Comparative analysis of Boys, Girls and Achievement (Becky Francis) and Choice, Pathways and Transitions Post-16 (Stephen J Ball et al)
In 2000, two books were published that were based on research funded by the Economic and Social Research Council. These were Boys, Girls and Achievement: Addressing the classroom issues by Becky Francis and Choice, Pathways and Transitions Post-16: New youth, new economies in the global city by Ball, Maguire and Macrae. Both addressed issues of educational attainment and both took as their starting point for observation the behaviour and attitudes of adolescents in their mid-teens.
In other respects, however, these books, and the research that underpins them, are very different from one another not only in terms of the findings they produced but also in fundamental aspects of scope, methodology, and ideological background. These differences are the more noticeable and significant because so many aspects of the two studies do share similarities.
Boys, Girls and Achievement is an examination of the way gender roles are constructed by secondary school children and how this might be connected to educational achievement for both boys and girls. Choice, Pathways and Transitions is a study of the lives of school leavers, looking at the options open to them and the routes they take, whether in education, employment or neither.
Both studies took place in London. Francis investigated gender attitudes in three schools in different parts of London, ranging from the inner city to Greater London. All three of these schools were mainly working class in composition, although they had different gender balances and ethnic mixes (Francis 2000: 21). Ball et al focussed on one particular school in south London, along with the local Pupil Referral Unit (Ball et al 2000: 14). This school also had a mainly working class composition, and a significant number of ethnic minority pupils. Ball et al give an enormous amount of information about the area in which the study took place and the backgrounds (both ethnic and social) of the pupils themselves (Ball et al 2000: 10-16, Appendices 1-3 and passim), while Francis merely sketches in the general ethnic and social background, and gives no information about the local areas except to say whether they are surrounded by council estates, or a mixture of estates and owner-occupied housing (Francis 2000: 21-2). This is in line with the differing methodological approaches: a case-based approach such as that of Ball et al requires a much deeper knowledge of the subjects, while a more variable-based approach, such as Francis’, makes up for the lack of background information by gathering data from a wider sample (see below for more on methodology). Despite this justification, however, Francis’ sketchiness about the ethnic and class backgrounds of her sample is a weakness, as we shall see.
Concerning the choice of London as the area from which to select the sample, Ball et al give two reasons. The first is that poverty and inequality is at its worst in London, with economic polarisation increased by recent Government policy (Ball et al 2000: 5). This is relevant to a study looking at how political and economic changes have affected the options available to young people. The second reason is that London is a “global city”, like Tokyo and New York. This means that it is linked much more closely than other large cities to global economic organisation and developments (Ball et al 2000: 5), and as such it presents different opportunities and even different economies from the rest of the UK – for those who are able to access them (Ball et al 2000: 148-9). This means that by setting the study in London they are able to look at the widest variety of choices that are (theoretically) open to young people. Francis, by contrast, gives no reason for setting the study in London, leaving the reader to guess whether the choice was made for ideological reasons, or purely for convenience. This may be an example of what Ball et al note as a tendency in social theorising to assume that trends and patterns in London apply generally to the whole of the UK (Ball et al 2000: 148-9).
The historical background to the research is important to both studies, and it is the same history to which they both refer, although of course they each take a different focus. The historical development in question is the change to the labour market in the second half of the twentieth century, and particularly since 1979 and the accession of Margaret Thatcher as Prime Minister. As Ball et al note, the adolescents in their study are “Thatcher’s Children” (Ball et al 2000: 1). During this period manufacturing jobs decreased as industry declined, and the service sector provided a larger share of the job market. This meant that there were fewer manual jobs requiring few or no qualifications, which has seriously reduced the options available to school leavers, and particularly boys (Francis 2000: 7). This background is relevant to the work of Ball et al because it reduced the options open to the young people whose choices they studied. They focus in particular on the “McDonaldized” nature of many of the jobs available, and how this contrasts with secure, high-status professional and managerial jobs (Ball et al 2000: 10). Employment that falls between these two extremes has become rarer, and so many of the young people in Choice, Pathways and Transitions are faced with employment that is low-pay, low-status and low-security.
Francis focuses more on what this change in the labour market has done for the prospects of boys. Manufacturing jobs that required no qualifications were mainly filled by men, because of the physical nature of the work. On top of this, the growing service sector is seen as requiring more ‘feminine’ skills such as empathy and communication (Francis 2000: 7). This means that the traditional jobs for male school leavers are disappearing and the jobs that are replacing them are unattractive on gender grounds. Francis is concerned to look at how this may be affecting the educational attainment of boys and girls.
Unlike Ball et al, Francis also looks at the educational history of the period, with the introduction of the National Curriculum and standardised testing in the Education Reform Act 1988. This was the move that made it possible to directly compare boys’ attainment with girls’ throughout the country, by means of league tables showing performance in compulsory subjects (Francis 2000: 6). Ball et al, on the other hand, are more concerned with the ‘learning society’ the political initiative to encourage people to obtain and continue to develop marketable skills throughout their lives so that they are suited to the needs of the labour market (Ball et al 2000: 8). The unrealistic nature of this one-dimensional view of young ‘learners’ as fodder for the labour market is dealt with throughout the work.
Although the historical background and social changes that inform and indeed inspire the two books are similar, when it comes to the underpinning social theory, more differences start to emerge. In fact, both works are informed by a post-structuralist understanding of identity; people ‘construct’ who they are in terms of the particular society, culture and background they find themselves in. “We take it that the self is discursively and interactively constituted”, as Ball et al put it (Ball et al 2000: 20). Francis, however, goes further down this line by taking a social constructionist view of gender, something that Ball et al do not address, although they do look at the particular gender-based social limitations that face some women in the study. Francis believes that “gender difference is socially produced” (Francis 2000: 19), a belief that is very important to the purpose of her book, which is to break down the gender constructs that hinder learning. Francis is a feminist, and her work is strongly influenced by this philosophical position. She believes that the construction of gender difference leads to the subjugation of women and says, “As a feminist, I believe that this situation is wrong and we should work to change it.” (Francis 2000: 19)
This declaration of purpose is another difference between the two texts. Francis has an acknowledged aim: to deconstruct the gender constructions of secondary school pupils that are detrimental to the education of both boys and girls. To this end, the final part of Boys, Girls and Achievement is devoted to practical strategies that can be used with the whole class or with groups, to deconstruct preconceived notions of gender roles and expressions of gender such as behaviour. She believes that this should “spur educational achievement” (Francis 2000: 152). Ball et al’s work is much less prescriptive in nature. While they make it clear that they are uncomfortable with the Government’s notions of young people as “individual, rational calculators”, they do not say whether this is a view that preceded, or arose out of, the research, nor do they make any prescriptions about what young people, or the Government, should do. Instead, the book seeks to “describe and analyse the choices and constraints” of the young people (Ball et al 2000: 1). The overall purpose, therefore, is descriptive rather than prescriptive. It is for others to use their data and analysis as best they may.
Unlike Francis, Ball et al do not appear to have a hypothesis that they are trying to prove or disprove. Their work is inductive rather than deductive, resembling grounded theory. Francis does appear to have a hypothesis, although it is never explicitly stated as such. She mentions that she has previously carried out similar work in primary schools, looking at gender construction and power (Francis 2000: 19), and throughout the book she notes when the results correspond to the earlier research and when they do not. She also begins the book by discussing the notion that boys are ‘failing’ in school because girls are outperforming them. Her hypothesis, then, seems to be that, as in primary schools, secondary school children construct their gender, and this is the reason why boys appear to be progressing less rapidly than girls. As a feminist, though, she believes that the construction of gender disadvantages girls as well as boys, and should be challenged (Francis 2000: 19).
The different approaches and scopes of the two studies are reflected in the research methods used and in the sample selected.
The research of Ball et al is purely case-based. Each of the chapters of their book demonstrates a different potential outcome for young learners, but the data is presented in the form of narratives, following the lives of particular individuals – typically three or four per chapter. The young people’s names (pseudonyms) are given, and their whole history throughout the study is presented as a unit; it is not split up to sit under different topics. The study was also longitudinal (Ball et al 2000: vii). This type of research produces rich and complex knowledge of how different variables interact, but it is of course harder to generalise from than statistics would be. It is also not well suited to testing a hypothesis, since the young people are being treated as holistic individuals, rather than as points on either side of an argument. Case-based research requires a relatively low number of subjects because so much information is drawn from each subject, and because it all has to be analysed qualitatively for meaning, which is very time consuming (Della Porta 2008: 208). Ball et al’s study began with 110 students, but of those only 59 were interviewed over four years for the in-depth study, and only 24 are featured in the book Choice, Pathways and Transitions as representative examples (Ball et al 2000: 14).
Francis, by contrast, carried out classroom observation on twelve classes at three different schools (Francis 2000: 24). Assuming a class size of approximately 30 pupils (for Francis does not specify) gives a sample of 360. Each class was observed three times over “a relatively short time” (Francis 2000: 23), making this study transverse rather than longitudinal. Interviews were then conducted with a random sample drawn from all the classes, numbering 100 pupils – 50 boys and 50 girls (Francis 2000: 28). This larger sample, combined the shorter space of time and the use of only one interviewer (Ball et al used a number of interviewers) means that the interviews had to be conducted in less depth, but the results could be generalised more easily. For example, Francis is able to provide a table of the attributes of an ideal pupil with figures showing how many times each attribute was mentioned by girls, and how many by boys (Francis 2000: Appendix 3). This pooled information is isolated from the identities of the individual interviewees, and could be used as an indication of the general attitudes of adolescent boys and girls, in a way that Ball et al’s narratively presented interviews could not be.
Francis’ study is variable-based rather than case-based, although the data cannot always be reduced to statistics. Perhaps because of the difficulty of accurate recording when there is only one observer, the classroom observation part of the study is used mainly to confirm the presence of certain language (such as ‘bitch’), behaviours (such as personal grooming) and attitudes (such as homophobia), and to demonstrate tendencies and trends (for example, “boys were far more physically active in the classroom than were girls.” Francis 2000: 42). When collecting generalised information like the data in Boys, Girls and Achievement it is helpful to have a larger sample size so that it is possible to generalise from the data with greater confidence (Della Porta 2008:198). Francis’ study was designed to indicate something about boys’ and girl’s attitudes to gender in Britain as a whole, so a larger sample is useful to her.
Interview technique, and the interpretation of results, is another are where these two books differ, and again the differences in method are attributable to different theoretical or philosophical positions. Francis, in accordance with her feminist beliefs, takes an approach to interviewing which might be labelled romanticism or emotionalism (Alvesson 2011: 14). Rather than trying to reduce the influence of the interviewer as much as possible, she enters into a relationship with her interviewees, being “winsome, warm/jokey” with the boys, while “laughing conspiratorially and nodding” at the derogatory things the girls said about boys (Francis 2000: 29). This is in line with the feminist approach to interviewing which seeks to break down the power imbalance between the interviewer and the interviewee, and to establish a relationship of trust where the truth can be revealed collaboratively (Alvesson 2011: 15). It does raise some problems, however, as Francis herself recognises. By giving the boys special treatment to get them to “open up” she risks being guilty of accommodating males while assuming co-operation from girls, effectively submitting to and reinforcing gendered power roles (Francis 2000: 29). More worryingly for the integrity of her results, the intrusive interviewer may elicit responses that have been influenced by her own views. Certainly, there are times in the transcripts of the interviews where Francis appears to put words in the mouths of the participants, for example when she (rather than the interviewee) suggest that it is “all right for [boys] to be thick” and when she inserts the word “slag” into the conversation rather than allowing the interviewee to supply it (or an alternative) (Francis 2000: 73 and 82).
Ball et al seem to have initially taken a more positivist approach to interviews, keeping the format very prescriptive and the power balance very unequal. Later, however, they allowed the subjects of interviews to be led more by the interviewees, eliciting information which they had not anticipated and which could then be followed up with other subjects. This later approach can be characterised as interactive rationalism (Alvesson 2011: 12-13). Although more trust had been built up and the interviewees have more freedom, there is no suggestion of collusion or the intrusion of the interviewer into the data. Ball et al acknowledge the problems of interview context that they faced, with varied and sometimes unsuitable venues, and often the presence of family members, which inhibited the responses. Despite this, they still seem to take a positivist approach to the data gathered, treating it as valid, if curtailed, information, rather than as simply the product of the particular context in which the interview took place (Della Porta and Keating 2008: 23). This may be a little over-confident, especially considering the influence that the opinions of family members have over the young people, as evidenced by the study. These variable and often unsatisfactory interview contexts are a weakness in Ball et al’s work, compared to the uniformity available to Francis in her work in a school setting.
The other major weakness in Choice, Pathways and Transitions is how heavily it relies on interpretation. This is, of course, both the strength and weakness of case-based studies. The young people are allowed to represent themselves rather than statistics, which means that the scope of the study is much wider and the information obtained is richer, but in order to reach any conclusions about what the lives of particular young people have to say about the situation of young people in London or Britain as a whole, it is necessary to supply a large amount of interpretation. Other researchers might have taken different lessons from the same results, and this subjectivity is a weakness that Ball et al acknowledge (Ball et al 2000: 142).
Certainly the weaker of the two studies, however, is Boys, Girls and Achievement. The problem of leading interviewees has already been discussed, but perhaps more serious is the attitude towards the ethnic and class backgrounds of the children involved. Ball et al found that the single biggest factor that increased or curtailed a young person’s chances was class (Ball et al 2000: 177), while ethnicity, especially for women, could also be a major factor; the way a girl’s gender affected her choices was strongly linked to cultural background (for example the cases of Rena and Delisha, Ball et al 2000: 35-41). Francis pays no attention to class whatsoever, except to say that her sample is mostly working class (Francis 2000: 21). As for class, her approach is very inconsistent. She claims that, while a full analysis of ethnicity and class are outside the range of a study such as this, she will attempt to include an analysis of ethnicity where possible (Francis 2000: 23). To this end reference is made to racial stereotypes (found in the schools) of Afro-Caribbeans as sexually potent (Francis 2000: 38), speakers are identified by sex and ethnicity, and where ethnicity seems to be relevant to gendered behaviour it is remarked upon (e.g. Francis 2000: 44). However, no reference to ethnicity is found in the final discussion, and nothing which could fairly be called “analysis” of ethnicity and gender appears anywhere in the book. The references to the young people by ethnic background seem to be a red herring, and are therefore merely distracting and unhelpful. Particularly unhelpful is the term “Anglo” which is never defined and is by no means self-explanatory.
The two books examined here make an interesting comparison because of the similarity of the subjects they study and the almost identical social setting, combined with very different methodological approaches and ideological interests. There is nothing to criticise in the choice of either a case-based or variable-based research model, nor in the focus on gender specifically as opposed to all of the factors affecting young people’s choices. These are all equally valid and are useful in different ways. However, even making allowances for these differences, Choice, Pathways and Transitions succeeds better on its own terms than does Boys, Girls and Achievement.
M Alvesson (2011) Interpreting Interviews London: Sage
S J Ball, M Maguire and S Macrae (2000) Choice, Pathways and Transitions Post-16: New youth, new economies in the global city London: RoutledgeFalmer
D Della Porta (2008) “Comparative Analysis: case-oriented versus variable-oriented research” in D Della Porta and M Keating Approaches and Methodologies in the Social Sciences: A Pluralist Perspective Cambridge: Cambridge University Press 198-222
D Della Porta and M Keating (2008) “How many approaches in the social sciences?” in D Della Porta and M Keating Approaches and Methodologies in the Social Sciences: A Pluralist Perspective Cambridge: Cambridge University Press 19-39
B Francis (2000) Boys, Girls and Achievement: Addressing the classroom issues London: RoutledgeFalmer
M Keating (2008) “Culture and Social Science” in D Della Porta and M Keating Approaches and Methodologies in the Social Sciences: A Pluralist Perspective Cambridge: Cambridge University Press 99-117