A national survey of ability grouping practices in secondary school physical education in England

ABSTRACT

This research sought to generate large-scale yet sophisticated data relating to ability grouping practices in physical education (PE) in secondary schools in England, with the intent of extending knowledge of the various ability grouping practices being adopted within and across schools. The prevalence of particular ability grouping practices, processes associated with their application, and factors influencing their use and non-use were explored. Data were collected via a web-based survey of all (3197 at the time of study) mainstream state-funded secondary schools in England. A total of 903 responses were received, giving a response rate of 28.2%. The findings reveal that overall, mixed-ability grouping is the most common ability grouping practice in PE, although the extent and nature of this practice (and other ability grouping practices) varied by year group, Key Stage, gender of students, and/or curriculum activities. Setting was the predominant approach in PE in Year 8 (aged 12–13) and Year 9 (aged 13–14). The use of variants on these practices (including mixed-ability grouping with a separate top and/or bottom set) and descriptions of how and why different grouping practices are enacted in specific contexts, illustrate the significance of discourses of ability, gender, and pragmatism in grouping practices in PE.

KEYWORDS:

• Ability grouping
• setting
• streaming
• mixed-ability grouping
• physical education

Introduction

Ability grouping practices have a long and contentious history in educational policy and practice in England, with trends reflecting the influence of various factors, including political priorities, societal values, and/or educational ideologies of the time (Bradbury and Roberts- Holmes 2017; Ireson and Hallam 2001). The late 1960s and early 1970s, for example, saw a shift in educational priorities in England, from a focus on the attainment of the most able to a focus on child-centred teaching and the provision of equal opportunities for all students. The merits of streaming were thus increasingly called into question (Barker Lunn 1970; Jackson 1964; Lacey 1970; Willig, 1963). Seminal reports commissioned by the government showed that in contrast to streaming, mixed-ability grouping ensured that all students had equal access to teachers, curriculum materials, and resources (Department of Education and Science 1967; Ministry of Education 1963). The reports concluded that schools, and primary schools particularly, should abolish all forms of ability grouping (as well as between-school segregation) in favour of mixed-ability grouping (Department of Education and Science 1967; Ministry of Education 1963). By the mid-1990s, research showed that mixed-ability grouping was widespread in primary schools and was also increasingly used in the early years of many secondary schools in England (Lee and Croll 1995; Sukhnandan and Lee 1998). By contrast, in the late 1990s, the incoming Labour Government made trenchant criticisms of mixed-ability teaching, arguing that in most cases it had ‘failed both to stretch the brightest and to respond to the needs of those who had fallen behind’ (Department for Education and Employment 1997, 38). Primary and secondary schools were therefore exhorted to employ setting to enable teachers to tailor their instruction to a narrower range of attainment, and in doing so ensure that students were ‘better engaged in their own learning’ (Department for Education and Skills 2005, 58). In the run-up to the 2010 general election, the Conservative Party (2007, 2010) also advocated for greater use of setting in schools, and while there has been no explicit mention of setting (or any other form of ability grouping) in policy documents and ministerial statements in the period since, research continues to show that setting is an established practice in many primary and secondary schools in England, particularly in the core subject areas of mathematics, English, and science (Baines, Blatchford, and Kutnick 2003; Office for Standards in Education 2002, 2013; Taylor et al. 2022; Towers et al. 2020).

Although there is substantial literature on ability grouping practices in schools in England, the literature is largely restricted to mathematics, English, and science, particularly in secondary schools. To date, little research has directly explored the nature and extent of various forms of ability grouping in other subjects of the curriculum, including art, music, drama, and physical education (PE). Instead, these subjects have tended to be grouped together as ‘practically based’ or ‘other’ in research, obscuring potential differences in the ability grouping practices used between these subjects. This point is acknowledged by Taylor et al. (2022), who in noting that their study did not distinguish between ‘other subjects’, suggested that further research was needed to ‘provide a more detailed picture of grouping in “specific” subjects other than English and mathematics’ (p. 214, our emphasis). The research reported in this paper sought to do this for PE in secondary schools in England. We contend that PE provides a particularly important context for exploring various forms of ability grouping and the rationales for their use. In contrast to the relative privacy of classroom-based subjects, PE is a teaching and learning context in which students’ bodies and competencies are very publicly on display. The organisation of PE in secondary schools in England is also often gender-differentiated, with boys and girls taught separately by a teacher of the same sex (Stride et al. 2022; Wilkinson and Penney, Forthcoming). Furthermore, it is a subject setting within which particular notions of ability, centring on proficiency in motor skill and/or sport performance (itself narrowly defined), are acknowledged as normalised, often privileged, and inherently problematic (Evans 2004; Penney and Lisahunter 2006). Thus, we identify PE as a context that calls for research that avoids characterising ability grouping practices as ‘a simplistic dichotomy between mixed-attainment and setting’ (Taylor et al. 2022, 203), emphasises nuance, complexity, and variations in the ability grouping practices defined below (see Table 1), and that critically engages with normalised practices.

Table 1. Key terms and definitions.

Streaming	Streaming is a practice of assigning students to classes based on an overall assessment of their general ability, with students remaining in these classes for all or most subjects (Sukhnandan and Lee 1998; Wilkinson and Penney 2022b).
Banding	Banding is a less restrictive and differentiated form of streaming, in which students are allocated to broader ability bands, rather than to single classes (Ireson and Hallam 2001). Typically, schools have two or three bands, and each includes more than one class group, which may then be regrouped into sets for different subjects (Ireson, Clark, and Hallam 2002).
Setting	Setting is more flexible and nuanced than streaming and banding because students are divided into classes based on their prior attainment in individual subjects only (Francis, Taylor, and Tereshchenko 2020; Wilkinson and Penney 2022b). Hence, setting recognises that students may be low attaining in one subject area, but high attaining in another (Francis, Taylor, and Tereshchenko 2020; Hallam and Ireson 2006).
Mixed-ability grouping	Mixed-ability grouping differs from steaming, banding and setting because classes are formed (either randomly or purposefully) to ensure that each includes students with a broad range of prior attainment (Francis, Taylor, and Tereshchenko 2020; Wilkinson and Penney 2022b).
Within-class grouping	In conjunction with any of above ability grouping practices, teachers may also use within-class grouping to divide students within the same class into flexible, smaller groups for specific activities and/or purposes (Francis, Taylor, and Tereshchenko 2020).

Notes: Streaming, banding, setting and mixed-ability grouping are all forms of between-class grouping, where students are organised into different whole-class groups (Francis, Taylor, and Tereshchenko 2020). Schools may also use variations of these practices, including mixed-ability grouping with a separate top and/or bottom set (Taylor et al. 2022).

Our research was designed to provide original insight into contemporary ability grouping practices used in PE in secondary schools in England through an inquiry with national reach that challenged dichotomous representations of grouping and probed the processes and criteria featuring in decisions about grouping in PE. Indeed, our inquiry recognised that although streaming, banding, setting, mixed-ability grouping, and within-class grouping are distinct practices, some of these practices may be blended and/or enacted in different ways in different schools (Francis, Taylor, and Tereshchenko 2020). It also sought to extend understanding of the rationale for specific grouping practices being employed in PE with particular year groups and/or in association with teaching particular activities, and how particular practices were applied in any given setting. In pursuing matters of ‘why and how’ associated with grouping practices in PE, the inquiry was directed towards critical interrogation of the underlying understandings and/or assumptions about ability being expressed and enacted in grouping practices in secondary PE. Hence, the scale and sophistication of this inquiry into grouping practices distinguished it from previous studies in the PE field, while simultaneously providing an important extension to broader educational research addressing contemporary grouping practices in secondary schooling.

While focused on grouping practices in secondary schooling in England, the research addresses matters of international relevance, particularly in the context of limited research engaging with PISA data showing 38% of students in Organisation for Economic Co-operation and Development (OECD) countries attending schools where between-class grouping (setting, streaming) was used for some subjects, and subject-specific grouping in OECD schools increasing by 4% between 2006 and 2012 (operation and Development 2016). Contemporary research evidence relating to ability grouping practices in all subjects in secondary schools in England provided a more expansive backdrop to this study.

Background: ability grouping practices in secondary schools in England

Research identifies setting as frequently adopted in secondary schools in England, particularly for mathematics and English (Ireson, Clark, and Hallam 2002; Office for Standards in Education 2002, 2013; Taylor et al. 2022). In a national survey of ability grouping practices in 375 secondary schools in England, for example, Taylor et al. (2022) found that ‘the majority’ of schools were using setting in mathematics, with the prevalence increasing as students moved through school, from 60% in Year 7 (aged 12–13) to 88% in Year 11 (aged 15–16). 33% of schools were using setting in Year 7 for English, rising to 52% by Year 11 (Taylor et al. 2022). Taylor et al. (2022) also reported that around 10% of schools were using steaming, at least at the start of the lower secondary school.

As indicated previously, mixed-ability grouping has frequently been reported as the most common practice in other subjects in the secondary school curriculum, including humanities, art, music, drama, and PE (Hallam, Rogers, and Ireson 2008; Ireson, Clark, and Hallam 2002; Office for Standards in Education 2002, 2013; Taylor et al. 2022; Wilkinson, Penney, and Allin, 2016). Hallam, Rogers, and Ireson's (2008) survey of 48 secondary schools in England, for example, found that 92% of art, 86% of music, 80% of drama, and 58% of PE classes were taught in mixed-ability groups, with a range of practices used in the remaining classes, including setting, streaming, and banding. Wilkinson, Penney, and Allin's (2016) survey of 155 secondary schools in the North-East of England by comparison revealed higher levels of setting in PE, with 62% of secondary schools fully or partially (in combination with mixed-ability grouping) using setting between Year 7 and Year 11 (aged 11–16). Setting was particularly prevalent in Year 8 (aged 12–13) and Year 9 (aged 13–14) and varied according to different topics of the curriculum, with higher levels of setting for team games than individual games (Wilkinson, Penney, and Allin 2016). Wilkinson Penney, and Allin's (2016) survey also highlighted gender-differentiated patterns of ability grouping practices in PE, with one co-educational secondary school using setting in girls PE only (with mixed-ability grouping in boys’ PE) and ten co-educational secondary schools using setting in boys’ PE only (with mixed-ability grouping in girls’ PE).

The need to move beyond a simplistic binary distinction between setting and mixed-ability grouping, and instead recognise and further examine the multifaceted and overlapping nature of ability grouping practices in schools, is increasingly evident. Taylor et al. (2022), for example, revealed schools using combined approaches in mathematics and English, such as mixed-ability grouping with a separate bottom set (sometimes known as a nurture group) or top set (sometimes known as a stretch group), or a broad mixed-middle with separate top and bottom sets. Insight into such approaches being used in other subjects, and specifically in PE, is lacking. Further, very little is known about how teachers in other subjects perceive different forms of ability grouping and the various factors, including curriculum requirements and assessment approaches, that may influence these perceptions. Previous research indicates that there may be pedagogical reasons why some subjects are more frequently taught in sets and others in mixed-ability groups, although this remains unclear (Francis, Taylor, and Tereshchenko 2020; Taylor et al. 2022). Hallam and Ireson (2003), for example, suggest that differences may relate to the extent to which learning in specific subject areas is ‘perceived as linear and building directly on prior knowledge, [as well as] the extent to which differentiation can occur through learning outcomes rather than the setting of differentiated tasks’ (p. 354). Linear conceptualisations of learning are associated in turn with perceived suitability of classes that contain students of similar ability and/or attainment who can work through learning materials at a similar pace (Hallam and Ireson 2003; Sukhnandan and Lee 1998; Wilkinson, Penney, and Allin 2016).

While Wilkinson, Penney, and Allin's (2016) research has provided valuable insight into ability grouping practices in PE, mixed-ability grouping received only cursory attention (often appearing tangentially in discussion of within-class grouping) and the research did not collect data on streaming, banding, or variants of mixed-ability grouping. Moreover, Wilkinson, Penney, and Allin's (2016) research was limited to a regional sample and was conducted at a time when ability grouping was high on the political agenda (Conservative Party 2010; Office for Standards in Education 2013), with many teachers feeling under pressure to use setting to raise standards of attainment in PE. As indicated, this study aimed to achieve national reach in investigating the various ability grouping practices being used in PE across secondary schools in England, the processes and criteria associated with their use, and the reasons cited for particular practices being used in specific school, year level and teaching settings in PE. As reflected in the research questions that follow, we anticipated variations in practices within and between schools and sought to generate data that would reveal the understandings and assumptions about ability in PE that are expressed in (and legitimated by) current grouping practices.

Research questions and methodology

The study addressed the following research questions:

• How prevalent are various ability grouping practices in PE for Years 7-11 (aged 11-16) in mainstream state-funded secondary schools in England? (Including but not restricted to setting, streaming, mixed-ability grouping, and/or combinations of these).

• What processes and criteria are used to group students in PE in mainstream state-funded secondary schools in England?

• Why are these ability grouping practices used in PE in mainstream state-funded secondary schools in England? And, what discourses and assumptions relating to ability are explicit and implicit in teachers’ descriptions of grouping practices in secondary PE in England?

Design and administration of the survey

The research questions were addressed through an online survey administered using the platform Jisc Online Surveys (https://www.onlinesurveys.ac.uk/). The survey was adapted from previous studies of ability grouping in secondary schools in England (Taylor et al. 2022; Wilkinson, Penney, and Allin 2016). Specifically, additional free-text questions were added to explore teachers’ justifications for using particular ability grouping practices, the processes and criteria used in enacting different practices, and teachers’ attitudes towards different ability grouping practices in PE. Modifications were also made to some questions to distinguish between ability grouping practices used in boys’ and girls’ PE, as well as core (non-examined) PE and General Certificate of Secondary Education (GCSE) and/or Business and Technology Education Council (BTEC) PE in Key Stage 4 (Years 10–11). While much previous research has only included two or three options for type of grouping (typically setting or mixed-ability grouping and/or within-class grouping), following the lead of Taylor et al. (2022) the survey included a broader range of options to provide a more nuanced and comprehensive picture of ability grouping practices in PE. Space was also provided for comments on any grouping practices that were not included in the range of options listed. The range of options and their descriptions are summarised in Table 2. The survey also included questions about the extent and form of within-class grouping practices in PE, although it is beyond the scope of this paper to report these data.

Table 2. Ability grouping options presented in the survey.

Grouping option	Description
Setting	Students are grouped by ability/attainment in PE (not by ability/attainment in other subjects).
Fully mixed-ability grouping	All PE teaching groups include students from the full ability/attainment range for the year group.
Mixed-ability grouping and top set	The highest ability/attaining students are taught in a separate group for PE. All other groups include students from the remaining full ability/attainment range.
Mixed-ability grouping and bottom set	The lowest ability/attaining students are taught in a separate group for PE. All other groups include students from the remaining full ability/attainment range.
Mixed-ability grouping and top and bottom sets	The highest and lowest ability/attaining students are taught in separate groups for PE. All other groups include students from the remaining full ability/attainment range.
Streaming/Banding	Students are grouped for most or all subjects, including PE, by a general measure of ability.
Other	Other.

The survey was piloted with a convenience sample of six former Subject Leaders of PE prior to the commencement of the research and ethical approval to conduct the study was granted by the Ethics Committee of Northumbria University. Minor changes were made to the survey following the pilot study to improve the clarity and comprehensibility of some questions.

The survey was distributed by email to all (3197 at the time of study) mainstream state-funded secondary schools in England. Schools were first identified using the Get Information about Schools website, the Department for Education’s (DfE’s) register of schools in England. School contact details were then obtained from the DfE, and where available, contact details for Subject Leaders of PE were sourced from individual school websites. The survey targeted Subject Leaders of PE because they are typically responsible for policy-making within the department. Two emails were then sent one week apart, either directly to the Subject Leader of PE (where a personal school email address was found), or to the school administrator, with a request that they be forwarded to the Subject Leader of PE. The Subject Leader of PE was asked to forward the survey to another member of the PE department if applicable (i.e. if they had only been in post for a short period of time). The first email explained the purpose of the study, provided assurances of confidentiality, requested voluntary participation, and informed recipients that they would be receiving another email in one weeks’ time. Meta-analytic studies of online surveys show that personalised pre-notification emails can have a significant positive effect on response rates (Cook, Heath, and Thompson 2000). The second email provided a hyperlink to the online survey and detailed instructions on how to complete the survey. Informed consent was indicated by voluntary participation in the survey. Participation was incentivised through the opportunity to be entered into a draw to win one of eight Amazon gift cards valued at £50 each. The survey was opened from September 20^th, 2021, to October 22^nd, 2021, and one reminder email was sent to non-respondents two weeks before the survey closed. A total of 903 surveys were completed, representing 28.2% of the sample.

Characteristics of responding schools

The schools captured by the survey responses received were largely representative of secondary schools nationally, particularly in terms of admission policy, gender of entry, Ofsted inspection rating, and educational achievement. In the case of school type, there was a slightly greater proportion of academy converters and a smaller proportion of University Technical Colleges than in the national population. There was also a slightly higher response rate (based on the number of schools in the region) from schools in the North-East and South-East of England, schools with higher percentages of students eligible for free school meals (FSMs), and medium to large-sized schools. Table 3 presents the characteristics of responding schools.

Table 3. Characteristics of responding schools.

		Survey (n = 903) Frequency %		National (n = 3197) Frequency %
Location	East	100	(11.1%)	326	(10.2%)
	East Midlands	74	(8.2%)	283	(8.9%)
	Greater London	123	(13.6%)	503	(15.7%)
	North-East	65	(7.2%)	136	(4.3%)
	North-West	120	(13.3%)	457	(14.3%)
	South-East	165	(18.3%)	496	(15.5%)
	South-West	81	(9.0%)	296	(9.2%)
	West Midlands	97	(10.7%)	384	(12%)
	Yorkshire and Humber	78	(8.6%)	316	(9.9%)
School type	Academy converter	472	(52.3%)	1412	(44.2%)
	Academy sponsor-led	187	(20.7%)	709	(22.2%)
	Community	76	(8.4%)	360	(11.3%)
	Voluntary	65	(7.2%)	261	(8.2%)
	Foundation	52	(5.8%)	203	(6.3%)
	Free	50	(5.5%)	187	(5.8%)
	Technical college	1	(0.1%)	48	(1.5%)
	Studio school	0	(0%)	17	(0.5%)
Gender of entry	Co-educational	818	(90.6%)	2857	(89.4%)
	All girls	46	(5.1%)	190	(5.9%)
	All boys	39	(4.3%)	150	(4.7%)
Number of students	Less than 700	141	(15.6%)	627	(19.6%)
	From 701 to 1,300	507	(56.1%)	1682	(52.6%)
	More than 1,300	240	(26.6%)	781	(24.4%)
	Not recorded	15	(1.7%)	107	(3.4%)
Admission policy	Selective	44	(4.9%)	163	(5.1%)
	Non-selective	859	(95.1%)	3034	(94.9%)
Ofsted rating	Outstanding	128	(14.2%)	463	(14.5%)
	Good	468	(51.8%)	1644	(51.4%)
	Requires improvement	122	(13.5%)	433	(13.5%)
	Special measures	10	(1.1%)	50	(1.6%)
	Serious weaknesses	4	(0.4%)	36	(1.1%)
	Data is not available	171	(19.8%)	571	(17.9%)
Progress 8 score	Well above average	111	(12.3%)	387	(12.1%)
	Above average	150	(16.6%)	442	(13.9%)
	Average	330	(36.5%)	1117	(34.9%)
	Below average	139	(15.4%)	535	(16.7%)
	Well below average	73	(8.1%)	303	(9.5%)
	Data is not available	100	(11.1%)	413	(12.9%)
FSM proportion	Less than 15%	369	(40.9%)	1139	(35.6%)
	From 15% to 30%	361	(39.9%)	1301	(40.7%)
	More than 30%	156	(17.3%)	652	(20.4%)
	Not recorded	17	(1.9%)	105	(3.3%)

Where data is not available, this is because these schools recently changed their type (i.e. became an academy), re-opened after the closure of a previous school, or opened as a new school. A small proportion of mainstream state-funded secondary schools in England are academically selective grammar schools. The national data were calculated using school information from the Get Information about Schools website.

Analysis of survey responses

Data analyses were performed in two ways. First, the quantitative data from closed questions were analysed using descriptive statistics to calculate frequencies and percentages of responses. Percentages reported in results are in relation to the total sample (n = 903) unless otherwise stated. Where the number of responses to a question was different to the total sample, valid percentage results were calculated and are reported. Secondly, the qualitative data from written responses to optional, open-ended questions were analysed using content analysis. This entailed reading and rereading responses to identify patterns and relationships amongst the data. Responses were then coded into different categories and rank ordered by the most frequently occurring category to determine the level of agreement between respondents.

Results

The prevalence of particular ability grouping practices in PE

Tables 4 and 5 provide full details of the frequencies and percentages of ability grouping practices by year groups and classes (e.g. GCSE and/or BTEC) in PE in Key Stage 3 (Years 7–9) and Key Stage 4 (Years 10–11). As shown in Table 4, there was a notable increase in the use of setting and a decrease in the use of mixed-ability grouping as students moved through PE in Key Stage 3. 38.3% of schools were using setting in PE in Year 7, rising to 46.8% in Year 8, and 48.2% in Year 9, whereas 44.3% of schools were using fully mixed-ability grouping in PE in Year 7, falling sharply to 32.7% in Year 8 and 28.7% in Year 9.

Table 4. Ability grouping types by year groups in PE in Key Stage 3.

Type of grouping	Year 7	Year 8	Year 9
Setting	346 (38.3%)	423 (46.8%)	435 (48.2%)
Fully mixed-ability grouping	400 (44.3%)	295 (32.7%)	259 (28.7%)
Mixed-ability grouping with a top set	40 (4.4%)	67 (7.4%)	73 (8.1%)
Mixed-ability grouping with a bottom set	13 (1.4%)	19 (2.1%)	19 (2.1%)
Mixed-ability grouping with top and bottom sets	24 (2.7%)	25 (2.8%)	33 (3.7%)
Streaming/Banding	85 (9.4%)	84 (9.3%)	80 (8.9%)
Other	34 (3.8%)	27 (3%)	42 (4.7%)
Not applicable	3 (0.3%)	3 (0.3%)	1 (0.1%)

This was a multiple selection question, so the answers are more than the total number of responses. Where respondents answered not applicable, this was either because their school was new and therefore did not have any students in Year 7 or Year 8, or their school was a University Technical College and only enrolled students aged 14 and over.

Table 5. Ability grouping types by year groups in PE in Key Stage 4.

Type of grouping	Year 10 (Core)	Year 10 GCSE/BTEC	Year 11 (Core)	Year 11 GCSE/BTEC
Setting	274 (30.3%)	129 (14.3%)	231 (25.6%)	131 (14.5%)
Fully mixed-ability grouping	375 (41.5%)	610 (67.6%)	408 (45.2%)	605 (67%)
Mixed-ability grouping with a top set	71 (7.9%)	18 (2%)	55 (6.1%)	21 (2.3%)
Mixed-ability grouping with a bottom set	11 (1.2%)	3 (0.3%)	10 (1.1%)	2 (0.2%)
Mixed-ability grouping with top and bottom sets	31 (3.4%)	8 (0.9%)	33 (3.7%)	8 (0.9%)
Streaming/Banding	68 (7.5%)	42 (4.7%)	69 (7.6%)	42 (4.7%)
Other	92 (10.2%)	63 (7%)	108 (12%)	61 (6.8%)
Not applicable	6 (0.7%)	35 (3.9%)	13 (1.4%)	39 (4.3%)

This was a multiple selection question, so the answers are more than the total number of responses. Where respondents answered not applicable, this was either because their school was new and therefore did not have any students in Key Stage 4, they did not offer or failed to recruit students to GCSE and/or BTEC PE courses, or they offered something different to core PE – typically a PE-related qualification such as the Sports Leadership Award or the Cambridge National in Sport Studies.

Open ended responses that were received clarified that setting in PE was often delayed until Year 8 because teachers were reluctant to rely on or received no prior attainment data from feeder primary schools. In either case, it was explained that there was a need for a period of assessment of students in Year 7 before decisions could be made about their allocation to sets for PE in Year 8.

A much small proportion of schools were blending elements of mixed-ability grouping and setting in PE in Key Stage 3 (mixed-ability grouping with a separate top set the most common of these arrangements) or were using streaming or banding, or ‘other’ forms of grouping (see Table 4). These other grouping practices were diverse and included grouping based solely or in part on attitude, competitiveness, motivation, confidence, choice of curriculum activities (in Year 9 only), behaviour, and/or friendships.

As shown in Table 5, setting was used to a lesser extent in PE in Key Stage 4 than in Key Stage 3, with only 30.3% of schools using setting in core PE in Year 10 (aged 14–15), falling to 25.6% in Year 11, and a much smaller number of schools using setting in GCSE and/or BTEC PE in Year 10 (14.3% of schools) and Year 11 (14.5% of schools). The open-ended responses indicated that in some schools, there was no scope to use setting in GCSE and/or BTEC PE because there were only enough students to form a single teaching group that was thus necessarily deemed mixed-ability.

Overall, slightly fewer schools were using streaming or banding in Key Stage 4 in comparison to Key Stage 3. As with PE in Key Stage 3, mixed-ability grouping with a separate top set was the most frequent of the blended practices reported for PE in Key Stage 4. There was also a notable increase in the number of schools reporting other grouping practices in PE in Key Stage 4, particularly in core PE. These grouping practices mostly related to a system of pathways that were organised around different types of curriculum activities (i.e. team, individual, and/or health- oriented), with students self-selecting their ‘mixed-ability’ group based on the pathway they wanted to follow. This explanation from one respondent echoed Wilkinson and Penney’s (2022a, 2022b), research revealing the different orientation of different pathways:

We use a system where students select a pathway that is suited to their interests. They can choose either the recreational pathway (more individual activities) or the competitive pathway (more team activities). So, we end up with mixed-ability groups (Male Subject Leader of PE).

A relatively small number of schools (n = 185) were using the same grouping practice in all year groups and classes in PE in Key Stage 3 and Key Stage 4. Of these schools, 74.1% were using fully mixed-ability grouping, 16.7% were using setting, 6.5% were using streaming or banding, and 2.7% were using other forms of grouping based on attitude, motivation, and/or behaviour.

A small proportion of schools in the sample were also using more than one type of grouping practice in PE in Year 7 (4.3%), Year 8 (4.1%) and/or Year 9 (4.2%). This was particularly so in situations where different numbers of groups were timetabled PE together and/or where different ability grouping practices were used in boys’ and girls’ PE. We explore some of these nuances in approaches below. Fewer schools were using more than one type of grouping practice in core PE in Year 10 (2.1%), core PE in Year 11 (0.7%), GCSE and/or BTEC PE in Year 10 (0.6%), and/or GCSE and/or BTEC PE in Year 11 (0.7%).

Gender emerged as a factor influencing ability grouping decisions in PE in a small number of co-educational schools in the sample. Of the co-educational schools (n = 818), 2% were using setting for boys’ PE only (with fully mixed-ability grouping for girls’ PE) and 0.2% were using mixed-ability grouping with a separate top set for girls’ PE only (with setting for boys’ PE). Most of the respondents in these schools believed that girls were more cooperative and collaborative than boys and associated these values with the principles of mixed-ability grouping. Relatedly, most respondents in these schools expressed concern that setting would prospectively lead to feelings of anxiety for girls because it would separate them from many of their friends and thereby reduce their social support network. Two respondents explained:

In our experience, girls make greater progress in mixed-ability groups. We find that girls have lower levels of competition, better communication, better collaboration skills, as well as more empathy. So, they are more receptive to mixed-ability grouping (Female Subject Leader of PE).

The girls are more reluctant than the boys and a lot of the time friendship groups drive their motivation not ability groups. It would cause too much upset if they were split up (Male Subject Leader of PE).

Only a very small number of respondents were considering changes to their current ability grouping practices in PE in Key Stage 3 and/or Key Stage 4 (see Table 6).

Table 6. Potential changes to current ability grouping practices in PE in Key Stage 3 and Key Stage 4.

Potential new grouping type	Key Stage 3	Key Stage 4
Setting	12 (1.3%)	10 (1.1%)
Fully mixed-ability grouping	14 (1.6%)	9 (1%)
Mixed-ability grouping with a top set	12 (1.3%)	4 (0.4%)
Mixed-ability grouping with a bottom set	3 (0.3%)	2 (0.2%)
Mixed-ability grouping with top and bottom sets	8 (0.9%)	3 (0.3%)
Streaming/Banding	2 (0.2%)	3 (0.3%)
Other	15 (1.7%)	10 (1.1%)

The section that follows provides further insight into the range of ability grouping practices reported across Key Stage 3 and Key Stage 4 by addressing how schools arrived at particular arrangements for grouping in PE.

The processes and criteria used in enacting different ability grouping practices in PE

The organisation of mixed-ability grouping in PE was complex and varied between schools. Of those schools providing details of the processes and criteria used to create mixed-ability groups in PE (n = 562), 39.7% were using setting in another subject, particularly mathematics and/or English, 34.3% were using tutor groups, and 13.9% were using pathway choices. Grouping students according to their set placement in another subject can be problematised in relation to the spread of abilities consequently arising in ‘mixed-ability’ PE classes. Nonetheless, respondents recognised this arrangement as constituting mixed-ability grouping because they believed that ability was contextually specific to PE. The following comment was typical:

All groups are mixed-ability (in terms of practical ability) because students are in their maths sets for PE. Ability in one subject doesn’t necessarily relate to ability in another (Male Subject Leader of PE).

Mixed-ability groups in PE were also formed purposefully in some schools to ensure an even spread of ability or a particular mix of students in each group. For example, a small number of respondents (6.2%) explained that care was often taken to ensure that mixed-ability groups included a balance of students with learning and/or behavioural difficulties and/or students who would work well together. Two respondents commented:

We keep friendship groups that work positively with each other together. We also split disruptive groups of students to disperse any behaviour issues (Male Subject Leader of PE).

We space out SEND and vulnerable students so that staff have equal numbers in their class (Female Subject Leader of PE).

Of those respondents providing details of the processes and criteria used to create sets in PE (n = 410), many were using baseline activities (57.8%) and/or baseline tests (35.1%), often alongside their own observations and judgements. Baseline activities were subjective in nature, with students assessed by PE teachers across a range of curriculum activities, most notably team games. By comparison, baseline tests objectively measured students’ skill-related fitness components, including agility, speed, coordination, balance, and reaction time. A smaller number of these respondents reported using discussions with other teachers (2.4%), discussions with students (2.4%), and/or primary school data (1.5%) as the basis of setting decisions in PE.

Several of the respondents expanding upon their processes and criteria used to create sets (15.1%) were also keen to clarify that assessment was ongoing to ensure that students could be moved between sets where necessary and appropriate, including when they made progress, if they were deemed to be incorrectly placed in a set, if they misbehaved, and/or when the curriculum activity changed. One respondent elaborated:

The groups are not set in stone. We are very flexible in our approach. So, if a student progresses, we can move them up. We also move them based on the activity they are undertaking. They might excel in athletics but might not be very good at team games (Male Subject Leader of PE).

The mode and context of assessment used by respondents meant that setting decisions in PE were mostly based on students’ practical ability, and more specifically their motor skills in team-based activities. From the explanations of processes and criteria used in forming sets for PE, it was also clear, however, that considerations other than ability (however conceptualised and/or assessed) came into play in some instances. A number of respondents noted that they considered a variety of factors other than or in addition to ability in setting decisions, including attitude (9.5%), behaviour (7.8%), friendships (7.6%), confidence (6.1%), students with statemented SEND (1.7%), and/or attainment in maths, English and/or science (1.2%).

Similar processes and criteria were reported as being used in those schools using a combination of elements of mixed-ability grouping and setting in PE. Of the respondents providing information on these processes and criteria (n = 27), most used baseline activities (55.6%) and/or baseline tests (29.6%) to organise students into groups at the beginning of Year 7. The highest attaining students would then be allocated to the separate top set, and so on depending on the approach taken and the number of students required in each group. Practical ability was again mentioned as the main factor informing the organisation of students into these groups (74.1%), although several respondents (22.2%) also considered behaviour, friendships, and/or students with statemented SEND.

A small number of respondents (n = 37) provided details on how students were organised into streams or bands in their school. 67.6% of these respondents indicated that students were streamed or banded based on their Cognitive Ability Test (CAT) results and 32.4% indicated that students were streamed or banded based on their Key Stage 2 (aged 7–11) Standardised Assessment Test (SAT) results. SATs are compulsory tests in English (reading, grammar, punctuation, and spelling) and mathematics. By comparison, CATs are non-statutory tests that are generally used to assess students in three different areas: verbal reasoning (problem solving with words); quantitative reasoning (problem solving with numbers); and non-verbal reasoning (interpreting shapes and space). CATs were administered in the first term of Year 7 and students were streamed or banded based on the resulting data.

Reasons for different ability grouping practices in PE

The respondents were asked to provide details of the reasons for using different ability grouping practices in PE, to which 873 (97.3%) responded. The reasons given by respondents for using mixed-ability grouping in PE were predominantly pragmatic. Indeed, timetabling constraints was by far the most frequently reported reason for using this practice (19.7%). These respondents explained that PE was often timetabled against other subjects in the curriculum, particularly mathematics and English, and so had to share groupings. The lower status of PE in some schools also meant that grouping decisions were often determined by those made in other subjects in the school. One respondent summed up the situation and their resulting frustration as follows:

PE is usually timetabled against maths and English and these subjects take priority in the school. So, the groups that we have in PE are dictated by these subjects. We have no control over how students are grouped. It’s frustrating (Female Subject Leader of PE).

Other less frequently reported reasons for using fully mixed-ability grouping in PE were that it was school policy (7.6%) there were only a small number of students or teaching groups (6%), to reduce the chances of students being stigmatised and labelled as ‘low ability’ (4.6%), to encourage cooperative and collaborative learning (3.7%), to foster positive and supportive peer relationships (2.3%), to provide a sense of continuity for Year 7 students (2.3%), and to support fairness and equality of opportunity (1.6%).

There were also a wide range of reasons given for setting students in PE. 12.8% of respondents highlighted that setting reduced the spread of ability within classes and thereby better enabled them to vary their pace and level of instruction to cater for students’ needs. Specifically, this approach enabled teachers to stretch and challenge more able students (9.7%) and provide additional support for less able students in PE (10.1%). From teachers’ perspective, students were therefore more engaged in PE lessons, which resulted in them making more progress in their learning. One respondent explained:

We use setting because it helps us pitch lessons more appropriately to the needs of our students. It creates a culture of engagement that drives progress and attainment. Higher ability students are pushed more because they are working with others of a similar ability level (they are also not being held back by less able students) and less able students are supported more because they are able to work at their own pace (they are also not intimidated by more able students) (Male Subject Leader of PE).

The nature of the learning environment was also reported as an important factor in the decision to use setting in PE. Several respondents (6.3%) expressed the view that less able students were more comfortable working with and performing in front of similar ability peers in PE. Setting was also justified on the grounds of safety, with some respondents (3.1%) suggesting that it was the safest approach to teaching students in competitive team-based sports, particularly contact rugby. Two respondents summed up their feelings:

PE is a very public subject in the curriculum because students are required to perform in front of one another. This can make things a bit uncomfortable, particularly for less able students. Setting provides a bit of a safe space/safety blanket for them to express themselves and make mistakes because they are separated from more able students (Male Subject Leader of PE).

We also use setting for health and safety reasons. Due to the physical differences between some students, it is safer to split them up when delivering activities such as contact rugby (Male Subject Leader of PE).

Other less frequently reported reasons for using setting in PE were that it was the preferred approach of students (2.9%), timetabling meant that there were enough groups to enable setting to be used (1.4%), and to promote fair and balanced competition (1.3%).

Very few respondents commented on the reasons for using variants of mixed-ability grouping in PE. 1.4% explained that mixed-ability grouping with separate top and bottom sets was associated with catering for students at the extremes of the ability range. These respondents explained that the top and bottom sets were often deliberately smaller in size (than the mixed-ability middle) to enable teachers to stretch and challenge high ability students and maximise support for low ability students and those with statemented SEND. In a similar vein, mixed-ability grouping with a separate bottom set was used to support low ability students and those with statemented SEND in their learning (0.7%) and mixed-ability grouping with a separate top set was used to stretch and challenge high ability students in their learning (0.7%). 0.5% of respondents also noted that mixed-ability grouping with a separate top set was used to avoid singling out certain groups of students as ‘low ability’ and thereby reduce the impact of labels and stigma. This was explained in the following terms:

We don’t want to create a ‘bottom set’ because it carries a stigma – i.e., that you aren’t very good at PE. Labelling students like this can be demoralising and discouraging. We find that having a ‘top set’ challenges the more able and a ‘mixed-ability middle’ makes the rest feel more comfortable because the more able aren’t taking over the lesson (Female Subject Leader of PE).

A very small number of respondents (0.6%) similarly explained that they used grouping based on attitude, behaviour, friendships, confidence, and/or choice to avoid labelling students by ability in PE.

A very small number of respondents also indicated that streaming or banding in their school was imposed by the SLT (1.2%) and/or influenced by school timetabling arrangements (0.9%).

The use of the pathway approach in PE in Key Stage 3 and Key Stage 4 was associated with engaging students (4.6%), encouraging students to find an activity they enjoyed and fostering their subsequent lifelong participation (2.7%), and enabling students to exercise choice and control over their own learning (1.6%). One respondent explained:

Students are encouraged to take ownership of their learning by selecting their own pathway. We find that this approach improves levels of engagement because students are selecting the activities that they want to do. The idea is that they will find something that they enjoy and will continue to participate in throughout their lives (Female Subject Leader of PE).

Preferences for particular ability grouping practices in PE

Overall, substantially more respondents expressed a preference for setting than any other form of ability grouping in PE in Key Stage 3 and Key Stage 4. As shown in Table 7, this included 43.3% of respondents in Key Stage 3 and 28.9% of respondents in Key Stage 4. There was also a notable rise in the number of respondents reporting a preference for ‘other’ grouping practices (particularly curriculum pathway groups) in PE in Key Stage 4 compared to Key Stage 3. Table 7 provides the frequencies and percentages of grouping preferences in PE in Key Stage 3 and Key Stage 4.

Table 7. Ability grouping preferences in PE in Key Stage 3 and Key Stage 4.

Grouping preference	Key Stage 3	Key Stage 4
Setting	391 (43.3%)	261 (28.9%)
Fully mixed-ability grouping	78 (8.6%)	94 (10.4%)
Mixed-ability grouping with a top set	51 (5.7%)	49 (5.4%)
Mixed-ability grouping with a bottom set	12 (1.3%)	7 (0.8%)
Mixed-ability grouping with top and bottom sets	23 (2.6%)	19 (2.1%)
Streaming/Banding	7 (0.8%)	11 (1.2%)
Other	32 (3.5%)	68 (7.5%)
No preference	309 (34.2%)	394 (43.7%)

544 respondents provided comments on their reasons for preferring a particular ability grouping arrangement in PE in Key Stage 3 and/or Key Stage 4. The most frequently reported reasons for preferring setting in PE related to matters of workload, planning, and pedagogy, which were often interrelated. There was a belief among some respondents (18.8%) that it is much more time-consuming to prepare teaching for mixed-ability groups than setted groups in core PE, most notably because the wider range of abilities encompassed in mixed-ability groups required greater differentiation of work. However, several of these respondents expressed concern about the lack of time to effectively plan for and meet the needs of students in mixed-ability groups in PE. One respondent summed up the situation as follows:

You need to spend a great deal of time (that I often don’t have) developing differentiated materials for teaching mixed-ability groups. Sometimes you are teaching one lesson, but you are teaching something six or seven times over in different ways. So, it’s difficult to meet the needs of all students in mixed-ability groups (Female Subject Leader of PE).

Setting was perceived by many respondents (33.8%) to make planning and teaching easier and less stressful because it enabled instruction to be targeted more closely to the needs of students. This point was explained in the following way:

Setting is so much easier on teachers’ workloads. It makes your job so much easier in terms of planning and delivery because you don’t have that same level of differentiation. You can challenge all students at their level and make sure they all make progress (Male Subject Leader of PE).

Discussion

The findings of this study reveal the complex and varied nature of ability grouping practices in PE in secondary schools in England, with very few schools using the same approach across all year groups and formats (core, GCSE/BTEC) of PE, and multiple factors identified as shaping the grouping practices employed in specific contexts. Here we look across data relating to reported use of grouping practices, expressed preferences for particular practices, and reasons cited for their use, to critically engage with trends evidenced by this research. We highlight that while our data clearly affirms a need to move beyond simple, dualistic descriptions and conceptualisations of grouping practices in PE, it also speaks to the sustained dominance of certain discourses pertaining to grouping practices and pedagogy more broadly in PE. We draw particular attention to the need to probe the representations of ability and gender inherent in grouping practices in PE and hence, the function of grouping practices in supporting or inhibiting advances in equity in PE.

Mixed ability grouping: smoke and mirrors?

While fully mixed-ability grouping was the most common practice reported as being used in core PE in Year 7, Year 10, and Year 11, and the dominant practice in GCSE and/or BTEC PE in Year 10 and Year 11, setting was the most common practice in core PE in Year 8 and Year 9. These findings have some alignment with those of other studies showing high levels of mixed-ability grouping in PE (Hallam, Rogers, and Ireson 2008; Office for Standards in Education 2002, 2013), but they also point to much more extensive use of setting than has previously been reported. Furthermore, reference to ‘mixed ability’ grouping in GCSE and/or BTEC PE needs to be read within the context of student selections (and eligibility) for these courses. It is also notable that the survey revealed a high prevalence of ‘other’ grouping practices in Key Stage 4 particularly, including students self-selecting a ‘pathway’ group. Previous research indicates that pathways may express and/or result in de-facto ability grouping, with pathway descriptors (i.e. Performance or Participation) and curriculum content conveying a hierarchy between ‘mixed-ability’ pathway groups (Wilkinson & Penney, 2022a, 2022b).

Elsewhere in our data, there is further evidence that the label of mixed ability grouping obscures the role of other discourses in framing grouping practices in PE. For example, a small number of schools reported using a combination of ability grouping approaches, most notably mixed-ability grouping with a separate top set in PE in Key Stage 3 and/or Key Stage 4. In this instance, judgements about ability frame a differentiated grouping practice as appropriate to meet the needs of those learners deemed to be most able, while a ‘mixed ability’ arrangement is posited as appropriate for all other learners. Our findings also point towards the persistence of a gendered dimension to ability grouping practices in PE in some schools, with higher levels of setting in boys’ PE (with mixed-ability grouping in girls’ PE) and mixed-ability grouping in girls’ PE (with setting in boys’ PE). These differences appeared to be mostly tied to assumptions about girls’ learning preferences in PE, with girls perceived to be more cooperative and social than boys and some respondents associating these values with the underlying principles of mixed-ability grouping. In this respect, our data echoed previous research in evidencing the ways in which binary gender discourses continue to frame pedagogical practices in PE (Stride et al. 2022; Wilkinson & Penney, Forthcoming). We suggest that such commentaries on grouping practices highlight the pressing need for professional discourse and pedagogy in PE to embrace gender diversity and for schools and teachers to avoid grouping practices that legitimate stereotypical gender discourses in PE.

As we discuss below, schools used a variety of different approaches and criteria to allocate students to streams, sets, and mixed-ability groups in PE. Here it is pertinent to reiterate that in many instances, mixed-ability grouping reflected discourses of pragmatism rather than a pedagogical rationale, with factors such as the organisation of the timetable, the number of classes per year group, and/or the availability of staff leading identified as precluding exploration of other grouping practices. Several respondents also reported that they based mixed-ability groups on students’ attainment and subsequent set placements in other subjects in the school, particularly mathematics and English. Grouping students in this way raises questions about the spread of abilities consequently arising in ‘mixed-ability’ groups in PE.

Groupings and ‘ability’ in PE

Setting was by far the preferred grouping approach amongst respondents, including those in schools not currently using the practice in PE. In this respect our findings are concordant with previous research showing that teachers tend to exhibit more positive attitudes towards setting than other forms of ability grouping (Francis, Taylor, and Tereshchenko 2020; Hallam and Ireson 2003; Hallam, Rogers, and Ireson 2008; Taylor et al. 2017). Respondents in our study also indicated that this preference was particularly associated with the perception that a reduced range of ability made teaching easier and more manageable. Again, we are prompted to question how ability is being conceptualised and consider whether narrow notions of ability underpin the anticipated ease and manageability in teaching. While a direct response to this is beyond the scope of our data, we emphasise the importance of bringing a critical perspective to talk of ability in grouping debates and research in PE (see also Evans 2004; Hay and Penney 2013; Wilkinson and Penney 2021, 2022a; Wilkinson et al. 2021).

In most cases grouping decisions were reported to be based on ability and/or attainment, although very few schools were solely using these criteria to place students into groups in PE. Instead, as has been reported in previous research in classroom-based subjects (Francis, Taylor, and Tereshchenko 2020; Hallam and Parsons 2013a, 2013b; Ireson, Clark, and Hallam 2002), a variety of other factors were also considered, particularly social and working relationships between students, and as noted above, gender.

In allocating students to sets in PE, most respondents identified student performance in baseline activities and/or baseline tests as the prime point of reference. Some respondents also drew on their own observational assessments or knowledge of students, often in conjunction with baseline activities and baseline tests to organise students into sets in PE. While we cannot claim to have a full understanding of the scope of the baseline activities, tests, and/or observational assessments used in various schools, descriptions that some respondents provided of baseline activities and tests pointed towards a tendency for these to privilege a narrow conceptualisation of ability that centred on movement skills and values associated with games-based sports, including competitiveness and coordination. It is also pertinent to note that the limited use of setting in Year 7 was often associated with a perceived need for secondary school teachers to have time to formulate a ‘baseline’ judgement of students’ ability/ies in PE. We suggest that this in turn poses questions about the nature and particularly breadth of the Year 7 PE curriculum and the scope that it presents for all students to demonstrate their abilities in varied movement settings. More specifically, we highlight the need to probe the skills, knowledge, and understandings that assessments and associated judgements about ability are overtly and/or subtly privileging or in contrast, marginalising or excluding.

Conclusion

The research reported in this paper provides a snapshot of current ability grouping practices in PE in mainstream state-funded secondary schools in England. The findings particularly prompt critical engagement with the nuances of grouping practices across the years of secondary schooling and across schools, that take us beyond traditional binary approaches to ability grouping practices (i.e. setting or mixed-ability grouping) and that can extend understanding of the consequences of different forms of ability grouping in schools (Taylor et al. 2022). The findings have also revealed the ways in which grouping discourses and practices provide a mechanism for the expression and legitimation of established ability and gender discourses in PE. Grouping is therefore identified as an important focus in work directed towards disrupting the dominance of discourses that limit advances in equity in PE and this research is posited as presenting important new data and perspectives for teachers, teacher educators, agencies, and authorities supporting curriculum enactment, professional learning providers and researchers to consider.

While the research has provided valuable national level data and findings offering new insights into grouping practices in PE, some limitations should be acknowledged. As pointed out by others (Ireson and Hallam 2001; Taylor et al. 2022), it is difficult to establish the exact prevalence of different forms of ability grouping because grouping decisions are often made on a day-to-day basis to respond to the particularities of the learning context and/or the changing needs and interests of students. Further, ability grouping practices are not mutually exclusive, and as shown in this research and elsewhere (Taylor et al. 2022; Wilkinson, Penney, and Allin 2016; Wilkinson & Penney, 2022a, 2022b), schools often use more than one approach simultaneously and/or blend elements of one approach with another (i.e. mixed-ability grouping with a separate top set). A further level of complexity arises due to the existence of more subtle and less formal grouping practices, particularly within-class grouping, which we have not addressed in this paper. Thus, we acknowledge that the day-to-day reality of grouping students in PE is likely to be much more complex and nuanced than has been captured in the data reported. It is also important to note the absence of student voice in this study and a corresponding need for future research that captures students’ preferences and experiences of the different forms of ability grouping reported here. This is an area of research that is attracting increasing attention, particularly in mathematics and English (Francis, Taylor, and Tereshchenko 2020; Hallam and Ireson 2006; Tereshchenko et al. 2019), but that remains under-researched in PE and other subjects of the curriculum.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Notes on contributors

Shaun D. Wilkinson

Shaun Wilkinson is an Assistant Professor in the Department of Sport, Exercise and Rehabilitation at Northumbria University. He is a qualified physical education teacher and has previously taught in secondary schools. Shaun’s research interests relate to grouping and equity issues in physical education.

Dawn Penney

Dawn Penney is a Professorial Research Fellow in the School of Education, Edith Cowan University, leading projects focusing on policy developments and issues of equity in the education and sport sectors.