Researching ‘Hard to Reach’ Groups: Some Methodological Issues.


Lesley Jones and Barbara Allebone

Goldsmiths University of London.


The aim of this paper is to discuss some of the methodological issues which have arisen in the context of two research projects focussing on the education of Somali children in London. The projects are concerned with the links between children’s learning of numeracy at home and at school. In both projects the research is designed to collect data from parents, children and the schools. One focussed on children in Reception year, the other on children in Y7.


Many of the Somali immigrants in the U.K. have entered the country relatively recently and many of the parents are only just beginning to learn English. For these reasons the group who form the focus of the research can be considered to be a ‘hard to reach’ group (Toomey, 1996). Tomlinson (1993) takes issue with this term, seeing it a euphemism for those working class groups regarded as problems, but recognises that teachers
‘have difficulty informing themselves about the lives and backgrounds of ethnic minority parents and continue to resort to stereotyped beliefs about ethnic minority homes.’
Views of the parents’ role.

The role of parents in their children’s education has long been recognised as a significant factor in educational success and school improvement (Epstein, 1996, Safran, 1996). In recent years we have reached the stage where certain educational organisations and international conferences concentrate almost entirely on the issue of partnership between schools and parents (E.g., Parents in Education Research Network, European Research Network about Parents in Education. Education is Partnership Conference, Copenhagen, November 1996). Within this area of interest lies a vast spread of concerns and purposes.


Parents clearly have rights in terms of their children’s education. Hughes (1994) provides interesting insight into the notion of ‘parents as consumers’, a view promoted in the U.K. under the conservative government and continued under the present government. Within this view is the associated idea of education as a commodity. Parents are given the right to ‘shop’ for their children’s education in the school of their choice. Vincent and Tomlinson (1997) suggest that this view of parent power, together with the notion of schools’ partnership with parents is little more than rhetoric. In reality there is little opportunity for parents to exercise an individual or collective ‘voice’ which will have an effect on the children’s school experience.

An alternative view of parents is that they can be seen as a ‘problem’ for teachers. In particular children are frequently judged to come from ‘poor backgrounds’, from a home environment which is unsupportive to the school and unsupportive of the educational process. e .g. Tizard and Hughes, (1984), p18. refer to,

a widespread belief amongst educationalists that working class parents do not stimulate their children adequately and in particular do not develop their language.’
Such views of parents were radically challenged by researchers such as Tizard and Hughes (1984) and Walkerdine, (1988) who recognised the substantial contribution made to their children’s pre-school education by parents from working class as well as middle class backgrounds.

However, these views persist and very recently David Reynolds, in a discussion about the introduction of a national campaign for family numeracy was attributed with the remarkable statement that,

‘At present most children get no support at home. … Many parents would feel able to help with their children’s reading, but it’s clear from everything we know that parents find it much harder to engage with numeracy or maths. At home there is almost zero provision. If we were able to generate some provision, some parental skills where at the moment there are low levels, the family background would have an enormously increased effect.’ (T.E.S, 14.11.97)
Many projects have sought to develop work with parents starting from this ‘deficit’ view of the family background and set out to educate the family in order to help to educate the child. Even where the deficit model is less apparent projects are designed to suggest activities which develop school methods at home ( e.g. Merttens and Vass, 1990, National Numeracy Project, 1997). However, as argued elsewhere (Jones, 1996) this appears to offer the parents the chance to participate in the culture of the school, but offers no opportunity for them to recognise the contribution of their own knowledge and social background to their children’s education. More recently projects have developed in which there is a more equal notion of partnership developed between the school and the community and in which the richness of the home environment is recognised. (E.g. Bouchard, 1996, Civil, 1996, Macbeath, 1996) Findings from the first Somali project mentioned above suggest that these children were indeed receiving support from the family at home and that the oral tradition was particularly helpful in helping children to acquire numeracy. (Jones & Farah, 1995)

It was thus with a view of parents as contributors to their children’s education that the projects were designed. We knew that parents and older siblings took an active part in developing young children’s number knowledge, but that the parents of young children had expressed concern about their ability to support children as they progressed through KS2 and into KS3.


Somali children in the U.K.

However, reaching parents of particular groups continues to be problematic. In particular it can be problematic and expensive to work with ethnic groups in England, where English is not spoken fluently in the home. It is acknowledged amongst researchers into family involvement that there are groups who are regarded as ‘hard to reach’. It was with such groups that the present studies were conducted. Many London Boroughs work with children who come from ethnically diverse backgrounds. It is not unusual to find children from 20 different ethnic groups in one school, some of whom will have been in England for only part of their life and whose language skills are still developing.

There is clear evidence that children in the U.K. are underachieving in major aspects of mathematics in comparison with their counterparts in other countries (Reynolds & Farrell, 1996) and that within this pattern is a very mixed experience for children of different ethnic backgrounds (Gillborn & Gipps,1996, Smith and Tomlinson, 1989). It is difficult to locate precise information about specific ethnic groups, but information gathered through the first project (Jones and Farah, 1995) suggested that the education of Somali children was a concern to their parents when they reached the age of entering secondary school. Evidence from this project suggested that the Somali culture, with its rich oral tradition, offered very useful learning experiences in the home environment, which could prove very helpful to children in their development in mathematics in school. The focus of the current project then became to identify factors which influenced these children’s attainment as they progressed through the U.K. school system. In particular we wanted to identify significant home numeracy practices, so that the findings could be disseminated to teachers and used to inform their practice. The view of learning taken in the research is one which does not separate the psychological from the social aspects. (Cotton and Gates, 1996) The research was to focus on one particular cultural group and to attempt to track down significant cultural practices which supported or otherwise affected school experience. However, it was anticipated that it would be possible to extend the findings to other groups and find effective ways for parents to develop and continue their own cultural practices in a way which would support the children’s work in school.

The research methodology was designed as an ethnological case study, originally intending to focus on children in year 6, but eventually using a small group of children in Year 7 in two single sex schools. Data was collected by participant observation within mathematics classes, informal interviews with children and teachers at school and semi-structured interviews at home with parents and children. The appointment of a research assistant fluent in the Somali language was considered an essential prerequisite to the success of the project.


Methodological issues.

Locating the target group

In both projects there were great difficulties in initially locating a target group of pupils. Comparatively large numbers of Somali pupils were identified in the two London Boroughs concerned. Staff in the education office or at the ESOL centre gave assurances that there were large numbers in particular schools. However, despite these reassurances, in both projects we had the experience of tracking down a school which sounded as if it contained a sizeable sample of Somali pupils, only to find that the families had recently moved and the numbers of Somali pupils in the school diminished. Somali families, it seems, retain their nomadic lifestyle on arrival in London. Eventually a decision had to be made which changed the target age from Y6 (final year in primary school) to Y7 where there was a sufficiently large group of pupils to make the research viable.


Making Contact with the Homes

In the first project we worked with children of Reception age. The research assistant was convinced that initial contact by letter would not be helpful. We did not know whether the adults in the family could speak or read English or Somali. We solved this problem by ‘hanging around’ the school gate or the classroom door and making direct contact with the parents or the older children who collected the infants. This personal contact worked well and guaranteed a warm welcome when we arrived at the homes at the appointed times.

In the second project we had the advantage that the research assistant was working within one of the schools and knew a number of the Somali families through his additional community work. It was clear was that here again personal contacts were the most effective and it was through this method that the first home visits were arranged. It was clearly an advantage to have prior contact with the community, and this prior knowledge of the families helped to build an ethnographic picture of the children in their home circumstances.


Gender and Culture

Neither of these projects could have taken effect without the services of an interpreter. In each project a research assistant was appointed and a major part of their role was to interpret and communicate with the families. In the first project the research assistant was a Somali woman who had previous research experience, but no previous experience within the English school education system. In the second project the research assistant was a Somali man who was working in a state school and in a community school, but had no previous research experience.

A new issue which emerged in the second project was the question of gender and culture in relation to the research staff. In the previous project one of the authors (female) had worked with the female research assistant and conducted interviews in the homes with the mothers of the families. These had generally been conducted when the fathers were not at home, but when the men were home they had remained in a different room. In Muslim society it is not appropriate for an adult male to spend time with an adult female outside the family. In the new project the engagement of a male research assistant led to very different arrangements and relationships within the research process, particularly in the home context. We were able to interview men at home, which had not previously been the case and the men were more fluent and confident in their spoken English. This meant that the research assistant did not need to translate and it was easier to maintain a flow of conversation during the interview. There were times in the interviews with the women in the homes when the male research assistant would represent the interviewee’s views briefly, then give elaborate interpretations on his own behalf. In a society where gender roles are clearly defined and very different from each other the areas of expertise and interpretation of events seemed to be very different according to the gender of the parent. Each interview was conducted with only one parent from the family and in each case this meant there was a set of interactions from which the researchers were excluded and another set of interactions which we were permitted to enter.

In this kind of research the researchers are necessarily learning a great deal about the culture and background of the target group as they proceed. The researcher who acts as interpreter is also mediating your inculcation into the culture. Recognising your interest in all things connected with the culture, he/she enthusiastically ‘fills you in’ on many aspects which help you to build up the ethnographic picture which enables you to appreciate the central aspects of the research. For this reason we introduce the term ‘Inside Researcher’ to refer to the person who is working from inside the cultural framework. Their role becomes that of an enthusiastic tourist guide wanting you to love and appreciate their country and customs as much as they do. Where, then, in all of this, does reflexivity stand, as a research ideal? The ‘Outside’ researchers, in this case, white, English, middle class and female recognise (hopefully) the limitations of their educational and cultural background in the context of the present research. By the nature of the ethnographic design of the research their aim must be to enter into the cultural framework as far as possible. The Inside Researcher, who has been trained in research techniques and encouraged to allow the interviewees to speak for themselves, finds it hard to break free of his/ her cultural and gender role. The irony of the research methodology is that whereas the ‘outside’ researcher needs to get inside the culture and to be able to see the world through the eyes of the group being researched, the ‘inside’ researcher needs to break free of his/her cultural paradigm in order to be able to represent the views of the interviewees, particularly if they belong to a different sub-group.

Certain key issues emerge. Siraj Blatchford and Siraj Blatchford, (1997) refer to, ‘Powerful social actors who are able to impose their meanings on others'. They suggest that ‘researchers who are members of socially dominant groups will be unable to work effectively with certain respondents’ and that ‘In some projects it will be desirable to recruit and train interviewers drawn from the respondent’s own age, ‘race’ gender or ethnic group.’ Our findings suggest that of these factors, gender is a major consideration when working with a culture in which there is clear gender differentiation within its structures. However, the very nature of social dominance means that it is from the group who are dominated that it is most difficult to recruit researchers. One result is that for a generation, ‘hard to reach’ groups remain hard to reach.