not. We use modelling to try to ascertain whether belonging to one or more associations reinforces immigrants' adherence to values that are oppositional to the French cultural model; and whether this has an influence, at least indirectly, on their integration through access to work.

I. Immigration, integration and volunteer associations: current thinking and sources

  • 1 We should note that Schnapper was able to convey this view when writing: "Public policy can only ai (...)

6Studying the links between the integration of foreign populations and the vitality of associations is a research priority. Yet, putting this analysis into practice is difficult, to say the least, since the statistical sources and information required are largely inaccessible. At the same time – and in line with France’s established relationship with foreign immigration – the refusal to include integration within the framework of community-based discourse (relative to Anglo-American approaches) has shifted the issue to the discipline of urban policy or specific educational policies (ZEP: Educational Priority Areas). This results from the fact that France’s suburbs – banlieues – have gradually come to symbolise the question of social integration due to the growing proportion of immigrant populations that have found housing there (Schnapper, 2007: 204). The dialectic linking community associations and integration has consequently been rendered even more complex in part through the action of the state. Successive French governments since the 1980s have never stopped creating, rebuilding or redefining the public framework of the political priority that integration has become. The vast array of initiatives, together with the diverse responses they have elicited, have helped give shape to the notion of integration. In the early 1990s, this dynamic led Dominique Schnapper to sum up the essence of integration as: “The process by which individuals take part in collective life through professional activity, learning about consumption patterns, adopting family and social behaviours and forming relationships with others” (1992: 18). More recently, Serge Paugam has emphasised that this notion cannot be reduced to the issue of immigration alone and that: “The explanation about problems of integration is to be found not only in the intrinsic difficulties experienced by immigrant populations in relation to the prevailing norms in the country they have settled in but also in the functioning – or, rather, the malfunctioning – of the institutions that are supposed to help integration” (2014: 5).1

7This article will add to our knowledge about the processes of integration by focusing on the structure of associations and looking at them from both a spatial and thematic perspective. Drawing on the data supplied by the Journal Officiel de la République Française (the official gazette of the French government), we commit to a functional approach (as suggested by Marie Poinsot (2000: 49)): the various sectors of activity of the associations inform us not just about the dynamics common to all four groups under review but also about the specific nature of their initiatives or engagement, observed especially at regional level. In addition to this spatialised reading, this research takes a fresh look at the dynamics of associations through the prism of the identity positioning and values that drive them by leveraging the results provided by the Trajectoires et Origines (Trajectories and Origins) survey. The TeO is the first genuine large-scale study conducted in France on the impact of a person’s origins on his or her living conditions and social trajectory. The results on the decisive nature of adherence to oppositional values or identity positioning echo the findings of Dominique Baillet in the late 1990s, when he differentiated society-based associations from community-based “communalist” associations (2000: 56). Accordingly, the scientific challenge is sizeable even if we are well aware that any research on associations, and any attempt at categorisation in an immigration situation, is fundamentally flawed since, as Jocelyne Streiff-Fénart reminds us: “Assessing immigrants' values depends on the point of view from which they are considered, and this point of view is largely shaped by national conceptions of immigration and the relationship to foreigners, as well as by the institutional arrangements influenced by these conceptions, and the government policies that implement them” (2006: 854).

The presence of foreigners and the association movement: historical background and a growing problem

8In an article marking the centenary of the Waldeck-Rousseau Law of July 1, 1901 on freedom of association, François Boitard wrote that France has always had a conflicting relationship with associations, “wavering in turn between freedom and control, especially when foreigners are involved” (2001: 5). The 1901 law authorising freedom of association in France was the result of a long process (in comparison with Great Britain, which authorised such a law in the 17th century), with each stage characterised by feelings of mistrust towards potentially conspiratorial collective dynamics. However, when the law was enacted, it did not impose any restrictions on association for foreigners. During a period of instability in Europe, and when there was a desire to prevent “any loss of cohesion inside France by means of utterly reprehensible acts” (Niboyet, 1946: 53), the French state took the decision to promulgate the statutory law of April 12, 1939, which saw freedom of association withdrawn from foreigners. It was to be over forty years before the law was repealed by François Mitterrand’s government on October 9, 1981 (Belorgey, 2000; Boitard, 2001). Patrick Weil gives a very clear account of France’s relationship with foreigners living in the country in his book La France et ses Étrangers: “Those who engage in the fight for social justice, rather than practising their religion in the places of worship provided for them in the workplace, are deemed to be causing a breach of the peace and so are engaged in combat” (1991: 104). Although forbidden by law, political events in some of the countries of origin of foreigners living in France have always led to forms of mobilisation and grouping by nationality. Antoine Dumont (2010) cites the example of Italians who were active in the anti-fascist associations during the inter-war period. Weil, for his part, stresses that these mobilisations were monitored extremely closely by the state. He points to the “cooperation” that grew up between the national police and police forces in “some countries of origin regarding the trade union and political activities of their nationals” (Weil, 1991: 104).

  • 2 These notions echo the problem of nationhood as defined by Benedict Anderson. As Antoine Dumont poi (...)

9During the 1980s, the issue of associative life, migrants and their descendants became a topic of research in France and many other European countries. This movement emerged as a carbon copy of the debates that were shaking the world of academia around the term integration. This period was notable for the creation of the Haut Conseil à l’Intégration (Higher Council for Integration) in 1989. Notwithstanding the attention paid to the issue by the government, it only began to fund research work once the shortcomings of the institutions intended to enable integration (to use Serge Paugam's words) started to emerge in public debate (the 1983 “March of the Beurs”, for example). Research questions covered an ever-expanding field of analysis and their dynamism was obvious. The gradual overlapping of the issues of integration and urban policy, and the emergence of the question of identity and the trans-nationalisation of feelings of belonging and mobilisation testify to this.2 In France – as in the major European immigration countries more widely – in spite of this work and regardless of the approach prioritised by government policy (community-based or not), most research centred on the same question of ambivalence towards associations (Cattacin, 2007). As Riva Kastoryano points out: “Public recognition of the cultural specificities of foreign populations is felt by indigenous people as a threat to the homogeneity of the nation-state. Given France’s Jacobin tradition, public opinion is turning toward the question of the ‘ability’ of immigrants from North Africa to ‘assimilate’. The emphasis is on the legacy of the colonial ties between the Maghreb and France. The French and Germans insist on the cultural distance, on the link between religion and politics in Islam, and on the different conception of private law” (1989: 15). More generally, these debates are polarised around two schools of thought: the “national-republican” approach and a line that focuses on a generic notion of integration combining participation and cultural integration. André Réa and Maryse Tripier distinguished the following features: “Participation covers inclusion in the economic, political and national spheres based on the mobilisation of stakeholders. Cultural integration is more concerned with the challenges of cultural conflict” (2008: 92-93).

10The study of the diversity of association dynamics in France is mainly based on surveys conducted at local level. In the late 1990s, the journal Migrations Études stressed that this research was shaped by “highlighting the considerable integrative work carried out by associations as active elements in local sociability, which, based on the world experience of the people living in the quartiers and the community organisation of everyday life, help to maintain social ties and generate a civil bond, which (one might think) is the means for these poorly-integrated populations ‘to achieve’ a local and participative form of citizenship. Short of a civism based on involvement in the affairs of the local community, and short of a vision of social proximity as a civic or counter-power matter, it is important to see that these populations are rebuilding a ‘civil community’ in a somewhat modest and invisible manner, where ordinary reciprocal rights between individuals are reconstructed in daily human interactions” (Eme et al., 1997: 2).

11The work as a body gives a picture of an associative process that expresses itself as a dynamic of “acting together”, which may just as easily find expression in communitarian choices as in a desire for social segregation. The research by M.-A. Hily and M. Poinard, which falls into this category, is based on studying the Portuguese association movement. It explores the “dynamic density” of the associative movement in order to sketch out the moral relations developed by individuals in the sense attributed by Durkheim, according to the authors: “In other words, they not only exchange services or compete with each other but also live a communal life” (Hily, Poinard, 1985).

  • 3 From the notice issued by the Haut Conseil à l'Intégration (p.13) entitled: Investir dans les assoc (...)

12Given the large number of associations in France (more than 1.3 million, with over half dedicated to sport and culture), it is difficult to fully grasp the diversity of the association process in a migratory context. The dearth of reliable data on the exact number of associations involved in migration makes it impossible to measure the density of the network. Nevertheless, one of the final reports produced by the Haut Conseil à l’Intégration, Investir dans les associations pour réussir l’intégration, estimated that in the early 2000s only 1,300 associations were still state partners in the field of integration, to which should be added “9,000 associations funded by the Agence Nationale pour la Cohésion Sociale et l’Égalité des Chances (French Agency for Social Cohesion and Equal Opportunities: ACSE) and the General Secretariat of the Inter-Ministerial Committee for Towns and Cities (SG-CIV) as part of urban policy”.3

From lists to surveys or the TeO “revolution”

Example of a form declared in the Journal Officiel

Example of a form declared in the Journal Officiel

13To work towards a better understanding of this issue, our methodology focused on exploiting data provided by the Journal Officiel, which lists all the associations created under the 1901 Law.4 We explored the database to extract the data for associations related to the four countries of origin that we had selected. The inventory created from this database allowed us to locate the headquarters of the relevant associations and to ascertain the dates they were set up and their areas of activity (see box opposite). The information provided was of high quality; nevertheless, it should be noted that the data in the Journal Officiel has a weakness: although associations are required to report any changes in their functioning, those that no longer exist do not always notify the prefectural services. This first source of information helped us identify over 4,350 associations (see Table 1). We chose to pair it with data from the survey Trajectoires et Origines. Enquête sur la Diversité des Populations en France (Trajectories and Origins: Survey on Population Diversity in France; TeO survey). This decision was based on the fact that the TeO is the first large-scale survey carried out to “improve knowledge about the social trajectories and living conditions of migrants and their descendants in French society” (see website and box below). These two sources combined formed the main body of the analysis for a two-fold interpretation of the association process, producing an unprecedented geographical reading for the four groups in an attempt to highlight the specificities of the integration process.

The TeO survey aims to identify the impact of an individual’s origins on his or her living conditions and social trajectory while factoring in other socio-demographic characteristics, such as social context, neighbourhood, age, generation, gender and level of education.
Issues of
integration and discrimination play a significant role in public debate. However, France still lacks national statistics for studying these issues. The TeO survey is designed to fill these gaps.
The TeO focuses on all populations living in metropolitan France, their current
living conditions and life courses. The survey takes a particular interest, however, in populations that may encounter obstacles in their trajectories due to their origins or physical appearance (immigrants, descendants of immigrants, persons from French Overseas Departments and their descendants).
The TeO survey was conducted jointly by
INED and INSEE. It was rigorously controlled by the agencies that oversee public statistics (CNIS and CNIL). It scrupulously respects the right of respondents, with all participants taking part in the survey on a voluntary and anonymous basis.
Data collection (from 22,000 respondents in metropolitan France) took place between September 2008 and February 2009.”


II. Is there a relation between the geographical location of associations and the integration of members of the four communities?

The contrasting geographies of the associations of the four population groups

14The geographies of the associations of the four population groups share some common features. Paris and the Ile-de-France region are generally the places with the highest concentration of associations, with high primacy indices (proportion of the region with the highest concentration in the national total) for associations related to Algeria and Vietnam (Table 1). On the other hand, the network of associations is relatively weak on an axis running from north-eastern France to the Massif Central; in Brittany; in the northern and eastern parts of the greater Paris Basin (Normandy, Picardy and Champagne-Ardenne); and in Corsica – i.e. on the axis where the population densities are lower on average.

Table 1: Associations linked to the four communities






Associations in Ile-de-France





Associations in France





Ile-de-France share





Source: Journal Officiel de la République Française, 2013; authors' calculations.

15There are some differences from one group to another (Map 1):

  • The network of associations for the Algerian community is concentrated mainly in the Paris-Lyon-Marseille axis (Rhône-Alpes and PACA).

  • Outside the Ile-de-France region (22% of the associations listed in France), the Portuguese network is concentrated in the south.

  • The Turkish network of associations is more nuanced with two major sites of concentration (Ile-de-France and Rhône-Alpes) but the primacy of the Ile-de-France region is relatively low (29%).

  • For the Vietnamese group, the main areas of concentration are found in the outlying regions, from Brittany to PACA.

Is it possible to identify the determinants of the geography of the community associations?

16The most significant explanatory variable for the geography of the network of association is the spatial distribution of the relevant populations. There is a very high correlation between the relative weight of each French region (expressed as a percentage) in the total population of the four groups and the number of associations (all areas) recorded in each region compared to the total number for France as a whole. The figures for the Algerian, Portuguese, Turkish and Vietnamese population were collected in the LiLi survey (see above). In all cases, these figures by region concern foreign or French persons residing in France but born in the country of origin. Accordingly, descendants born in France are not taken into account.

17Given the small number of regional units considered (only about twenty French regions), it is not possible to calculate a robust determination coefficient between the geography of the communities studied and the geography of the associations on offer. However, by comparing the geographical distribution of these two variables, we noted that there is probably a relationship, with some exceptions. The correlation is not clear for the Vietnamese community, doubtless because of its conspicuous concentration in the Ile-de-France region (44% of the community and 41% of associations). In the case of Turkish associations, the geographical correlation is less clear-cut also when excluding the Ile-de-France and Rhône-Alpes regions. Variables other than the geographical distribution of the population play a role in the geography of the network of associations but they were not identified.

18In calculating the number of inhabitants per association, we see that the ratio of associations varies significantly from one region to another within the same community (Map 2). The lower the number, the higher the ratio. This means that the number of associations in relation to the number of inhabitants considered is high: for the Algerian community, the number varies from 403 inhabitants per association (Pays de la Loire) to 2,665 (Lorraine); for the Portuguese community, from 317 (Nord-Pas-de-Calais) to 2,152 (Basse-Normandie); for the Turkish community, from 87 to 1,225; for the Vietnamese community, from 15 to 824. In other words, there are significant differences from one community to another. The Vietnamese community is the most likely to create associations, with 128 people per association, ahead of the Turkish (477), Portuguese (784) and Algerian (1,029) communities. Moreover, the density of the network of associations varies considerably from one region to another and from one immigrant population to another. The regions with the greatest numbers of associations in absolute terms are generally not those with high densities (Map 2) because the target population is usually the largest.

Variety of thematic profiles depending on regions and groups

19The Journal Officiel ranks associations in more than twenty different sectors, divided into numerous sub-fields. This information can be used to draw a fairly precise picture of the thematic range of the network and to spot nuances between groups and regions. Map 3 represents, for example, the number of associations in two fields for the four populations: cultural and artistic associations on the one hand, and social intervention associations on the other. The map shows that the relationship between social intervention and cultural activity is not the same from one group to another. Social action predominates in the Algerian group while the situation is reversed for the other three groups.

20Some common features may be highlighted by classifying associations by field:

  • Some fields often come at the top of the list: culture, arts-based activities and cultural activities dominates in the Vietnamese (23%), Portuguese (58%) and Turkish (30%) groups. The social intervention field dominates in the Algerian group (45%) and also ranks highly in the other groups. These two fields are in the top two for three population groups: Vietnamese, Turkish and Algerian. Associations in the area of social clubs, affinity groups and self-help groups or education and training often come top of the ranking, as do sports associations (10% of the total in the Portuguese population).

  • Some fields are under-represented: tourism, defence of economic interests, assistance with employment, religious, philosophical and spiritual associations, legal, etc.

  • Paradoxically, some fields that are theoretically intended to promote the integration of immigrant populations are under-represented (defence of economic interests, assistance with employment). Conversely, some well-represented fields do not, in theory, promote integration into the French population. Rather, they seem to favour self-segregated communities (social clubs, affinity groups and self-help groups together with culture, arts-based activities and cultural activity groups).

21The thematic range shows significant differences from one region to another that are not always easy to explain. Some important facts include:

  • The thematic range is complete only Ile-de-France, since this is the region where the populations of the four groups are the largest and the number of associations is the highest. It is followed by Rhône-Alpes and PACA and Nord-Pas-de-Calais in some cases (for Algerians and Vietnamese).

  • The ranking of associations in absolute numbers varies considerably from one region to another within a group. In the Algerian group, the social intervention field is predominant in all regions except Limousin and Corsica. For Portuguese associations, the culture, arts-based activities and cultural activities field easily dominates in 17 regions although not in Lorraine, Franche-Comté and Midi-Pyrénées, where the thematic profiles are more balanced.

Does the density of the network of associations have an impact on economic and social integration?

  • 5 “L’insertion professionnelle des immigrés et de leurs descendants en 2010”, Infos Migrations No. 31 (...)
  • 6 In this section, the populations considered are made up of persons born abroad.

22Immigrants are more strongly affected than nationals by unemployment, especially immigrants from non-EU countries.5 According to INSEE, the unemployment rate of active immigrants6 aged 15 and over was 16% (20% for non-EU immigrants and 8% for EU immigrants), while it was only 9% for non-immigrants. Differences are observed at all skill levels and across all age groups. There are also strong variations depending on the country of origin: for immigrants aged 25 to 64, it is only 7% for EU nationals, compared to 25% for Turks, 22% for North Africans and 19% for other Africans. Only the acquisition of French nationality is correlated with a lower unemployment rate: these individuals have been living in France for longer, are more familiar with the host society or have had time to build a network that can help them find a job.

23It may be hypothesised that the community-based network of associations strengthens relations within a community and does not have a positive impact on access to employment. However, this is not exactly what we see when we compare and contrast the figures. Simple linear regressions between the absolute number of community associations in the different regions of France and the unemployment rate of the members of the four communities shows that there is in fact no positive or negative correlation between the density of the association network and access to employment. For three of the groups included in the INSEE statistics (Algerians, Portuguese and Vietnamese), the unemployment rate was compared to the number of associations, followed by the unemployment rate and density of associations (the number of people per association in the regions). This leads to the conclusion that community associations play neither a positive nor a negative role in the economic integration of the members of the communities. This method does not, however, produce robust results because the sample size is once again too small. Although this method is rough and ready, comparing the geography of the indicators should make it possible to see whether their distribution across France is superimposable. We applied this method to other indicators that provide information about the economic and social integration of immigrants: how long they have been looking for employment, the highest level of education and the proportion of homeowners. Once more, there was no correlation. The lack of conclusive results may be due to the highly aggregated nature of the data, which does not make it possible to establish a more precise causality ranging from a particular individual’s membership of an association to the probability of having a job, etc. The TeO database provides the means to analyse this causality.

III. Adherence to oppositional values and integration through employment: the role of belonging to an association

24To analyse the link between membership of an association and the probability of finding a job, we referred to the conceptual framework of the oppositional values model developed by Battu and Zenou (2010). We used the tendency among members of the four communities to belong to associations in general, community-based or otherwise, as a variable.

25Associations can be a vector of integration or, on the contrary, of social segregation. If they reduce transaction costs between a minority group and the rest of society, they integrate the individuals in the group. When they aggregate a minority group around communitarian values, they raise the costs of integration and segregate individuals. The group refuses the positions of the dominant group, and individuals bear the cost of this rejection since adherence to oppositional values reduces the probability of finding a job. This was the analysis undertaken by Dominique Baillet (2001) on Maghreb organisations in France, when he drew the distinction between organisations that promote communitarianism and thereby widen the gap with the dominant cultural group; and organisations that make it easier for their members to integrate into this group.

26Current thinking shows that adherence to oppositional values determines participation in and integration into the labour market. According to Berthoud (2000), ethnic-minority employment in Great Britain is based on identity variables, which are grounds for exclusion from the work place. Blackaby et al. (1997) suggest that some minority groups prefer to withdraw into themselves (communitarian and identitarian fall back or closure), which restricts economic opportunities and increases unemployment rates. These studies hypothesise that integration into the labour market depends on the cultural distance between the job seeker and the dominant culture of the host country. Variables such as the level of fluency in the host language and (past) exposure to racist behaviour and harassment, as well as membership of an association, have an impact on this distance and thus affect the probability of integrating into the labour market.

Econometric procedure and model

  • 7 The results are not published here for reasons of space; they are available, however.

27As with Battu and Zenou (2010), we assume that membership of an association does not alter the probability of finding a job directly7 but indirectly, through adherence to oppositional values. In the four communities we studied, being part of an association influences variables such as the feeling of being French, being perceived as being French and distrust towards French institutions, which are markers of adherence to oppositional values and identity choice. They alter the probability of finding a job by affecting the distance to the dominant cultural group. They act as instruments, helping to identify the impact of adherence to oppositional values on the probability of finding employment. The influence of associations is indirect: it is because an individual perceives him or herself more as being French or has more trust in French institutions – and not because he or she joins an association – that the individual has more chance of finding a job.

28The identification strategy is a two-stage process. In the first stage (results in Table 1), the variable of interest, which is membership of an association, explains the adherence to oppositional values according to the nature and functioning of the associations that the TeO database allows us to consider (see box below):

Adherence to oppositional values = a2 membership of an association + b2 controls

29In the second stage (Table 2), adherence to oppositional values determines the probability of being in employment.

Employment = a1 adherence to oppositional values + b1 controls

List of variables used

Employment = 1 if an individual is professionally active and in employment.
Adherence to oppositional values 
Legal system = 1 if an individual states that he or she has “little” or “no” trust in the French legal system.
Government policy = 1 if an individual states that he or she has “little” or “no” trust in French policy.
Public bodies = 1 if an individual states that he or she has “little” or “no” trust in public employment bodies in France.
Schools = 1 if an individual states that he or she has “little” or “no” trust in schools in France.
Feels French = 1 if, in response to the statement “I feel French”, an individual answers “tend to disagree” or “strongly disagree”.
Feels at home = 1 if, in response to the statement “I feel at home in France”, an individual answers “tend to disagree” or “strongly disagree”.
Seen as French = 1 if, in response to the statement “I am seen as French”, an individual answers “tend to disagree” or “strongly disagree”.
Frequency of police ID checks = 1 if, in response to the question: “How often do you get asked about your origins?”, an individual answers “often” or “sometimes” (as opposed to “rarely” or “never”).
Membership of an association
Sport = 0 if an individual is in a sports association.
Parent-teacher = 0 if an individual is a member of a parent-teacher association.
Charitable = 0 if an individual is a member of a charitable and mutual aid association.
Human rights = 0 if an individual is a member of a human rights and anti-racism association.
Culture = 0 if an individual is a member of a cultural association.
Religion = 0 if an individual is a member of a religious association.
“Control” variables
Not married = 1 if an individual is single.
Local neighbourhood % = 1 if, in response to the question: “What percentage of immigrants live in your neighbourhood?”, an individual answers almost all or more than half.
Immi1 = 1 if the immigration status of an individual is “immigrant”.
Immi2 = 1 if the immigration status of an individual is “descendant of one or two emigrants” or “descendant of one or two parents born in a DOM (French Overseas Department)”.
Immi1c if the status of an individual’s current spouse is “immigrant”.
Immi2c if the status of an individual’s current spouse is “descendant of one or two emigrants” or “descendant of one or two parents born in a DOM”.
Police checks = 0 if, in response to the question: “How often are you stopped by the police (or gendarmerie) for identity checks?”, an individual answers: “Never”.
Racism = 0 if, in response the to the question: “Have you ever been the target of racist remarks or attitudes in France?”, an individual answers: “No”.
Forced marriage = 1 if an individual states that he or she did not want to marry but was forced to do so by family pressure.
Religious education = 1 if, in response to the question “What was the importance of religion in the education you received?”, an individual answers: “Very important”.
Woman = 1 if an individual is a woman.
Age is the age of an individual.
Age on arriving in France is the age at which an individual came to France.
French level = 1 if the level of French is very good.
No qualifications = 1 if an individual has no qualifications.
CEP (French certificate of primary education) = 1 if an individual has a CEP or equivalent foreign qualification.
Brevet (equivalent to GCSE) = 1 if an individual has a brevet des collèges (junior college certificate), BEPS, brevet élémentaire (elementary certificate) or equivalent foreign qualification.
CAPBEP = 1 if an individual has a CAP (equivalent to an NVQ), BEP (certificate of vocational proficiency) or an equivalent qualification.
BAC (French baccalaureate) = 1 if an individual has a technological or vocational baccalaureate or a qualification of the same level or a general baccalaureate (A, B, C, D, E, ES, L, S), a brevet supérieur (higher certificate), capacité en droit (legal studies diploma), a DAEU (access to university qualification) or an equivalent foreign qualification.
Higher Ed = 1 if an individual has a qualification that is at least the equivalent of a BAC + 2 (two-year higher education diploma).

30This two-step identification strategy has the advantage of addressing the endogeneity between the choice of oppositional identity and the fact of having a job or not. Being unemployed or not having a job clearly has an impact on the feeling of being French, being perceived as French, feeling at home in France or the trust an individual has in French institutions. Furthermore, whereas adherence to values that reject French identity may explain insertion into the labour market, the opposite is just as true. We are thus faced with a circular relation, which is in no way a causal relation but a simple correlation between two variables that expressed or measured the same thing.

31The hypothesis that underpins our method is that membership of an association reduces oppositional behaviour. frequency of police ID checks, exposure to racist reactions and cultural activities that are far removed from the dominant practices in France will also promote the adoption of these behaviours. Some associations will promote integration or assimilation behaviours, just as other variables (racism, different cultural activities) discourage them. All these variables form part of the equation of adherence to oppositional values, and are instrumental, because they only explain employment through their impact on the construction of an identity that rejects or accepts French society.

32This hypothesis is strong and certainly debatable. It is highly possible that membership of an association directly alters the probability of finding a job, because an association may function as a network. If the tests on the validity of our instruments mean we can accept the hypothesis of exogeneity, according to which the impact of being a member of an association on employment is only indirect, we do not believe that these statistical tests have a definitive value. Furthermore, we highlight the fact that our results are valid only within the limits of the modelling that we selected.

33The two variables explained – rejection of the dominant culture and the probability of finding a job – are dichotomous variables (taking two values: 0 and 1). The two econometric models are therefore probit models. Estimating the first step makes it possible to construct a rejection of the dominant culture variable as predicted (or theoretically) by the instruments. This prediction is used in the employment equation instead of the various variables we selected to measure adherence to oppositional values: the feeling of being French, the feeling of being perceived as French (feels French, at home, seen as French) and distrust of French institutions (legal system, government policy, public bodies and schools).


34The results are shown in Tables 1 and 2. Table 1 explains adherence to oppositional values. It has eight columns corresponding to eight proxies of adherence to oppositional values (see box for descriptions of the variables used): I feel French; I feel at home in France; I am seen as French; assessment of the justified or excessive nature of police checks on an individual’s origins; level of trust in the French legal system; level of trust in French policy; level of trust in public employment bodies in France; and level of trust in schools in France.

35The results show that belonging to certain associations has an impact. The proxies of adherence to oppositional values: I feel French, I’m seen as French and I feel at home in France are influenced by membership of sports (Sports) and Charitable (Charitable) associations. The former improve integration, while the latter affect populations that perceive themselves as different, and therefore tend to reduce integration. Belonging to cultural associations (Culture) has a particularly high explanatory power: it strengthens trust in the legal system (Legal system), government policy (Government policy) and public employment bodies (Public bodies). Finally, belonging to a parent-teacher association (Parent-teacher) plays an integrating role by improving trust in the French legal system (Legal system), and, of course, in the school system (Schools).

Other variables

  • Unwarranted ID checks by the police or gendarmerie (Police ID checks) and acts described by respondents as racist towards them (Racism) are significant in almost all the specifications, and logically reinforce the distrust of foreigners in the sample in the host country.

  • Cultural activities that are overtly alien to dominant practices in France, such as forced marriage (Forced marriage), have no impact, curiously, on the variables measuring integration. A religion-focused education (Religious education) does not play a role either.

  • A good level of French (Level of French), on the other hand, favours integration by reducing adherence to oppositional values. This result is recurrent in the literature (see, for example, Constant and Zimmerman, 2008).

  • When the percentage of foreigners (local neighbourhood %) living in a “quartier” is over half, it favours adherence to oppositional values.

  • Women (Women) coming to France when they were young (Age on arriving in France) and single people (Not married) are better integrated.

  • Finally, the level of education (No qualifications, CEP, Brevet, CAPBEP, Bac, Higher Ed) also plays a part.

36Table 2 describes the impact of different proxies of adherence to oppositional values on the probability of finding a job. Regardless of the marker of adherence to oppositional values (not feeling French, not feeling at home in France, not being perceived as French, not trusting French institutions), the probability of access to employment is significantly reduced.

37The other variables have the expected indicator regarding the probability of finding a job. Thus:

  • The nationality of the spouse or parents of the spouse does not play a role.

  • The proportion of foreigners living in the neighbourhood and the level of French influence the probability of being employed in the expected direction (negative and positive respectively).

  • Women, young persons, unmarried people and late arrivals in France are employed less.

  • Finally, education is positively correlated with access to employment.

Table 1: Adherence to oppositional values










(non participation in an association)

Do not feel French

Do not feel at home

Not seen as French

Freq. of police checks

Distrust the legal system

Distrust the police


public bodies

Distrust school




















































Human rightsa




































































Police checks


































Forced marriage

















Religious Education



















































Level of French

















Local neighbourhood %



















































Not married

















Age on arrival

in France


































No qualifications



















































Higher Ed.











































Standard deviation in parentheses
*** p <0.01, ** p <0.05, * p <0.1
a: A positive and significant sign (one or more asterisks, in bold in the table) indicates that non-participation in an association (variable takes the value of one) increases distrust in institutions and decreases the perception, by the subject and others, of being French, as well as of feeling at home in France.

Table 2: Probability of accessing a job









At homeb



Feels Frenchb



Seen as Frenchb



Freq. of police checksb



Legal systemb



Public bodiesb










































Local neighbourhood %

















Level of French



















































Not married

















Age on arrival in France


































No qualifications



















































Higher Ed












































Standard deviation in parentheses
*** p <0.01, ** p <0.05, * p <0.1
b: A negative and significant sign in the employment equation indicates that adherence to oppositional values decreases the probability of access to employment.


38Do associations work on behalf of integration, soften the fate of those suffering from discrimination who, in certain geographical areas, are mostly of foreign origin? The Haut Conseil à l’Intégration draws our attention to the fact that people born abroad of foreign nationality at birth, and descendants of immigrants, account for over half of the people aged between 18 and 50 living in ZUS (Sensitive Urban Areas). Against this worrying background, should we invest in associations to help achieve integration? More broadly, does France still know how to integrate immigrants? Surveys confirm that the French people are mostly open to immigration, on the express condition that immigrants and their children adapt to French values and customs; in other words, provided that they integrate. We may hypothesise that associations would be an effective lever.

39There are considerable difficulties involved in working on the integration of immigrants. There are no adequate databases, meaning we had to make do with what already existed in order to evaluate the effectiveness of associations in terms of integration. For example, the national directory of associations does not have a specific heading dedicated to integration, and associations are listed under thematic headings such as defence of fundamental rights and civic activities; cultural activities; education and training; social intervention; etc. It is no great surprise, therefore, not to find a correlation between the density of associations and integration indicators about the labour market, if the objective of the associations is not to integrate. The TeO database used in the third section suffers from the same problem. Moreover, it does not provide information on the ethnicity or community membership of an association, which is a well-known characteristic in France.

40A brief reminder about the history of immigrant associations in France helps us quickly understand that integrating foreign populations – especially economic integration – is not the officially-stated objective. Rather, the aim is (for example) to improve housing for North African families who settle in France or to fight for equality between French and foreigners or to combat racism (SOS Racisme). This article shows that belonging to non-community associations influences identity choices, which in turn affect the probability of finding a job. An individual is less likely to integrate into the job market if he or she does not feel French, does not feel at home in France, is not perceived as French and does not trust French institutions. It is through these feelings and perceptions that membership of an association is likely to play a role as a social integrator.

41The results should be interpreted with caution. They do not make it possible to decide precisely which associations should be funded: associations that claim to be engaged in integration or those more official associations that are in a partnership with the state or local authorities. Nor can the results be used to arbitrate between a policy based on subsidising associations and the wide variety of measures taken in recent years that have been translated at national and regional level. But they certainly do put into perspective the decline in the number of associations funded by the state. Some time before the Haut Conseil à l’Intégration was dissolved, its 2011-2012 report to the Prime Minister stated: “In France, in ten years, the number of associations financed by the state and working specifically in the field of integrating immigrants and their direct descendants has fallen by 80%. There were over 6,000 of these associations in 2000, receiving grants totalling 1.26 billion francs or nearly 200 million euros. In 2010, there were only 1,300 with a budget of 100 million euros. Nevertheless, this marks an increase in average funding per activity.” (HCI, 2012: 16). In 2013, the French government decided to refocus its support exclusively on recently-arrived immigrants. This change of policy will not be without consequences: “Under this new system, many associations have lost their footing and have either been refused funding or have given up suggesting projects of their own accord. This has also been the case for associations involved in culture and cultural mediation and the field of schools and parenthood. Associations that wanted to continue to receive funding have been forced to refocus, usually artificially, on the single category of newcomers, performing gymnastics in order to segment their members at the same time as being obliged to seek other sources of funding to maintain their activities for wider audiences” (Bernardot, 2017: 107).

Haut de page


Amiraux, V., Simon, P., 2006, "There are no Minorities here : Cultures of scholarship and public debate on immigrants and integration in France", International Journal of Comparative Sociology, Vol.47, 191-215

Baillet, D., 2000, « Les associations issues de l’immigration, un outil d’intégration ? », Migrations société, Vol.12, No.72, 55-66

Baillet, D., 2001, « Militants d'origine maghrébine et intégration », Sud/Nord, Vol.14, 91-103

Belorgey, J.-M., 2000, « Les étrangers face à la liberté d’association ou quelle laïcité ? », Migrations société, Vol.12, No.72, 45-54

Bernardot, M.-J., 2017, « Les associations et l’administration nationale chargée de l’intégration des immigrés: une analyse depuis l’intérieur », Migrations Société, Vol.170, No.4, 91-108

Berthoud, R., 2000, "Ethnic employment penalties in Britain", Journal of Ethnic and Migration Studies, vol.26, 389-416

Blackaby, D., Drinkwater, S., Leslie, D., Murphy, P., 1997, "A picture of male and female unemployment among Britain’s ethnic minorities", Scottish Journal of Political Economy, vol.44, 182-97

Beauchemin, C., Lagrange, H., Safi, M., 2011, "Transnationalism and Immigrant Assimilation in France : Between Here and There?", Paris, Ined, Working Paper, No.172, 42 p.

Boitard, F., 2001, « L’état et les associations, entre méfiance et allégeance », Hommes et Migrations, No.1229, 5-9

Brubaker, R., 2001, "The return of assimilation? Changing perspectives on immigration and its sequels in France, Germany and the United States", Ethnic and Racial Studies, Vol.24, 531-548

Cattacin, S., en collaboration avec La Barba, M. (en collaboration avec), 2007, Migrations et Associations. La vie associative des migrants – une exploration de leur rôle et développement en Europe occidentale, Genève et Neuchâtel, Département de sociologie/SFM.

Constant, A.F., Zimmerman, K.F., 2008, "Measuring ethnic identity and its impact on economic behavior", Journal of the European Economic Association, Vol.6, 424-33

Dumont, A., 2010, « L’état des recherches sur les associations de migrants internationaux », Revue européenne des migrations internationales, Vol.26, No.2, 117-137

Eme, B., Neyrand, G., 1997, Associations de proximité et processus d’intégration, Paris, Direction de la Population et des Migrations

Haut Commissariat à l’Intégration (HCI), 2012, Investir dans les associations pour réussir l’intégration, Paris. La documentation française

Hily, M.-A., Poinard, M., 1985, « Fonctions et enjeux du mouvement associatif portugais en France », Revue européenne des migrations internationales, Vol.1, No.1, 25-35

Kastoryano, R., 1989, « L'État et les immigrés : France, Allemagne, Grande-Bretagne et États-Unis », Revue européenne des migrations internationales, Vol.5, No.1, 9-20

Khelil, M., 2005, Sociologie de l’intégration, Paris, PUF

Niboyet, J.-P., 1946, Le décret-loi du 12 avril 1939 relatif aux associations, Travaux du Comité français de droit international privé, 6e année, 1938-1939, 53-56

Noiriel, G., 1988, Le creuset français, Paris. Seuil

Poinsot, M., 2000, « Associations de l’immigration et mouvement associatif français », Migrations société, Vol.12, No.72, 45-54

Rea, A., Tripier, M., 2008, Sociologie de l’immigration, Paris, La Découverte

Safi, M., 2006, « Le processus d’intégration des immigrés en France : inégalités et segmentation », Revue française de sociologie, Vol.47, 3-48

Schnapper, D., 2007, Qu’est-ce que l’intégration ?, Paris, Seuil

Paugam, S., (dir.), 2014, L’intégration inégale. Force, fragilité et rupture des liens sociaux, Paris, PUF

Streiff-Fénart, J., 2006, « A propos des valeurs en situation d’immigration. Questions de recherche et bilan des travaux », Revue Française de Sociologie, 47-4, 851-875

Weil, P., 1991, La France et ses étrangers. L’aventure d’une politique de l’immigration 1938-1991, Paris. Calmann-Lévy

Zenou, Y., Battu, H., 2009, "Oppositional identities and employment for ethnic minorities: Evidence from England", The Economic Journal, Vol.120, 52-71

Haut de page


1 We should note that Schnapper was able to convey this view when writing: "Public policy can only aim to integrate all populations through citizenship and participation in economic activity, supplemented by the protection of the welfare state. No democratic government would dream of pursuing a policy of exclusion. It is the terms and conditions of these policies, therefore, that require analysis" (2007: 203).

2 These notions echo the problem of nationhood as defined by Benedict Anderson. As Antoine Dumont points out: “Rather than abandoning the idea of associative ethnicity, it would be better to refound it on what is at the root of many migrant associations, namely their attachment to their national origins and their sense of national belonging. This is the feeling that Benedict Anderson calls ‘nation-ness’ as opposed to nationality: the latter opens the door to political rights, while the former refers to the feeling of belonging to a ‘national community’ defined by a history, culture, religion and language (Anderson, 2002). This distinction helps us understand that possessing the nationality of a state does not always go together with a sense of belonging to the nation represented by that state” (2010: 128). According to Dumont, to understand associative nation-ness we need “not only to cross-reference the theories of nationality and ethnicity but also to take into account the fact that this nation-ness is expressed at a distance, based on a state other than that to which this nation-ness is bound” (2010: 128).

3 From the notice issued by the Haut Conseil à l'Intégration (p.13) entitled: Investir dans les associations pour réussir l’intégration, 2012, 82p.


5 “L’insertion professionnelle des immigrés et de leurs descendants en 2010”, Infos Migrations No. 31, DSED, January 2012.

6 In this section, the populations considered are made up of persons born abroad.

7 The results are not published here for reasons of space; they are available, however.

Haut de page

Table des illustrations

Titre Example of a form declared in the Journal Officiel
Fichier image/, 93k
Titre Map 1
Fichier image/, 287k
Titre Map 2
Fichier image/, 269k
Titre Map 3
Fichier image/, 264k
Haut de page

Pour citer cet article

Référence électronique

William Berthomière, Mathilde Maurel et Yann Richard, « Volunteer Associations and Integration of Immigrants in France », Cybergeo : European Journal of Geography [En ligne], Espace, Société, Territoire, document 749, mis en ligne le 06 mars 2019, consulté le 22 mars 2019. URL : ; DOI : 10.4000/cybergeo.31814

Haut de page


William Berthomière

Directeur de recherche CNRS
PASSAGES - UMR 5319 CNRS – Maison des Suds, Pessac, France

Mathilde Maurel

Directrice de recherche CNRS,
Université Paris 1 Panthéon-Sorbonne, Centre d’Economie de la Sorbonne, France

Yann Richard

Professeur, Université Paris 1 Panthéon-Sorbonne, France

Articles du même auteur

Haut de page

Droits d’auteur

© CNRS-UMR Géographie-cités 8504

Haut de page