Reporting back – key findings from Adults’ and Children’s Friendships across social and ethnic difference

By Sarah Neal

This two year qualitative project, funded by the Economic and Social Research Council, which explores the friendships of 8/9 year old children, and their parents,  in  ‘super-diverse’  localities  in  London, England came to the end of its ‘live’ period this summer. While the data collection and analysis phase of the project is complete the research team’s thinking and writing of outputs continue.

The research team members were Carol Vincent and Humera Iqbal both Institute of Education, UCL and Sarah Neal, University of Surrey. We outline here some of our key findings on friendship. Further details of findings and the full end of project report can be found on the project’s website

The project uses primary school sites as a lens through which to understand social relations in localities which are experiencing rapid urban population change. We examine the ways in which local primary schools work as sites of everyday encounter, exchange and friendship making in multicultural and rapidly gentrifying London geographies.

Through a focus on three primary school classrooms, we explore how difference is experienced, and whether and how friendships and friendship networks are made and maintained across ethnic and social class differences. Our aim is to identify what friendships reveal about the nature and extent of ethnic and social divisions in contemporary multicultural society. In short, does living in diverse areas mean local populations have diverse friendship groups?


Our research shows that the adults and children who took part in the study did make and maintain friendships with those socially and ethnically different to themselves. The children, especially, mixed across ethnic difference and social class; indeed very nearly all the children had close friends from a different ethnic group to themselves.

Even for the children however, the frequency of  cross  class  relationships  drops  when  ‘best  friends’  are  considered.  We  found  that  the   cultural and social diversity of both the local areas and the schools were valued by the vast majority of adult and child participants. Many of the adults also saw school as providing a source of friends for themselves as well as their children.

However the adults were more likely to have friends – both in their school and non-school networks – who were similar to themselves, and friendships  across  class  were  especially  infrequent.

Despite  this  propensity  for  ‘sameness’  in  adult   friendships, even light, casual relationships, such as an exchange of greetings when meeting daily in the playground were felt by research participants to be important for creating friendly atmospheres and a sense of belonging in the school and its immediate surrounds.

Collecting the data

The project is based on data from three primary schools in three different areas of London. The data comprise 114 interviews with children, parents, governors, and school staff plus approximately 300 hours observation and 78 children’s social  maps  illustrating  their   friendship networks.

In each school we focused on one Year 4 (8/9 year olds) class. The three primary schools are based in super-diverse localities with different degrees of ethnic and social mix as well as increasing, but varying, levels of gentrification, described below.

Key Findings – children’s  friendships  

    • Children’s   friendships   are   often   understood   by   adults   as as   fluctuating and short-lived. However, friendships   were   a   central   part   of   the   children’s   lives   and   the   children   told   us   how   important   friendships were to them. They spoke of their friendships causing incidents of conflict and pain, as well as enjoyment and pleasure.
    • Close mixed friendships across ethnicity, and to a lesser extent, social class were common amongst the children.
    • When we looked at who the children said their closest friend was, there were still a significant number of friendships across ethnic difference  (nearly  three  quarters).  There  were  far  fewer ‘best friend’  friendships  across  social  class  difference  (just  over  a  quarter  of  the  children).
    • Children’s   shared   classroom   play cultures   and   practices   (e.g. games,   music   and   so   on)   facilitated   mixing across difference, but this was less apparent where differences of gender and disability were concerned.
    • Those friendships which moved outside of school were less mixed than those within school, as parents  organized  the  children’s  out  of  school  time. 
Many of the children successfully negotiated multiple identities. For example those whose families had origins in other countries typically identified as Somali, Bangladeshi and so on, but they were also  clear  that  they  ‘came  from’  London. 
The children recognised ethnic and social difference and viewed diversity as commonplace, and an ordinary part of their everyday lives.

Key findings – adults’ friendships

    • The adult participants found schools to be a source of friendships for themselves, as well as their children. Their relationships with other parents varied from interactions consisting of casual greetings to close friendships.
    • However, adult friendship relations that emerged from encounters in the playground were characterised by fluidity, temporality and instrumentalism. That is, school based adult friendships were often understood as likely to last only as long as the children were in that particular school, and to be fostered, partly to help the children’s social integration.
    • We found that adult friendships operated on different levels of emotional intensity and had ‘thicker’   and   ‘thinner’   levels   of   intimacy.
    • In   the   study,it   was   clear   that   both   ‘enduring’   friendship   and more casual, friendship forms based on seeing others regularly can work as a wide ranging resource for forming social-local connections (e.g. having someone to ask for childcare help; having someone to ask for information about a school event or outing; having someone to take the children to go to the park or swimming; having someone to socialise with, involvement in local community events, etc) and capacity building (involvement in school events and activities,  ‘know   how’  and  awareness  of  the school world, knowing teachers and school staff ).
    • However, the generation of social capital resources (e.g. knowledge and information about education) tended to be limited to networks of adults similarly positioned in terms of ethnicity and in particular social class; that  is  ‘bonding’  social  capital.
    • Importantly, the vast majority of parents to whom we spoke voiced approval of the diversity of the areas in which they lived, and some had made a point of choosing the school because of its diversity. However, in their own friendships, both in and out of school, the parent participants mixed less than the children with those different to themselves. The adult friendships that were made across difference, were, like those of the children, more likely to cross ethnic difference than to cross class difference.
    • When we discussed with the parent participants why the degree of adult mixing across difference was small, they responded that making friendly overtures to people who were different involved considerable effort and risked social awkwardness (e.g. language barriers, not knowing what to say). Such encounters could be anxiety-inducing.  Making  friends  with  ‘people  like  me’  appeared  to   offer shared interests, a basis for trust, and points of reference from the start.
    • Some adults questioned and reflected on why they had relatively homogeneous (similar) friendship networks when they lived in diverse areas. Others  felt  that  having  friends  ‘like  me’  was   ‘natural’  and  ‘inevitable’
    • As a social site, the school world is stratified. By this we mean that the different forms and volumes of social and economic resources to which parents have access shape their relationship with the school and each other. For example, it was clear in all three schools that those parents who participated in the governing bodies and Parent Teacher Association (PTA) committees were middle class, and largely white British.
    • However, when children attend a primary school in a diverse locality, parents and children encounter and engage with social and cultural difference. Even if that engagement is mostly avoided or is slight, the routine, day-to-day presence of being in primary school worlds requires social interactions which can best be described as civic conviviality as diverse populations that are (loosely) connected through their collective use of a shared social resource/organisation such as schools.

We are grateful for the financial support of the Economic and Social Research Council (ESRC) (project grant ES/K002384/1). Our thanks also go to all the participants who took part in the project.


Please note: Blog entries reflect the personal views of contributors and are not moderated or edited before publication. However, we may make subsequent amendments to correct errors or inaccuracies.