History education in comprehensive schools: using school-level data to interpret national patterns

Abstract

This paper reports the findings from two large-scale national online surveys carried out in 2009 and 2010, which explored the state of history teaching in English secondary schools. Large variation in provision was identified within comprehensive schools in response to national policy decisions and initiatives. Using the data from the surveys and school-level data that are publicly available, this study examines situated factors, particularly the nature of the school intake, the numbers of pupils with special educational needs and the socio-economic status of the area surrounding the school, and the impact these have on the provision of history education. The findings show that there is a growing divide between those students that have access to the ‘powerful knowledge' provided by subjects like history, and those that do not.

Keywords:

• history education
• curriculum policy
• policy enactment
• history curriculum

Introduction

The history curriculum has been, and remains, a focus of intense debate around the world, regarding both the purpose and content of what should be taught (see Osborne, 2003; Barton & Levstik, 2004; Grever & Stuurman, 2007; Taylor, 2009; Taylor & Guyver, 2011). In some respects the debate has centred on the relative benefits of acquiring a detailed factual knowledge of the past as opposed to developing an understanding of the nature of historical knowledge (although this is to simplify a very complex debate). Added to this are concerns about which version of the past is presented to young people; this could range from a simplistic national story, with the potential to generate patriotic fervour, to one that presents multiple and competing views of the past. Unsurprisingly, there is a great deal of rhetoric associated with history education. In the UK, the recent Ajegbo Report (DfES, 2007) called for a greater emphasis on history education to support an understanding of diversity within society and to develop a greater sense of identity. In contrast, Michael Gove, the current Secretary of State for Education, has called for a stronger ‘national' story to be taught, lamenting the fact that:

Children are growing up ignorant of one of the most inspiring stories I know—the history of our United Kingdom. The current approach we have to history denies children the opportunity to hear our island story. The trashing of our past has to stop. (Newsonnews, 2010)

Both Ajegbo and Gove seem to appreciate the importance of history in shaping a sense of identity but wish to cultivate different forms of identity.

Gove's views about the importance of history education have prompted him to include it as one of the subjects within his proposal for the award (or at least a new standard) of the English Baccalaureate to those students achieving GCSE grades A^∗–C ¹ in English, mathematics, science, a modern or ancient foreign language and either history or geography (DfE, 2011a, b). While the establishment of this new standard for measuring a school's success may well prompt a surge in the numbers of students entered for a GCSE in the subject, and enhanced provision for its study at Key Stage 3, a study of curricular provision over recent years reveals that despite all the declarations about the importance of history education, provision for actually teaching the subject has been extraordinarily mixed.

Recent reports by the Historical Association (HA) (Burn & Harris, 2009, 2010; Harris & Burn, 2011) have highlighted huge variation in the provision of an historical education for students in English secondary schools. This can be seen in a number of areas, for example: in the actual amount of time allocated to the subject and the stability of this time allocation; in the nature of the curricular provision (i.e. whether history is presented as a discrete subject, merged with humanities or offered in an alternative, competency based curriculum); and in the numbers of students who continue to study history beyond the end of Key Stage 3 (a choice usually made at the age of 14, but in some schools now at the age of 13).

What emerged most clearly, however, was a divide between different types of school. Independent schools and state maintained grammar schools (where they exist within the UK) were far more likely to offer greater discrete subject time and to have high numbers of students pursuing history beyond the age of 14. Academies, ² in contrast, tended to give less time to the subject, to merge it with other subjects or into alternative curricular provision, and to have low numbers studying the subject for GCSE. The picture within the independent and grammar schools suggests a ‘traditional' approach to the curriculum, which seems to imply that change is seen as neither necessary nor desirable. However, the academies seemed to be engaging in more curriculum experimentation in response to initiatives instigated by the New Labour government (which was in power at the time the surveys were conducted), as a means of improving the low level performance that had provided the rationale for giving them academy status.

The picture in state maintained comprehensive schools, however, was less clear cut. Within this sector, which currently forms the backbone of the secondary education system in the UK, history provision was extremely varied, covering the extremes observed in both the independent and grammar school sector and the academies. This begs the question why there is such a high level of variation within this type of school. Many comprehensive schools were able to maintain a ‘traditional' style curriculum, with a large proportion of time allocated to history teaching, often resulting in large numbers studying history at GCSE, whereas others were engaging in curriculum experiments merging history with other subjects or simply reducing the time available to the subject. This presents an interesting puzzle, especially as there had been no directives from government demanding such changes take place. There are arms-length devices (Hodgson & Spours, 2006) that create incentives for schools to maximise particular outcomes, such as examination pass rates, but how schools respond to these is an individual school decision.

Recent research into policy enactment in secondary schools suggests that the process by which schools, and departments within schools, choose what position to adopt in response to policy initiatives is complex, and depends on a mix of ethos, history and positioning:

[A]n examination of policy enactment at a school-level is about examining connections and inter-dependencies. It is important to consider, firstly, that policies are processes, even when mandated, and policy texts can be differently worked on and with. Secondly, policy practices are specific and contextualised. They are framed by the ethos and history of each school and by the positioning and personalities of the key policy actors involved. And thirdly, and related to the contextualised aspect of practice, policies are mediated by positioned relationships: between government and each local authority, the local authority and each of its schools, and within, as well as between schools. (Braun et al., 2010, p. 558)

As a consequence of these factors, it appears that some schools are able to ‘resist' or are confident enough to ignore certain policy initiatives, whereas others feel the need to embrace, or be seen to engage with, a raft of initiatives. Ball et al. (2011a) draw a useful distinction between imperative and exhortative policies. While some policies—most obviously the drive to raise (or maintain already high) standards of achievement—are imperatives for all schools, others may be perceived in different ways depending on the circumstances of the individual school. While those with lower standards of academic achievement and under pressure to demonstrate improvement may regard certain policies as imperatives, other schools, particularly those that are regarded as successful, can regard the same policies as merely exhortative, and therefore feel able to ignore initiatives that do not match the school's priorities.

Given the diversity of provision within the state maintained comprehensives, it was decided to analyse their responses to the 2010 HA survey of history teaching more fully, with the aim of identifying specific factors that might be associated with different levels of history provision within this group of schools. The extensive study of policy enactment recently conducted by Ball and colleagues has suggested that there are four different kinds of contextual factor that help to explain how and why particular schools respond in different ways to particular initiatives; these contextual factors may be situated (e.g. school histories, intakes and settings), professional (e.g. values, teacher commitments and experiences, ethos/emotional capital and ‘policy management'), external (e.g. extent and quality of local authority support, pressures and expectations from broader policy context such as Ofsted ratings, league table positions, legal requirements and responsibilities), and material (e.g. staffing, budget, buildings and infrastructure) (Ball et al., 2011b). Given the nature of the data we were able to collect through an online survey, it would obviously be impossible for us to examine all these potential influences; however, we were able to look at some elements of the situated factors—the nature of the each school's intake and setting (in the form of socio-economic data for the area around the school) and some external factors in terms of league tables.

Methodology

The data reported here are derived from a national online survey undertaken in 2010 on behalf of the Historical Association. This is the second such survey conducted by the HA, with the intention of building up a longitudinal data set, identifying key trends and the impact of specific developments in history education. This survey was carried out in the spring term of 2010, with invitations to complete an online questionnaire sent to all secondary schools in England. Responses were received from 531 schools, including 345 comprehensive schools which could be identified by their postcode. ³

The schools' postcodes could be used to link the data from the HA survey with what was known from secondary data sources about individual school performance (taken from Department for Education performance tables (DfE, n.d., a) and socio-economic datasets (DfE, n.d., b). This made it possible to analyse each school's provision for history in relation to a number of situated and external factors. These included: individual CVA ⁴ scores and the percentage of students obtaining five A^∗–C grades at GCSE (including English and mathematics) from 2006 to 2009; school profile data, namely the percentage of students identified with special educational needs (SEN) (either at School Action, School Action Plus or with a formal statement of their needs); ⁵ and a measure of the socio-economic status (SES) of the area in which each school was located, using a social deprivation index (IDACI), ⁶ which provided a SES score and ranking for each school.

The GCSE results, CVA scores, SEN profile and IDACI scores/ranks provided the school-level factors that formed the basis of analysis for the participating schools. These were used to examine particular trends and issues that had emerged from the original analysis of the HA survey, namely the type of curriculum model adopted for history teaching, the amount of curriculum time allocated to the subject and whether this had recently increased or decreased, the proportion of non-specialist teaching provided and the uptake of history at GCSE. Mean values for these school-level and attainment and contextual data were calculated for schools reporting in each of the ordinal response categories for the items in the HA survey. Confidence intervals were calculated for the means in order to indicate whether differences in means might be considered to be statistically significant. Spearman's rho correlations were calculated to examine the strength of association between each of the history curriculum factors and the range of school-level raw attainment and contextual variables.

Findings

First it was important to establish how representative those schools that had responded to the survey were of comprehensive schools generally. The data could then be analysed to explore the nature of the relationship, if any, between the school-level factors and the trends and issues identified in the HA survey.

The descriptive data in Table 1 show how the group of schools participating in the HA survey compares to the group of all comprehensive schools in England based on the school-level context factors in the Table. Although they appear broadly similar, the comprehensive schools that responded to the HA survey have significantly higher (p<0.05) mean%5+ A^∗–C GCSE (including English and mathematics [EM]) for 2007–2009 and significantly lower (p<0.05) percentage of students with SEN in each category. There is no significant difference in CVA score between the HA sample and the national set of comprehensive schools.

Table 1. Comparison of comprehensive schools in the HA survey to comprehensive schools in England

School level factors used for comparison	Mean (and standard deviation) of measure for all comprehensives in England	Mean (and standard deviation) of measure for participating comprehensives in HA survey with matched postcodes
	(N=2735^1)	(N=345 ^1)
% 5+A^∗–C inc English and maths in 2006	46 (16.0)	45.8 (17.1)
% 5+A^∗–C inc English and maths in 2007	45.6 (16.2)	48.0 (16.6)
% 5+A^∗–C inc English and maths in 2008	47.9 (15.3)	50.6 (15.9)
% 5+A^∗–C inc English and maths in 2009	50.0 (14.5)	52.4 (16.0)
% of all students with SEN at school action	9.4 (5.6)	8.5 (4.8)
% of all students with SEN at school action plus or statemented	14.1(7.6)	12.5 (6.9)
CVA score	1001.1 (19.9)	1001.2 (18.6)
^1Not all schools have matched data in every case so this represents the maximum N

Examining relationships between responses on the HA survey and school-level context factors

Type of history curriculum

Responses from schools in the HA survey, across the three years of Key Stage 3 (KS3), ⁷ indicated that the modal response in terms of curriculum model was for history to be taught as a discrete subject; this made up 70% of responses for Year 7, which rose to approximately 80% for Years 8 and 9. History as a discrete subject within humanities was the next most popular response for Years 7 and 8, accounting for around 11–14% of the responses. For Year 9 10% of the participating schools responded to indicate that history was an optional subject for the last year of KS3 (compared with less than 1% in Years 7 and 8), a response which may reflect the accelerated KS3 curricula followed by a number of schools that commence Key Stage 4 (KS4) ⁸ programmes of study a year early. The proportion of schools indicating that history is taught within an integrated humanities curriculum appears to diminish across the three years of KS3 from 9% in Year 7 to 2.5% in Year 9. For each of the three KS3 years less than 1% of the participating comprehensive schools responded to indicate they were adopting an alternative curriculum.

Further analysis of schools' arrangements for their KS3 history curriculum in relation to a range of school-level contextual factors showed that schools adopting curriculum models other than history as a discrete subject were likely to:

-	have postcodes in slightly more deprived areas based on the IDACI score for the school postcode;
-	have a higher percentage of students with SEN;
-	have a lower percentage of students attaining 5+A∗–C grades at GCSE including English and mathematics.

However, none of these differences in school-level contextual factors was significant at the 95% level (p<0.05), except the mean%5+ A^∗–C GCSE (EM) pass measure for schools for the various curriculum types for their Year 7 curriculum. Those schools teaching history as a discrete subject for Year 7 had significantly higher mean%5+ A^∗–C GCSE (EM) measures (for each of the years 2006–2009) than both those teaching history as a discrete subject within humanities and those teaching an alternative curriculum for Year 7 students. There were no significant differences in mean%5+ A^∗–C GCSE (EM) measures for 2006–2009 between the various types of curriculum for Year 8 and Year 9 students.

Trends in curriculum time allocation

Table 2 analyses the trends in time allocation in comprehensive schools. Schools were asked to indicate whether the time allocated to history had been reduced, increased or stayed the same over the preceding three years. The results in the Table show that schools reporting a reduction in the amount of KS3 teaching time devoted to history have significantly higher proportions of SEN students at School Action (for changes to Year 7 and Year 8 time only) and significantly lower mean%5+A^∗–C GCSE (EM) year on year when compared to schools with no change to the amount of teaching time.

Table 2. Comparison of trends in curriculum time allocation by school level factors

	Year 7 change in curriculum time			Year 8 change in curriculum time			Year 9 change in curriculum time
	–	none	+	–	none	+	–	none	+
N	77	219	15	76	233	13	59	244	18
mean %SEN SA	14.3^∗	11.8	13.8	14.8^∗∗	11.8	13.7	13.3	12.1	11.9
mean %SEN SAP/St	9.1	8.4	8.3	9.1	8.4	9.7	8.8	8.6	7.4
mean IDACI Score	0.21	0.18	0.15	0.21	0.18	0.20	0.21	0.18	0.13
mean CVA	1000.2	1001.5	1002.4	998.5	1002.3	1000.4	1000.0	1001.9	999.4
mean %5+A^∗–C (EM) mean 2006	41.0^∗∗	47.9	49.6	38.9^∗∗∗	48.3	42.4	39.2^∗∗∗	48.3	48.3
mean %5+A^∗–C (EM) 2007	43.3^∗∗	50.1	53.3	41.4^∗∗∗	50.5	45.6	41.6^∗∗∗	50.2	51.6
mean %5+A^∗–C (EM) 2008	46.4^∗	52.4	52.7	43.6^∗∗∗	52.9	49.4	44.4^∗∗∗	52.5	53.4
mean %5+A^∗–C (EM) 2009	48.4^∗	53.9	53.2	45.9^∗∗∗	54.3	51.9	46.2^∗∗	53.8	56.2
^∗ = significant at the 95% level (p<0.05), ^∗∗ = significant at the 99% level (p<0.01), ^∗∗∗ = significant at the 99.9% level (p<0.001), significance of difference in means compared to mean value for schools with no change in curriculum time

Some care needs to be exercised with these findings as the schools reporting an increase in the amount of teaching time devoted to history do not always consistently have lower proportions of students with SEN at School Action nor consistently higher mean 5+A^∗–C GCSE (EM) percentages.

The IDACI score for the school postcode also seems to be associated with the survey responses for changes to Year 7 and Year 9 teaching time but not for Year 8. However, the differences in mean school postcode IDACI score are not statistically significant.

Actual teaching time

Schools were asked to indicate how much actual time was allocated to the teaching of history on a weekly basis, from a number of given time allocations; these were 1–30 minutes, 31–45 minutes, 46–60 minutes, 61–75 minutes, 76–90 minutes and over 90 minutes. Schools could also indicate whether no time allocation was given or whether the subject was not compulsory in that particular year group.

The associations between actual teaching time devoted to history each week and the various school-level context variables exhibit a non-linear relationship. Figure 1 is typical and the teaching time associated with the highest mean%5+ A^∗–C GCSE (EM) seems to consistently lie around the 61–75 minute category for all three years of KS3 and across the GCSE results from 2006 to 2009.

Figure 1 Actual time allocation in Year 7 by mean percentage GCSE results in 2009

Perhaps not surprisingly, given the link between deprivation and academic outcomes, a similar relationship is observed for the IDACI score for the school postcode (see Figure 2) but there was no similar pattern observed in the association between actual teaching time and percentage of SEN either at School action plus/statemented or at school action.

Figure 2 Actual time allocation in Year 7 by IDACI score

Proportion of non-specialist teaching

The survey sought to discover the extent to which students in KS3 received non-specialist teaching (i.e. from teachers who had not been trained to teach secondary level history). Respondents were given different categories to choose from, which indicated what proportion of history curriculum teaching time was covered by non-specialists; the categories ranged from 0% (i.e. there were no non-specialists teaching history), through 1–15%, 16–30%, 31–45%, 46–60% to over 60% (see Tables 3-5).

Table 3. Comparison of trends in extent of non-specialist teaching in Year 7 by allocation by school level factors

Extent of non-specialist teaching	0%	1–15%	16–30%	31–45%	46–60%	61+%	correlation (ρ)
N	130	56	31	25	20	28
mean %SEN SA	12.1	12.6	13.4	13.4	13.8	12.8	0.09
mean %SEN SAP/St	7.9	8.7	9.0	9.5	9.8	9.4	0.15^∗
mean IDACI Score	0.17	0.22	0.15	0.23	0.18	0.23	0.07
mean CVA	1001.3	1003.1	998.3	999.8	998.5	997.2	-0.07
mean %5+A^∗–C (EM) 2006	48.6	47.2	43.8	40.7	38.4	43.0	-0.17^∗∗
mean %5+A^∗–C (EM) 2007	51.2	48.3	45.8	42.0	42.9	45.7	-0.16^∗∗
mean %5+A^∗–C (EM) 2008	52.8	50.1	49.2	46.9	47.8	48.6	-0.11
mean %5+A^∗–C (EM) 2009	53.6	53.0	50.6	49.8	47.6	49.2	-0.11
^∗ = significant at the 95% level (p<=0.05), ^∗∗= significant at the 99% level (p<=0.01)

Although none of the differences in mean values of the school-level context factors were significant there are some associations to observe here including how the%5+A^∗–C figure is associated with the proportion of non-specialist teaching time. These associations are quite weak as illustrated by the values of the correlation coefficients. Generally it can be seen that higher GCSE results are associated with lower proportions of non-specialist teaching, although this association is weaker at the 46–60% and 61%+ categories..

Table 4. Comparison of trends in extent of non-specialist teaching in Year 8 by allocation by school level factors

Extent of non-specialist teaching	0%	1–15%	16–30%	31–45%	46–60%	61+%	correlation (ρ)
N	173	57	26	16	18	14
mean %SEN SA	12.1	12.3	13.5	18.1^∗	13.4	14.1	0.12^∗
mean %SEN SAP/St	8.2	8.5	10.2	10.4	8.7	10.3	0.14^∗
mean IDACI Score	0.17	0.18	0.23	0.25	0.19	0.26	0.13^∗
mean CVA	1000.4	1003.7	998.4	1002.0	1000.8	994.0	-0.02
mean %5+A^∗–C (EM) 2006	48.6	44.9	40.3	35.9	38.5	41.1	-0.22^∗∗
mean %5+A^∗–C (EM) 2007	50.7	47.6	40.9	37.8	45.7	42.4	-0.20^∗∗
mean %5+A^∗–C (EM) 2008	52.7	49.8	45.6	41.3	47.8	45.0	-0.19^∗∗
mean %5+A^∗–C (EM) 2009	53.5	53.3	47.4	46.3	47.4	44.7	-0.17^∗∗
^∗= significant at the 95% level (p<=0.05), ^∗∗= significant at the 99% level (p<=0.01)

When examining the extent of non-specialist teaching in Year 8 (see Table 4) in relation to different school-level factors, again there are associations to observe between the%5+ A^∗–C (EM) figure and the proportion of non-specialist teaching time. There is a similar anomalous behaviour of the categories at 46–60% and especially 61+%.

Table 5. Comparison of trends in extent of non-specialist teaching in Year 9 by allocation by school level factors

Extent of non-specialist teaching	0%	1–15%	16–30%	31–45%	46–61+%^1	correlation (ρ)
N	212	44	23	16	13
mean %SEN SA	12.4	11.3	12.9	15.5	17.9	0.09
mean %SEN SAP/St	8.4	7.9	11.3	8.7	11.4	0.08
mean IDACI Score	0.18	0.15	0.25	0.18	0.25	0.05
mean CVA	1000.9	1002.1	997.7	1002.2	994.8	-0.04
mean %5+A^∗–C (EM) 2006	47.6	45.9	37.4	42.1	36.8	-0.16^∗∗
mean %5+A^∗–C (EM) 2007	49.4	49.0	38.7	44.6	39.2	-0.15^∗∗
mean %5+A^∗–C (EM) 2008	51.9	50.8	41.0^∗	49.3	44.1	-0.15^∗
mean %5+A^∗–C (EM) 2009	52.9	54.6	44.8	50.7	41.6	-0.12^∗
^∗ = significant at the 95% level (p<=0.05), ^∗∗= significant at the 99% level (p<=0.01)^1 The last two categories were merged as both had fewer than 10 schools.

In Year 9 (see Table 5) the only significant associations observed with the proportion of non-specialist teaching are again with GCSE raw attainment at the school-level as measured by the percentage of students attaining 5+ A^∗–C GCSE grades including English and maths. Once again it is important to note that these associations are weak, although in Year 9 the weakness of the association seems to be due to those schools in the group reporting 16–30% non-specialist time in Year 9 having anomalously low mean raw attainment figures across the years surveyed. These schools also have a lower mean CVA score and a mean IDACI score suggesting they might be sited in more deprived neighbourhoods.

Year 10 GCSE history uptake

Towards the end of KS3, students in England are given a degree of choice about the subjects they will continue to study. Some subjects, such as English, mathematics and science are compulsory for all students, whereas others are optional. Although there is evidence that the vast majority of students enjoy history and regard it as useful (QCA, 2005; Harris & Haydn, 2006, 2008), nationally only about 30% of students choose to study it further. The reasons for this are complex (see Harris & Haydn, 2010) and will be discussed further below, but the analysis of the data (as shown in Table 6) shows there is an association between GCSE history uptake in Year 10 and a number of school-level factors. The strongest associations, as evidenced by the values of the correlation coefficients, are the relationships between school mean%5+A^∗–C GCSE (EM) and Year10 GCSE history uptake (see Figure 3). There are also significant associations between GCSE uptake and IDACI score and the proportion of SEN students (see Figure 4).

Table 6. Comparison of GCSE uptake in history for Year 10 pupils (aged 14–15)

Percentage of pupils who opted to study history at GCSE	0–15%	16–30%	31–45%	46–60%	61+%	Correlation (ρ)
N	42	96	98	68	31
mean %SEN SA	17.4	13.0	11.9	11.2	9.6	-0.27^∗∗
mean %SEN SAP/St	10.6	9.6	7.9	7.3	7.4	-0.25^∗∗
mean IDACI Score	0.25	0.21	0.16	0.16	0.15	-0.19^∗∗
mean CVA	1004.5	1000.7	999.7	1001.6	1000.5	-0.01
mean %5+A^∗–C (EM) 2006	32.0	40.3	47.9	42.9	57.5	0.43^∗∗
mean %5+A^∗–C (EM) 2007	34.9	43.4	50.8	54.2	57.6	0.39^∗∗
mean %5+A^∗–C (EM) 2008	38.5	45.8	52.2	56.6	59.3	0.39^∗∗
mean %5+A^∗–C (EM) 2009	39.6	48.3	53.2	58.1	61.3	0.38^∗∗
^∗∗= significant at the 99% level (p<=0.01)

Figure 3 Relationship between mean percentage GCSE pass rates and history GCSE uptake in Year 10

Figure 4 Relationship between percentage of students with SEN (School Action level) and history GCSE uptake in Year 10

CVA scores

Interestingly, there is no association between a school's CVA score and the kinds of variation in time allocation and uptake of history identified within the HA survey. CVA scores attempt to capture the school effect, i.e. the value added by the school to students' academic outcomes after a wide range of pupil and school-level contextual factors beyond the control of the school have been taken into account. What remains within the CVA score should be those factors within the control of the school which affect academic outcomes, such as those relating to teaching, learning, leadership and policy. However, there is no discernible relationship between schools that have high or low CVA scores and factors such as time allocation for history and the kind of curriculum models which schools adopt.

Discussion

Although much of the data suggests only tentative associations between the school-level factors and the ways in which schools have chosen to shape and structure students' experiences of history within the curriculum, there do seem to be some important correlations. In particular there seem to be associations between GCSE uptake in Year 10 and the school's overall GCSE pass rate, IDACI score and percentage of students identified with SEN. Analysis of the data reveals that schools with good pass rates are more likely to see higher numbers of students studying history at GCSE. Given the links between academic success and socio-economic background it is likely that most of the schools in the survey with high GCSE pass rates are located in more affluent areas, as indicated by their IDACI scores. The implication of this is that students in more socio-economically deprived areas are also less likely to study history. This is borne out by the strong association found in the data between less wealthy socio-economic areas and lower numbers of students studying history at GCSE. The similarly strong association between a high percentage of students with SEN and low GCSE uptake also suggests that schools with significant proportions of students with SEN are also less likely to enter students for subjects such as history that are often seen as more ‘academic'. Schools with high numbers of students with SEN and poor IDACI scores are also more likely to reduce curriculum time for history at KS3 and therefore have fewer students opting to continue with the subject at GCSE.

These associations do not imply the presence of any direct causality. The relationship between the uptake of GCSE history by pupils and school-level factors such as IDACI score for the school location or proportion of pupils with SEN will be complex, with a number of underlying factors influencing both. Other associations were also explored but these were found to be not as strong, for example each of the associations between the proportion of non-specialist teaching time in Years 7, 8 and 9 and the proportion of students taking up history were much weaker with values of the Spearman's correlation (ρ) of 0.21 or less (p<=0.01). Nonetheless, the HA surveys (Burn, 2009, 2010) have already shown that there is a strong statistical correlation between time allocation at KS3 and GCSE uptake, so that there seems to be a cumulative, if complex, relationship between the uptake of history and the combination of a reduced time allocation, low socio-economic status and high percentage of SEN.

Taken in isolation the trends identified in this paper may not appear very significant, but the accumulation of evidence through different studies (QCA, 2005; Burn & Harris, 2009, 2010; Haydn & Harris, 2010) is creating a picture of a subject that has come under particular pressures in recent years. The findings raise questions about the purpose of education and importance of the development of specific kinds of knowledge within that education, and the extent to which failure to provide access to an historical education may further disadvantage those who already face disadvantage (Harris & Burn, 2011).

This research therefore poses a number of questions:

1.	Has there been a deliberate steer by schools in terms of subject choices to maximise students' GCSE results or do students make their own ‘informed' choices?
2.	Does it matter if some students, especially those from areas of poor SES and students with SEN, receive less history teaching than others?
3.	If an historical education is considered important, to what extent does the form of this education matter?
4.	What is likely to be the response by schools to the introduction of a new policy context in the form of an ‘English Baccalaureate' and in what ways, in particular, will it impact on pupils facing economic disadvantage or with SEN?

Has there been a deliberate steer by schools in terms of subject choices to maximise students' GCSE results or do students make their own ‘informed' choices?

The extent to which the trends presented in this paper have been caused by schools deliberately steering students towards particular subjects or whether they are the result of pupil choice is unclear. Evidence already exists (Burn & Harris, 2009, 2010; Harris & Haydn, 2009) that in a number of schools, students are actively discouraged from studying history or guided towards subjects in which it is deemed to be ‘easier' to obtain a good GCSE pass or that carry a greater number of GCSE equivalent grades, and are therefore seen as more ‘value for money'. However, there are a complex range of factors influencing students' perceptions of, and decisions about, the value of GCSE history. Harris and Haydn (2006) have shown that, although 70% of students profess to enjoy history, only about 30% of them continue their studies when it becomes optional. Haydn and Harris (2010) also found that approximately 70% of students thought that history was a useful subject, although they found it difficult to express the ways in which it was useful. Given this professed level of enjoyment and value, why is it that less than half of those who claimed to enjoy and value the subject choose to pursue it when it becomes optional? According to Harris and Haydn (2006), students' enjoyment and therefore their willingness to pursue the subject further is specifically linked to particular teachers, to the pedagogical approaches adopted and to the choice of content. These findings support earlier work by McCrone et al. (2005), who identified teachers and pedagogy as important factors in students' choice of subjects. But their study also identified that students preferred to take subjects that they considered easier to pass, which were related to an interest outside of school and/or had relevance to later career choices. In a couple of these respects, history encounters difficulties. According to Haydn and Harris (2010), many students do not fully appreciate the purposes of studying history, nor, according to a QCA report (2005) are they able to relate it to future employment prospects (beyond obviously history-specific jobs such as history teacher and archaeologist). If students are also looking for an ‘easy' subject then the perception of history as an ‘academic' subject may be a barrier. Research commissioned by the Engineering and Technology Board (ETB) (2005) suggests that students from poor SES backgrounds were less willing to pursue subjects that were seen as academic, and, as Adey and Biddulph (2001) have shown, the majority of students believe history is a subject that requires extensive reading and writing at GCSE, and is therefore challenging. Although judging the relative complexity of GCSE subjects is difficult, using a Rasch model, Coe's (2008) analysis of results suggests that history is a relatively difficult subject in which to obtain high grades.

It seems that history has been affected by two pressures; first, the pressures on schools to improve examination results have meant some school leaders have tried to manage students' subject choices to maximise examination grades (and therefore boost the position of schools in national league tables); and secondly, the perception of history as a challenging subject means some students look for alternative options. Interestingly the research suggests that students' understanding of what history will be like beyond KS3 is not always accurate (Biddulph & Adey, 2003), nor are students making ‘informed' choices as their understanding of the subject, its nature and value, is limited (Adey & Biddulph, 2001; Haydn & Harris, 2010). Central to our discussion of this issue is an assumption that history matters—but is it really a matter of concern if schools steer students away from history or students choose not to study it further?

Does it matter if some students, especially those from areas of poor SES and students with SEN, receive less history teaching than others?

Clearly all students within state maintained comprehensive schools are supposed to be taught history in some form during the compulsory stage of the National Curriculum. However, as this research and previous studies have shown, the extent of their history provision varies considerably from school to school (Burn & Harris, 2009, 2010, Harris & Burn, 2011). Although there are general differences determined by type of school attended and, as the analysis in this paper shows, by the specific nature of the different comprehensives surveyed, students from poorer socio-economic backgrounds and/or with special educational needs are likely to receive less history teaching at KS3 and are less likely to study history beyond KS3. This restricted access to an historical education should be a matter of concern.

Several powerful arguments have been put forward to justify the place of history in the curriculum. The range of these arguments (and the tensions between them) are too numerous to explore in depth but, as the political claims cited in the early stages of this paper reveal, many of them relate both to the development of young people's sense of identity—their sense of their place in the world in which they live—and to their capacity to contribute to it as informed citizens. Barton and Levstik (2004, p. 40) argue, for example, that ‘history's place in the curriculum must be justified in terms of its contribution to democratic citizenship—citizenship that is participatory, pluralist and deliberative'. The GCSE Subject Criteria echo these ambitions in declaring that active engagement in the process of historical enquiry should enable young people ‘to develop as critical and reflective thinkers', aware of ‘how the past has been represented, interpreted and accorded significance for different reasons and purposes', and capable of organising and communicating their historical knowledge and understanding and of reaching ‘substantiated judgments'. Those pursuing GCSE History should explicitly recognise how that knowledge and understanding, and the skills that underpin them, ‘help them to understand the present and also provide them with a basis for their role as responsible citizens' (Ofqual, 2011, p. 3). There is a danger that restricting access to an historical education, particularly for students from areas of deprivation and/or those with special educational needs, may serve to disenfranchise them and deny them access to the powerful knowledge, both substantive and syntactic, which will help them to understand themselves and the world in which they live.

If an historical education is considered important, to what extent does the form of this education matter?

For young people genuinely to engage in democratic society they need access to and understanding of the issues being debated within that society; these occur within a historical context. In this sense Hirsch's (1987) idea of ‘cultural literacy' and Bourdieu's (1986) cultural capital are important elements needed to provide students with the means to understand and engage in a meaningful way in public discourse. Young people certainly need to understand at least something of the historical references that are routinely used in debating public policy (whether in relation to the economy or international affairs). In some ways, Gove's (Newsonnews, 2010) desire to promote history teaching should therefore be welcomed and the emphasis on teaching of a national story could certainly ensure that more students receive a more substantial history education.

However, there are a number of concerns about the current review of the National Curriculum. Much discourse around the proposals for a revised curriculum (Schama, 2010; BBC, 2011) has focused on a content-driven curriculum. Evans (2011) has attacked the ideas being expressed by Gove for their intention to create ‘a narrowly nationalistic identity' and claims that an undue focus on British history ‘is a radically ignorant form of dumbing down', because it adopts a very narrow view of identity. The proposals for the revised curriculum adopt what Moore and Young (2010) refer to as a neo-conservative approach to the curriculum, which emphasises the idea of a traditional canon and deference to the importance of authoritative knowledge. While knowledge of that canon, as Hirsh (1987) has argued, undoubtedly gives young people access to particular debates and discourses, it does not equip them either to understand why certain events or developments have been identified as canonical or give them the capacity to argue for alternative frames of reference. Young (2008, 2011) criticises such an approach as presenting young people with the ‘knowledge of the powerful' rather than genuinely ‘powerful knowledge', as it implies the transmission of knowledge as a ‘given'. Such criticisms form part of a growing debate about knowledge in the curriculum and the central importance of knowledge to young people, driven by social realist theory (Maton & Moore, 2010).

As part of this debate, Young and Muller (2010) discuss three different curricula models, which they refer to as ‘Future 1' (effectively inert ‘knowledge' transmitted to students), ‘Future 2' (an undifferentiated focus on skills or competencies—on the assumption that knowledge can be acquired as and when it is needed) and ‘Future 3' (based on a view of knowledge that combines content and concepts to enable students to appreciate how that knowledge has been constructed and the concepts by which it is organised and validated). ‘Future 1' best describes the proposals, mentioned above, regarding the revised National Curriculum. The impact of a curriculum which moves towards a ‘Future 1' model could potentially deny all students access to ‘powerful knowledge'. ‘Future 2' curriculum models already exist in many schools (see Burn & Harris, 2009, 2010), but competency-based curricula in which subjects are merged together are criticised by Young and Muller (2010) because they treat all kinds of knowledge as though they are the same. By treating all knowledge as mere ‘information' which students simply need to learn to access and deploy effectively, such curricula deny students the opportunity to appreciate why certain kinds of knowledge are valued, and how they themselves could come to contribute to the creation of that knowledge. For knowledge to be powerful, Young and Muller argue that curricula should be based around the ‘Future 3' model. Not only do students learn ‘content', but they learn about the generative processes by which claims to knowledge are made. As Wheelahan (2010, p. 99) argues: ‘students need access to the disciplinary “style of reasoning” … to move beyond a focus on isolated examples of content. Specific content is the product of disciplinary knowledge'.

Within the context of history education, students need to understand that there is no single fixed account of the past, whose meanings are clear and from which obvious ‘lessons' can be neatly abstracted. They need an awareness that historical knowledge is a construction—not a copy—of the past, and of the standards by which its validity can be assessed. They also need to understand enough of the fundamental processes of change through time— one of history's essential organising concepts—to appreciate the suspect nature of many of the historical analogies used in political debate.

To achieve such understandings all young people therefore need access to a history curriculum to which sufficient time and specialist attention is devoted, and which does not adopt a simplistic transmission model but focuses clearly on the discipline of history. Only time and expert teaching will equip students with sufficient substantive knowledge to create what Rogers (1987) has described as a ‘frame of reference', and an effective understanding of the grounds on which historical knowledge is constructed and may therefore be contested. The first alone is not enough. As Shemilt (2000, p. 100) has argued, in simply presenting students with a ‘received' version of the past—no matter how coherent or widely shared—the danger remains that:

… to subscribe to populist and mythic constructions of the past is to remain trapped in the codes and culture of the street gang, to invoke persuasive and partial histories that reinforce simple truths and even simpler hatreds.

Young people need to be able to test ideas and challenge assumptions and generalisations they encounter, either in school or everyday life. As Evans (2011) has argued, ‘History is a critical academic discipline whose aims include precisely the interrogation of memory and the myths it generates.'

The concern emerging from recent work (Burn & Harris, 2009, 2010; Counsell, 2011) is that the ‘powerful knowledge' provided by a historical education is becoming the preserve of a narrower range of students, essentially determined by the type and nature of the schools which they happen to attend. Although the government's desire to review the curriculum may create a greater emphasis on history education, there seems to be a tension within government policy; on the one hand there is a desire to promote history teaching in the curriculum (albeit in a different form to that which currently exists), whilst on the other hand there is a move towards granting more schools ‘academy' status, effectively freeing them from all obligations to follow the new curriculum. In such a scenario it is not at all clear what form history teaching would take, nor how much history teaching would actually happen. It is possible that the introduction of an ‘English Baccalaureate' might go some way to resolving this issue, but it also raises further questions.

What is likely to be the response by schools to the introduction of a new policy context in the form of an ‘English Baccalaureate' and what particular impact will it have on pupils facing economic disadvantage or with SEN?

The introduction of an ‘English Baccalaureate' by the current government as a measure of school success poses many questions for schools. By introducing a new measure by which to judge schools, namely a good GCSE pass in English, mathematics, science, a modern or ancient language, and a humanities subject (defined as either history or geography), the government has made a strong statement about what it considers to be an appropriate education for young people. Although there is no compulsion on schools to ensure all students study these five areas, this is a further example of the kind of arm's-length governance (Hodgson & Spours, 2006; Ozga, 2009) by which New Labour's emphasis on securing five A^∗–C grades previously encouraged many schools to steer students towards those subjects or types of courses in which they thought it would be easier for them to succeed. It would be a bold headteacher who chose to ignore this latest measure of school performance, and they will perhaps only do so in those contexts where a particular combination of situated and professional factors (Ball et al., 2011a) makes schools confident enough to follow their own educational agenda.

Exactly how schools will respond is, however, uncertain. In some cases, where students are actually able to study both history and geography, the introduction of the English Baccalaureate may serve to reduce choice. Those schools that have previously actively discouraged students from studying history or ‘steered' them towards ‘easier' subjects will either have to ignore the newest initiative (and run the risk of poor league table results) or determine whether to enter more students for history or geography. The choice that they make will again depend on the precise conjunction of local circumstances.

Currently, about one third of students study history at GCSE, with slightly fewer taking geography. This leaves a sizable proportion of the pupil population who study neither, and, as the results of this study suggest, this is often associated with schools in poor socio-economic areas or high numbers of students with SEN. The perception of both history and geography is that they require high literacy levels (Adey & Biddulph, 2001), which are likely to be seen as a barrier to many students with special educational needs (partly accounting, as we have argued, for the correlation in the survey between schools with high numbers of students with SEN and low GCSE uptake). But assuming that schools do choose to follow the English Baccalaureate model, will they guide students towards one particular humanities subject to position themselves better in league tables—or will they provide students with a free choice of history or geography?

Although both are often seen as ‘academic' subjects, Coe's (2008) analysis of examination grades shows that there are grounds for regarding history as marginally the more difficult of the two. There is thus a possibility even in conforming to the standard of the new English Baccalaureate that schools could create a ‘two tier' system in which some students are who are perceived to be less ‘academic' continue to be ‘steered' towards the ‘easier' subject. This raises two issues, again the question of whether it is appropriate to deprive the lowest attainers of education in a subject so critical to full participation in a democratic society; but also as to the impact that such a move might have on the status of geography.

The perceived difficulty of both subjects (Biddulph & Adey 2003) also raises legitimate concerns about what happens to students who are presumed not to be capable of achieving a good GCSE pass in either subject, focusing attention on the accessibility of current examination specifications. If students are presumed to be incapable of achieving a ‘C' grade in either subject, will there be any scope at all for them to continue with the study of history or any other humanities subject? Proper attention to this question requires a more concerted consideration of the shape and nature of the whole curriculum and assessment system from pre-school to the end of compulsory schooling; as Counsell (2011) argues, it is pointless changing one part of the education system in isolation, as has happened too frequently in the past. Yet there are models of successful practice, where curricula and assessment arrangements have been modified, making the assessment of historical knowledge and understanding less acutely dependent on levels of literacy. The OCR ‘pilot' GCSE appeared to offer one such successful model, as Ofsted (2011) highlighted:

Evidence from schools visited which were involved in the GCSE history pilot course indicates that it has had a positive impact on students and teachers. The course was started in September 2006; three cohorts of students have now received their final grades. Students particularly liked the approach to assessment and teachers noted that the option was attracting students of all abilities.

The model, however, was withdrawn with effect from September 2011.

Conclusion

This paper highlights the disparities of curricula provision between schools within the comprehensive school sector. Although differences between schools per se are to be welcomed, situated factors, particularly the nature of the school intake, the numbers of pupils with special educational needs and the socio-economic status of the area surrounding the school, clearly impact on the provision of history education. The problem is that there is a growing divide between those students that have access to the ‘powerful knowledge' provided by subjects like history, and those that do not. Given the importance of such knowledge, the danger is that certain types of pupils will be disadvantaged by the curricula decisions made by schools. Although the current government is introducing or consulting on new policies that may influence this situation, the factors that shape the interpretation of policy mean that the development of a two tier curriculum may well continue. And the potential losers in such a situation are those students from disadvantaged backgrounds who can least afford to be denied access to ‘powerful knowledge'.

Notes on contributors

Richard Harris is a Lecturer in Education at the University of Reading. His research interests focus on history teaching and history's place in the curriculum, and diversity. His recent publications have appeared in the Curriculum Journal, Journal of Curriculum Studies, Journal of Education Policy and Research Papers in Education.

Christopher Downey is a Lecturer in Education at the University of Southampton. His research interests lie predominantly at the interface between educational effectiveness and improvement, especially in how data are used by teachers and school leaders to inform school self-evaluation and development planning.

Katharine Burn is a Senior Lecturer in History Education at the Institute of Education, University of London. Her research interests focus on history education and teachers' professional learning. Her recent publications have appeared in the Oxford Review of Education, Research Papers in Education, Journal of Education Policy and the Journal of Education for Teaching.

Notes

1. GCSEs are the main examination type followed by students generally aged 14–16. Although a range of grades from A^∗ to G can be awarded, A^∗ to C are regarded as a ‘good’ pass, and these pass rates are frequently used to position schools within national league tables.

2. Academy schools were introduced into areas of high socio-economic deprivation (as designated by numbers of pupils eligible for free school meals) as an answer to perceived poor performance of existing schools under the New Labour government (National Audit Office, 2007). The recent coalition government has introduced a new type of academy school; all schools deemed to be good or better have the ability to apply for academy status which will give them greater independence from local authority control (Great Britain, Academies Act, 2010). These new academies did not exist when these HA surveys were conducted. Both types of academy schools have great freedom in designing their own curricula.

3. Responses were received from 411 comprehensive school teachers, but they could choose whether or not to disclose their school’s identity by entering a postcode. Even where they chose to enter the postcode it was not always correct.

4. CVA stands for contextual value added. This is a score developed by government agencies to measure the performance of a school. It is designed to take into account a number of school factors outside the school’s control, which may influence a school’s ‘raw’ data performance and put these into a context—for example, the number of students on free school meals (used as a measure of deprivation), the number of students with identified special educational needs or students with English as an additional language. A score below 1000 indicates a school has not added any value to a child’s education, whereas a score above 1000 suggests a school is adding value.

5. School Action, School Action Plus and ‘statement of special educational need’ all refer to different levels of support required for pupils identified with SEN. School Action means the needs are less severe and a school should be able to deal with any issues, whereas a statement means that additional funding and external support are needed to meet a child’s needs.

6. The Income Deprivation Affecting Children Index (IDACI) is a UK government produced measure of deprivation. The scale ranges from 0 to 1 with higher values showing greater levels of deprivation. The scores here were found by entering the school postcode into a search of the IDACI database and so do not represent the actual IDACI scores associated with the pupils known to attend each school. It represents instead the level of deprivation of the area in which the school is situated.

7. Key Stage 3 refers to the compulsory stage of secondary school in which schools are obliged to follow the National Curriculum (unless they are independent or academy schools). The age range is usually 11–14 years (although some schools have shortened this to 11–13). Year 7 refers to pupils aged 11–12, Year 8 to those aged 12–13 and Year 9 to 13–14.

8. Key Stage 4 refers to the final compulsory years of schooling, usually from the ages of 14 to 16. Pupils at this stage are expected to follow courses in a number of compulsory subjects and topics but have some choice over others.