Full Text Version

Sectarianism, Children and Community Relations in Northern Ireland.

  Chapter 4:
Towards an Alternative Methodological Approach


This chapter discusses how sectarianism and community relations work with children should be studied given the arguments contained in the previous chapters. By now it should be clear that the more structured and experimental tests, using essentially quantitative methods, are extremely limited in the insights they can offer. Rather, a methodological approach is needed which:

  • focuses on the perspectives and experiences of the children themselves;

  • grounds these experiences and perspectives within the broader social contexts within which the children are located.

Such a focus, as argued previously, will allow a more sensitive and nuanced understanding to develop of the complex, contradictory and context-specific nature of sectarianism in children’s lives. In doing so, it will also help to increase our understanding of the particular social processes and contexts which give rise to sectarian attitudes and behaviour and also of some of the underlying reasons why particular contact schemes are more successful than others.

Inevitably, such a focus requires more qualitative, ethnographic methods. It will be argued in this chapter that, alongside observing the children as they relate to one another and as they participate in cross-community contact, a significant emphasis needs to be given to conducting indepth, largely-unstructured group interviews with the children. It will be argued that interviews of this type provide an important source of data regarding the children’s experiences and perspectives. Such an approach raises certain methodological issues, however. Following a brief outline of the details concerning how such interviews were approached in the present, exploratory study, the chapter will then discuss two key concerns arising from this approach: the ethics of interviewing children directly about sectarianism and issues of how the data gained from these interviews can be validly interpreted.

p. 51 p. 52

Jump to Top

   Contents Page

The present study

The present study draws upon data from 30 largely unstructured group interviews with 10 and 11 year old children. The children attended P7 classes in two primary schools in Northern Ireland. The schools are located close to one another in an area that has experienced a relatively high level of violence and conflict over the past 25 years. To maintain confidentiality and also the anonymity of the children, psuedonyms are used throughout this report for the children, schools and local areas. The areas where the children live are simply referred to as ‘Northside’ (Protestant) and ‘Southside’ (Catholic). In addition, because of the sensitivity of the subject matter, certain non-essential details relating to the two areas have been altered to further disguise their identity. Access to the children was gained via the two schools. The children were interviewed in friendship groups of three with one child being chosen and asked to nominate two others to come with them for the interview. Interviews took place in a separate room away from the main class. All of the children from both classes were interviewed at least once and some were interviewed twice. Interviews usually lasted between 15 and 45 minutes. There was little structure to the interviews with children simply being asked what they did after school, where they went and what they enjoyed doing. These opening questions were usually all that was needed to stimulate discussion among the children which would soon take on a dynamic of its own. In order to gain some background information about the youth clubs that the children talked about and the surrounding areas, semi-structured interviews were conducted with the leaders of the two principle youth clubs which the children attended and the community workers for those areas.

One of the main strengths of using largely unstructured group interviews with the children is that we can begin to develop a feeling for and understanding of their own sub-cultures. As we will see in the chapters to follow, in allowing the children to discuss whatever they feel to be significant and/or important, we can therefore gain an invaluable insight into the central dynamics and influences of their own social worlds. Rather than imposing understanding and meaning on the children, which is always the danger of the more highly structured attitudinal tests discussed in Chapter One, we can therefore simply let the children ‘speak for themselves’. In doing so, when they do introduce the issue of sectarianism into the discussion, we can gain a good understanding of the particular social processes and situations which give rise to its expression. In other words, in allowing sectarianism to

p. 52 p. 53

Jump to Top

   Contents Page

be discussed ‘naturally’, within its own social contexts, we can begin to develop an understanding of its underlying causes. As will be seen, the interviews not only serve as important sources of data concerning incidents that the children have been involved in outside of school, but they also provide a good source of information about the dynamics of peer-group relations as played out in the arena of the interview and the influence that sectarianism can exert on these.

Ethical issues

One of the principle concerns when conducting these group interviews was the need not to inadvertently increase the children’s awareness and/or encourage them to think in terms of sectarianism. Neither did I want to place them in a situation where they felt obliged to discuss sectarian incidents which would then distress them (Connolly 1996). It was for this reason that the interviews were largely unstructured. The interviewer never raised the issue of sectarianism with the children and was confined to asking the type of general questions outlined above. Any discussions that then took place among the children about politics or sectarianism were therefore initiated by themselves. They therefore had full control about what topics they felt comfortable introducing and what they said about those topics.

A second major concern is that the act of simply allowing children to engage in sectarian discussion in an adults’ presence can easily give those children the impression that their views are acceptable and/or are being condoned by the adult (Troyna and Carrington 1989; Connolly 1996). As a result, whenever a child did introduce any sectarian ideas or beliefs, the interviewer would distance himself from them by interjecting into the discussion and asking the child(ren) what they meant by that statement, whether they thought it was a fair thing to say and/or how would they feel if they were treated like that. It is felt that such questions not only left the children in little doubt that their views were not supported by the adult interviewer but they also required the child to think about and justify beliefs that they may well have, until this point, taken-for-granted.

One understandable response to this type of approach could possibly be that we should not allow the children to discuss sectarianism at all and that, when they do attempt to introduce the topic, we should stop them and explain to them exactly why such beliefs are unacceptable. The problem with this approach, however, is that it simply ‘drives the problem underground’. While children will quickly learn

p. 53 p. 54

Jump to Top

   Contents Page

not to raise sectarian beliefs infront of the interviewer, it will not prevent them from engaging in and expressing them elsewhere. From informal conversations with the class teachers from the two sample schools, it was clear that the children were (quite rightly) not allowed to express sectarian beliefs in the classroom and, when they did, were appropriately dealt with by the teacher. However, as the following two chapters will demonstrate, this had little effect on the level of sectarianism found among the children themselves who simply learnt not to express it in the classroom.

Moreover, the data gained from the following two chapters illustrate just how deeply entrenched such beliefs are. If the report does nothing else, it should highlight the very real problem that sectarianism poses for children of this age. It is not a problem that is going to go away by refusing its expression in certain arenas and/or ignoring it. It is clear that we need to develop an understanding of exactly how sectarianism comes to influence children’s social worlds and the particular experiences, social processes and interactions that provide the principle contexts within which sectarianism progressively takes hold.

Validity of the data

The other main concern that needs to be addressed here concerns the status of the data gathered and how it can be interpreted and analysed. Two issues are often raised in relation to this issue: the problems of interviewer effects and the difficulty in developing generalisations from the data. I have addressed both of these concerns in detail elsewhere (see Connolly 1997a, 1998b) and do not intend to engage in a lengthy discussion here. It is important, however, to briefly outline the stance adopted in this present study on each of these issues in turn.

Interviewer effects

The first concern relates to the belief that what the children say in the context of a group interview is not an accurate reflection of their general attitudes and beliefs but is a reflection of the children’s attempts to ‘play to the audience’ - in this case the adult researcher. While the children may well express sectarian sentiments it could be argued that they are only doing it to either impress the adult researcher and/or undermine his or her authority. In response, I can only agree with this observation. It would be quite naive to believe that an adult researcher can sit around the table with three young children and still adopt the

p. 54 p. 55

Jump to Top

   Contents Page

role of a ‘fly on the wall’. The data generated from such interviews will inevitably, therefore, reflect the influence of the researcher’s presence. Whether this invalidates the data is another question however. Elsewhere I have argued that it only invalidates the findings of the research if this influence is not taken into account in the interpretation of the data (Connolly 1997). Thus we have to be mindful of the presence of the researcher and the likely effects of this on what the children say and do. As a result we cannot claim that the picture gained of the children’s social worlds from the interviews is a representative one. Neither, however, can we claim that it is a false one. In essence, once we accept that children are socially competent and able to interpret and respond to differing social situations in a variety of ways, then we can never claim to find the, true authentic voice of children. Rather, children can only ever be said to have a multiplicity of voices which are brought to the fore and adapted at differing times from one context to the next.

This is equally true for more structured attitudinal tests whose aim is to approach the study of children’s attitudes more ‘scientifically’ so as to uncover the underlying core of beliefs that they hold. However, even with this approach, it can still only be seen as relevant to the particular social context within which the data was collected (i.e. the ‘laboratory’). It would be equally naive to presume that the attitudes found within that particular attitudinal test could then be assumed to exert a consistent guiding influence on that child whatever the social situation is that they find themselves. With sociometric tests on children and racial prejudice for instance, it has been shown that a relatively high level of ‘same-race’ preference in relation to friendship groupings that children express in tests has been completely contradicted when those same children are then observed playing together in the playground (Denscombe et al. 1986). What the children say and do in one context, therefore, cannot be assumed to be generalisable to other contexts however structured and ‘scientifically-rigorous’ the methods used.

While we cannot therefore claim that what the children have said in this particular study is completely representative of their actions and behaviour, we can make three claims about the data which is equally important for the present study. First, in a climate where even the ability of young children to be sectarian is questioned, this data shows that, at least for some children, sectarianism is a major problem. While we cannot generalise from this to make claims about the scale of the

p. 55 p. 56

Jump to Top

   Contents Page

problem, we can use the data to effectively counter the claim that children are incapable of being sectarian and, in contrast, begin to develop the argument that anti-sectarian work (at an appropriate level) is therefore important among this age group (Playboard NI 1997; Connolly 1999). Second, we can gain an appreciation of some of the social contexts within which the expression of sectarianism is more likely to occur. While we do not have the ability to generalise and claim that they are the most frequent or significant, we can still, nevertheless, claim that the particular social processes highlighted from the data do contribute to the manifestation of sectarian attitudes and behaviour among the children. This understanding of the causes of sectarianism, however incomplete, is nevertheless important in its own right. Third, and finally, we can make claims about the importance that the children give to sectarianism in relation to their own developing identities. If we accept that the children will attempt to impress and/or challenge the authority of the researcher, then we can assume that the sectarianism they introduce to do this reflects what they value and aspire to in relation to becoming an adult. Presumably, they would not choose to express what they felt to be ‘childish’ ideas in order to impress an adult researcher. Similarly, one strategy that could be adopted to challenge the authority that the researcher has over the children in her or his position as an adult, is for the children to introduce what they believe to be ‘adult’ ideas normally considered as ‘taboo’ for children of this age. Of course, we cannot assume any of these motives as lying behind the children’s actions. However, we can conclude that what the children say, whatever the particular reason is for them saying it, does reflect what they consider to be important. As such, the influence of the adult researcher actually provides an important source of data in its own right.


The other major concern that I want to briefly address here is that of generalisibility. In essence, while the arguments above may be accepted in relation to interviewer-effects, it could still be argued that the data produced is still extremely limited as it is impossible to develop generalisations from. Again, I can only accept that the present research is limited if judged by its ability to generalise. However, it needs to be stressed that this is not the purpose of the present study. Rather than developing empirical generalisations, we are attempting to uncover the causes and effects of sectarianism among children and, as will be

p. 56 p. 57

Jump to Top

   Contents Page

seen in Chapter Six, reasons for the limited influence of the specific contact scheme organised between the youth clubs. This involves, necessarily, a focus on the broader social contexts within which children are located as argued in the previous chapters. Moreover, in order to understand why a child decides to behave in a certain way we need to understand the meaning they attribute to their actions. All of this, therefore, requires more in-depth, qualitative methods (Sayer 1992). Of course the problem with this approach is that the more in-depth it becomes, the more you loose the ability to generalise. In the end, as Hammersley (1992: 172) rightly contends, the choice of methods will always involve a ‘trade-off’ of this kind. This is equally true for more quantitative methods which in using larger, more representative sample sizes may well be able to produce sound generalisations but will do so at the cost of exploring the meaning that individuals attribute to their actions and the particular social interactions and processes that underlie this. In other words, they trade-off the need to generalise for the ability to address questions of why.

To offer an example of this we can take the article by Trew (1989) in which she reviews the present evidence in relation to the effects of contact schemes, derived in the main from various attitudinal tests carried out on children, and where she offers the generalisation that such schemes have had very limited long-term impact upon the children. While this appears to be a sound (and important) generalisation to make, Trew is unable to explain exactly why this is the case from the evidence before her. Again, for such a question to be answered, we need to develop an understanding of the complex range of influences that children experience within their social worlds and how these tend to encourage children to think and behave in sectarian ways. This, however, requires a very different methodological approach to the more quantitative one adopted by those studies reviewed by Trew (1989).


This chapter has proposed a more fruitful and appropriate methodological approach to the study of sectarianism and community relations work with children which is now to be modelled out in the following two chapters. It has been argued that the particular choice of largely unstructured group interviews with the children provides the most effective means of developing an understanding of the influences of sectarianism in their social worlds. In essence, the unstructured nature of the interviews allows the children to raise and discuss what they

p. 57 p. 58

Jump to Top

   Contents Page

feel to be important and enables the researcher to begin to develop an understanding of when and where the children draw upon sectarianism to make sense of their experiences. Such an approach does raise issues concerning interviewer-effects and the ability to generalise from the data. In response, it has been argued that while the present research does not claim to offer conclusions that are either representative or generalisable to all children it is capable of making a number of important contributions to our understanding of sectarianism and young children. In particular, it can:

  • show us what children at specific ages are at least capable of in relation to their ability to engage with and reproduce sectarian beliefs

  • draw attention to some of the broader social contexts that help to encourage some children to draw upon and make use of sectarianism in making sense of their social worlds

  • help us to understand some of the reasons why current approaches to community relations work (i.e. contact schemes) are or are not having the desired affects

  • and, as a result of the information gathered above, can contribute to the appropriate understanding necessary in order to begin to refine and develop existing community relations programmes

It is to these questions that the following two chapters now turn. Chapter five will focus on the nature and extent of sectarianism as found among the children and some of the reasons for its expression. Chapter six will then focus attention onto the youth clubs and their attempts to address this sectarianism among its children.