In the corridors and classrooms of my current placement school, resound teacher discussions, some boasting of achievements of the ‘most able’ pupils, others lamenting failures of those deemed ‘least able’. Having embarked on my teacher-training journey, I came to the realisation that ability labelling according to attainment (often in summative assessments) is a systemic practice in my placement school, typical to the UK secondary school environment.
Through lesson observations and teaching my own classes, I also realised the immense power of labels such as ‘top, middle and bottom sets’ to determine to which textbooks, curriculum content, teachers and assessment papers individual pupils are exposed. Hart (2004) explains how ‘judgements about ability are the points of reference against which teachers formulate expectations, make decisions about appropriate learning opportunities, decide how to interact with pupils and evaluate their progress.’ (Hart 2004:6)
The grave implications of pupils being assigned to a lower set-such as the unavoidable cap on grades due to the tiered assessment (at KS3-4) of a differentiated curriculum – puzzled and caused me to question the effectiveness or justifiability of setting by ability, a practice that I formerly considered an established organizational norm supported by solid theoretical rationale.
My interest in ability labelling and more specifically, within-school setting by subject according to attainment, stemmed from the recent wave of debate in the media regarding Theresa May’s plans to ‘end an “arbitrary” ban on the creation of new grammar schools that has been in place since 1998’ (Stewart, H. and Walker 2016), presaging the re-emergence of selection by ability in UK schooling. This ignited personal pondering and I began to question whether the system of ‘setting’ is a mere reflection of a pedagogic consensus on the existence of fixed, innate and measurable intelligence- the infamous ‘g’-or whether it’s a practical solution to diversity in the classroom. I also wondered whether there is empirical evidence that pupil achievement and motivation is enhanced in a homogenous learning environment, as both Labour and Conservative governments have supported in official papers, especially the Labour government during its three terms in office (1997-2010) (DfE 1997, 2001, 2004).
These considerations are of course are of primary importance to me as a trainee teacher, as theoretical and empirical knowledge on setting by ability will equip me with understanding of pupil reactions to the system and the possible effects on motivation, attitudes and behaviour-whether it fosters a competitive spirit among the most able or demotivation in the lower ranks. It may also account for pupil performance, whether this exceeds or falls short of expectations. Conversations with leading staff at my placement school reveal that it is proud of its set organisational structure especially for maths, English and languages, and regards setting as an effective means of ensuring that the brightest pupils are challenged and that the weakest are taught content appropriate to their levels as well as support.
Whatever the theoretical rationale, setting by ability is employed by the school, although it is claimed on its website that ‘years 7-9 is in mixed-ability groups and some students are given additional support in smaller groups where appropriate’ (Placement School). In fact, setting by ability is more prominent, where maths is taught in sets from years 8-11, as is English and modern languages. Science is taught in mixed-ability classes throughout years 7-11. At KS3 the school operates its own system of banding after the government removed national curriculum levels and allowed schools to form their own system of measuring progress (Capita Sims 2015). Bands vary from levels 3-8, and are reported to be ‘based on Bloom’s Taxonomy’ (Placement School 2016). For subjects such as mathematics, each topic of the curriculum has its own criteria for each band and for English the criteria vary for reading and writing. The performance of students is assessed against these bands, and eventually, for some subjects, pupils are placed into the bottom, or lower set.
Upon discussion with colleagues in the maths, MFL and English departments, it was brought to my attention that teachers often spoke of the effects of setting on pupil behaviour and motivation. In the case of modern languages, teachers alluded to the role of setting in demotivating those who previously displayed great enthusiasm for languages in year 7, before they had been placed in a low set. The language teacher expressed concern for witnessing such detrimental consequences on pupil enthusiasm for her subject, as a result of the imposition of setting and their accompanying feelings of failure at being labelled low achievers. Similarly in maths, teachers reported a general demotivation and an ‘I can’t do maths attitude’ among pupils, across all ability sets. For English, some spoke of the pressure on higher ability pupils to perform, especially from parents, and the preference among parents for setting especially for mathematics.
The questions arising from these conversations were primarily based on whether setting is justified as an organisational practice, and whether there is a theoretical foundation for parental preference, and its imposition by the Department of Education. The head of Ofsted Sir Michael Wilshaw’s position on setting is clear-‘Unless you have well-qualified, experienced people who know how to deal with mixed ability then it doesn’t work. And if that’s not happening, it is much better to move towards a setting system.’ (Paton 2013). I set out to discover whether there was an abundance of evidence to support the Ofsted and governmental position, by analysing the effects of setting on attainment as well as motivation.
In the literature, ‘setting’ refers to the allocation of pupils to specific classes across a year group, based on their attainment in each curriculum subject. It is distinguished from ‘streaming’ since the latter assumes performance in all academic subjects is determined by a ‘general ability’, so pupils are taught in the same class for all lessons (Ireson et al. 2002: 300). Boaler (2000:633) explains how setting is characterized by its ‘narrower scope’. Mixed-ability teaching implies classes are ‘formed on a random basis’ and pupil allocation to a class is not guided by attainment or measured ability (Ireson, op.cit.)
Ireson et al. (2002:299) describe how ability grouping (more specifically streaming) was the prevailing means of organization in secondary schools in the 1950s. The trend saw a reversal in the 1970s, driven by egalitarian concerns on education, the wave of comprehensivisation and research linking setting to working-class underperformance (Boaler 2000:632). Until the 1990s mixed-ability teaching invaded the educational landscape, however, interestingly, the 1990s were subject to resurgence in preoccupation with ‘academic success’ which saw a revival of setting (ibid.). Ofsted data reveal that for the first term of the labour government, – 1997-2001, inspections reveal a steady increase in total setting observed within a school (from 37% to 40%) (DfE 2012).
Today, setting is widespread in UK secondary schools, however OECD data reveals that high levels of ability grouping in mathematics is not unique to the British system. PISA data (2013) reveals that ‘across OECD countries, 67% (three out of four) students attend schools whose principal reported that students in mathematics classes study similar content, but at different levels of difficulty’ (OECD 2013: 81). Alongside the United Kingdom, Albania, Ireland, New Zealand, Australia, Israel, Kazakhstan, Singapore, the Russian Federation and Malaysia are examples whereby 95% students attend schools where students are grouped by ability across classes. (ibid.) Thus, the UK example is typical among certain OECD members.
Ability setting and attainment
At the advent of a New Labour victory in 1997 under Tony Blair, the Labour White Paper ‘Excellence in Schools’ highlighted the failure of ‘mixed-ability grouping’ to deal with ‘diversity within one campus’ (DfE 1997: 38). Setting by ability was presented as the solution to ‘stretch the brightest’, and to respond to the ‘needs of pupils who are struggling’
(ibid.) This coincides with attitudes deduced from informal conversations with some leading staff at my placement school, which see setting-especially for maths and English- as a means of ensuring appropriate challenge for the brightest and directed ‘one to one’ support for the rest.
In fact, very few research findings up until the late 1990s support the Labour government standpoint. In an early study of a selective grammar school, Lacey (1970) exposed imbalance in attention and resources in favour of higher ability groups to the detriment of the lower, and a follow-up study in the same school in 1974-after it adopted mixed-ability teaching-revealed better attainment among lower ability pupils in this type of structure (Lacey 1970, Dunne et. al. 2001:28). Evidently the generalisability of the results may be called into question considering that the selective nature of the grammar school implies that results apply to a very specific cohort of brighter pupils.
On comparing set and mixed- ability samples, Newbold (1977) failed to find similar benefits for upper sets, but similarly revealed that weaker pupils tend to thrive in a mixed-ability environment (Newbold 1977, Dunne et. al. 2001: 28). On a slightly different angle, in his review of twenty-nine studies, Slavin (1990:494) found that ‘between-class ability grouping had little or no effect on the achievement of secondary students, at least as measured by standardized tests’. This was especially applicable to KS3 (years 7-9), across all subjects.
Thus, research from the 1970s to the 1990s seems to contradict governmental aspirations of setting as a means of ‘stretching’ the most able and ‘supporting’ the weakest. It was found wanting in major effect on the attainment of the former, and the latter benefited more from heterogeneous settings. The reason for this trend is not precisely determined in the literature, however a possible explanation may be linked to teaching quality, which has been reported as better for higher sets and deficient for the lower, thus explaining why those struggling perform better in heterogeneous environments, where poor teaching standards may be eliminated. Despite these results, setting practices continued to increase during the first term of the Labour government (1997-2001) (DfE 2012).
While Lacey (1970), Newbold (1977) and Slavin (1990) failed to discover significant advantages of setting for either end of the ability spectrum, a more contemporaneous study by Boaler in 1997 when the government White Paper appeared, served to reinforce the unjustifiability of DfE insistence on setting as the ‘norm’ in secondary schools, at least in terms of attainment outcomes. Boaler’s (1997:591) findings similarly link success at summative assessment to mixed-ability schooling, as pupils were found to achieved better A-G results for mathematics at GCSE level than at the set school. The strength of the study lies in that the school samples were of similar working class composition, and had similar levels on entry, so it is unlikely that performance stems from social class or selection on entry. In light of the findings, governmental insistence that ‘unless a school can demonstrate better than expected results through a different approach…setting should be the norm in secondary schools”, (DfE 1997: 38) seems somewhat empirically unfounded, as most research up to that time did in fact reveal that a ‘different approach’, namely heterogonous class teaching, did indeed result in better attainment among weaker pupils or, as in the case of Boaler’s (1997) study, among all abilities.
Pupil and teacher views
Most interestingly, it is the exact merits of whole-class teaching highlighted by the White Paper – ‘less pressure on teachers, appropriate pace and challenge’ (op.cit., p.38 ), especially for maths, English and Science, that are contested by Boaler’s (1997:585) study. Interviews with pupils reveal strong opinions against the stratified classroom experience in maths, with both sides of the ability spectrum associating the ‘fixed pace’ of whole-class teaching to ‘a loss of understanding’. Girls in upper sets reported severe ‘maths anxiety’ and attributed lower performance in mathematics to the impracticalities of trying to ‘keep up’ with the rest of the class at all costs, as demanded by the teacher. Those in lower groups were equally discontent with the pace of their set environment, either complaining of boredom due to repetitive or unchallenging material or being held back by their peers (Boaler 1997:582). In this way, the central tenets of governmental support for setting- appropriate pace and challenge-were seriously brought into question by the pupil voice. However, only 24 interviews with pupils were conducted, thus generalisability of Boaler’s conversations is limited.
While pupils in the study condemn the fixed pace of homogenous classes, conversations with staff at my placement school reveal proximity of teacher views to those of the government. Setting is considered by educators, including myself (based on my teaching experience thus far) as a means to streamline curriculum content and to deliver homogenous material to the whole class as an entity or to at least offer more targeted attention to those struggling. Some colleagues (in the maths department) have linked setting to smaller class sizes and better behaviour management, especially among lower sets that are often associated with poor behaviour standards.
Pressure on the highest achievers emerged as a theme in discussions with teachers and colleagues from the MFL, Maths and English departments at my placement school, however, from the teacher perspective pressure stems from the pupils themselves or their parents, rather than teaching methods or classroom environment. This is an interesting reversal of attribution, where Boaler’s top set pupils-especially girls- hold teachers accountable for their anxiety-related failures stemming from the pressures of homogenous teaching but teachers I have spoken to attribute these failures to self or parent- exerted pressure among the highest achievers.
Thus far, we have explored how earlier studies find little effect of setting on attainment (Newbold 1977, Slavin 1990, Boaler 1997), or that positive effects are reserved for higher ability sets to the detriment of the lower (Lacey 1970). Despite the theoretical context, Department for Education publications from 1997 (DfE 1997) communicate clear preference for setting by ability as the ‘norm’ while contemporaneous studies such as that of Boaler (1997) went further to suggest that the detrimental effects of setting extend to the most able. Interestingly, pupil views on setting align with the literature and teacher views align with the governmental position.
2000s- Government reversal and further studies
By the 2000s, the government’s endorsement of setting is severely mitigated, for despite advocating setting in the 2001 Green paper ‘Schools Building on Success’-‘we want to see further increases in the extent of setting within subjects…’ (DfE 2001: 51), the White paper of the same year as well as the 2004 publication the ‘Five Year Strategy for Children and Learners’ (DfE 2004) fail to mention setting. Reasons for the change are not clear from the literature, however it could mark an attempt at a more ‘flexible approach’ by the government, as a response to the findings of the academic community. This is also reflected in the slight decrease in setting observed by Ofsted for 2001-2005 (DfE 2012) (40% in 2001-3, 36% in 2003/4, 36% 2004/5).
In fact, a study by Ireson et. al. (2002) during that period (using a stratified sample of 45 mixed comprehensive schools representing a range of grouping practices) revealed, yet again, the benefits of setting for the brightest and of mixed-ability teaching for those struggling. The findings merely confirmed the results of years of previous study and was reliable in that the data compared pupils of similar ability in different organizational settings. It was valuable in that it demonstrated how higher placed pupils attained higher scores in Key Stage 3 results than those of similar prior attainment placed in lower sets (Ireson et al., 2002). The relationship between individual sets and results was thus brought to the forefront of the setting debate. A natural inference would be that higher sets yield better results due to higher quality of teaching or the existence of a more competitive atmosphere in classes where all pupils are bright and aspirational. Whatever the causes, a study by William & Bartholomew (2004:291) served to mirror Ireson et al’s results. They found that being placed in a top set raised GCSE levels in that subject by half a grade and being in the bottom set lowered them by half a grade. To determine possible explanations for higher performance in advanced sets or lower performance in weaker sets or other possible combinations, we must analyse the effects of setting on pupil motivation, academic attitude and self-efficacy.
Ability setting and ‘academic self-concept’
Using the self-concept scale of Marsh (1990), Ireson & Hallam (2001) undertook a study of 3000 year 9 pupils (aged 13-14 years) in 45 mixed secondary comprehensive schools in England. The schools represented three levels of ability grouping in the lower school (years 7 to 9). ‘Self-esteem and general school self-concept were found to be higher in the partially set schools (no more than 2 subjects in year 7 and 4 subjects in year 9) than in fully set or mixed ability schools’ (Ireson & Hallam 2001: 323).
Interviews with Head teachers in the study reveal that moderate levels of setting are considered to create an environment of healthy competition and respect for achievement and progress
that may explain higher academic self-concept compared to mixed ability schools. The study is distinct in that it is one of the few findings of any important positive impact of setting.
However, it must be noted that this impact varies between subjects in the data, whereby setting in English raised the self-concept of the lower groups and lowered that of the higher-ability classes, but no similar effects were found in maths or science. The difference has been traced to more quantifiable assessment in maths and science that renders pupils, according to Ireson & Hallam, more ‘aware of their relative performance’ (Ireson &Hallam 2001:52).
Ireson and Hallam’s (2001) analysis of pupil self-concept in diverse organisational structures is compelling in that it implies the effects of ability setting vary with subject and assessment methods, where more open-ended, qualitative assessment in English seems to cause uncertainty among top set pupils when evaluating their own progress. Despite the fact that the research is based on pupil impressions of self-efficacy instead of actual attainment, it is comparable to Boaler’s study (1997) in the similar discernment of detrimental impact of setting for the brightest, running contrary to previous studies that described disadvantages as exclusive to lower sets. Where they differ is that Boaler found the effect to be applicable to mathematics and not exclusively English.
In Boaler’s (1997) study, girls in top set mathematics reported lower self-efficacy due to the pressures of being taught at a fixed pace, in a procedural manner and using teacher-centred methods (Boaler 1997:585) Interestingly, school-based conversations in the maths department at my placement school coincide with the findings of Boaler (1997) as colleagues have reported that the ‘I can’t do maths’ attitude is also prominent among the top sets. In this way, it seems that it is not only lower groups that suffer motivationally from setting practices. A study by William & Bartholomew (2004:290), reiterates pupil views that top sets suffer from the fixed, fast pace and lack of differentiation of homogenous teaching. Upper groups are found to be only mildly advantaged by setting. This calls findings from previous studies that top sets benefit from better quality teaching into question. The government standpoint as laid out in the 2005 White Paper, Higher Standards, Better Schools for all that ‘grouping students can help to build motivation, social skills and independence; and most importantly can raise standards because pupils are better engaged in their own learning’ (DfE 2005: 58) is also contested by these findings.
Lower ability pupils
Boaler’s (1997:587) study emphasises that lower ability groups also suffer from ‘lack of understanding, boredom, anxiety and disaffection due to restricted opportunities’. Interview data shows that this is linked to the differentiated curriculum accompanying setting and the fixed pace of whole-class teaching, with its lack of differentiation, which does not cater for pupils working at diverse speeds (ibid.). Most importantly, the motivational damage is reported to stem from feelings of the futility of working hard at the knowledge that there ‘is no point’- as grade boundaries are capped at a certain level (usually D or E) for each set (ibid.).
Another study by Boaler et al. (2000) reveals that ‘27% of students taught in the lower sets reported that work was too easy, compared with 7% of students in the upper sets and 14% of students in the schools using mixed-ability teaching in year 9’ (Boaler et al. 2000:637) inferring less aspirational teaching in lower sets. Interview data from the study by William and Bartholomew (2004:290) also report poor teaching standards for lower sets, less expectations and undemanding work. My own experience of teaching a lower ability year 9 Spanish class has also given me insight into the waning self-efficacy of the class, who overtly introduced themselves to me as the ‘unintelligent’ group, thus identifying with the stereotypical expectations of demotivated attitude inferred by their set.
The key themes that arise in the academic literature are that early studies find weak attainment advantage associated with setting. Studies by Lacey (1970), Newbold (1977), Slavin (1990), Boaler (1997) and Ireson et.al. (2002) show that weaker pupils thrive in mixed-ability settings while the research of Lacey (1970) and William & Bartholomew (2004) reveal the benefits of setting for the brightest, associating placement in a higher set with higher grades. Teaching quality may be a possible explanation for these results, as pupils tend to report poor teaching standards among lower ability sets, where higher sets are more likely to have experienced and highly-qualified teachers and lower sets are less likely to be taught by a specialist in the subject (William & Bartholomew (2004:291) . However, in terms of teaching standards, one must be cautious of relying solely on the pupil evaluations of pedagogic practices.
In terms of motivation, Ireson and Hallam’s (2001) study found setting to only affect English with no visible applicability to maths and science. Contrary to most studies where higher achievers attain highly in set environments, the work of Boaler (1997) and Ireson and Hallam (2001) sheds light onto the detrimental influences of setting on the attainment and academic self-concept of high ability pupils. The findings are reflected in my current placement school, as conversations with staff allude to general feelings of demotivation or pressure even among the higher ability sets.
Thus, setting is not consistently found to foster positive outcomes for higher ability learners, although there seems to be a general consensus that weaker pupils strive in a heterogeneous, mixed-ability environment where differentiation in the classroom allows everyone to work at their own pace and thus be free from the pressures associated with whole-class teaching.
Governmental views since 1997 under Labour started out with a strong advocacy of setting as the ‘norm’ in secondary schools, although this support waned in the early 2000s, possibly as a result of the academic evidence that seemed to contradict governmental aspirations of setting as a means to challenge the brightest and respond to the needs of the weakest. Interestingly, parental and teacher views tend to align with those of the government, and this is confirmed by conversations with colleagues at my placement school.
An evaluation of the effects of setting on attainment and motivation has provided me with great insight into the possible outcomes of an organisational practice imposed on a national level by the government. The prevalence of student opinion from the data seems to suggest preference for mixed-ability teaching, as this eliminates pressures linked to learning at a fixed pace and lack of differentiation. Such knowledge will possibly influence my own teaching practice in future, whereby I can find innovative means of ensuring pupils are permitted to work at their own pace, when they are in sets. This would be applicable for both higher and lower ability groups, as data has reveals that both sides of the ability spectrum suffer as a result of homogenous teaching, which pupils believe, treats them as if they were of identical ability. In light of the continued support of the current government alongside that of the head of Ofsted, Sir Michael Wilshaw, for the extension of setting as the predominant practice in secondary schools, new teachers such as myself must seek new practices in order to ensure that pupils benefit academically and motivationally from a set environment. Such practices may include more in-class differentiation for sets, or in-class grouping to enable pupils to work at a level and pace with which they are content.