High cost subjects analysis using TRAC(T) data: detailed commentary

Transcription

1 High cost subjects analysis using TRAC(T) data: detailed commentary Background 1. We have recently completed a two-stage consultation process to establish a revised approach to funding teaching and allocating student numbers in higher education (HE) in England for (the first stage) and from and beyond (the second stage). It has been necessary to change the way we fund teaching in HE following the Government's revised student fees policy as set out in the White Paper 'Higher education: students at the heart of the system' 1 2. For , new-regime students attract HEFCE funding at a rate based on the funding for their price group in , less the funding for price group C in that year. New-regime students studying in price groups C and D do not attract HEFCE funding in through the main subject-based element of teaching grant. 3. For and beyond HEFCE funding will meet some of the additional costs of teaching new-regime students in high-cost subjects. We identify high-cost subjects as those such as medicine, science, engineering and agriculture which have high costs that cannot be recovered through income from graduate contributions alone. 4. We have gathered evidence on the on the costs of teaching different subjects using Transparent Approach to Costing for Teaching (TRAC(T)) data. TRAC is an activity-costing system which draws on expenditure information in higher education institutions accounts to derive the costs of teaching, research and other activity, and TRAC(T) is the national framework for costing teaching in different subjects. Following a pilot of TRAC(T) in 27, data have been collected annually from all English HEIs, reporting costs for 27-8, 28-9 and TRAC(T) data is collected at the level of academic cost centre. Historically, we have grouped cost centres into four price groups, which have then attracted different rates of funding. This paper provides a detailed analysis of the TRAC(T) data. Methodology 6. The data underlying this analysis are drawn from the 27-8, 28-9 and 29-1 HESA student records, and the 27-8, 28-9 and 29-1 TRAC(T) returns. Combining three years data draws on as much information as possible, and gives us confidence that no single year s return is atypical. The data were collated as follows: a. Total 27-8 HEFCE-fundable student FTEs by cost centre were derived from the 27-8 HESA individualised student records, and the total 27-8 subject-related costs of HEFCE-fundable provision were taken from the 27-8 TRAC(T) return

2 b. Total 28-9 HEFCE-fundable student FTEs by cost centre were derived from the 28-9 HESA individualised student records, and the total 28-9 subject-related costs of HEFCE-fundable provision were taken from the 28-9 TRAC(T) return. c. Total 29-1 HEFCE-fundable student FTEs by cost centre were derived from the 29-1 HESA individualised student records, and the total 29-1 subject-related costs of HEFCE-fundable provision were taken from the 29-1 TRAC(T) return. d. The data have been cleaned by removing outliers and institutions which indicated that they did not meet the minimum requirements in the completion of the TRAC(T) return. Any observations of institution and cost centre pairings (for the 27-8, 28-9 and 29-1 data) with fewer than 2 FTEs, or with costs per FTE (subject-facts) of less than 1,5 or more than 2, (except in the medical cost centres, where no upper limit was set), have been removed. A total of 5,56 records (institution and cost centre pairing) are included in the combined 27-8, 28-9 and 29-1 analysis. e. In consideration of the time lag between the years that the data refers to and the year the new teaching funding method will be applied to, costs have been inflated to prices using the latest GDP deflators 2 issued by Treasury at the end of December 211 ( 7. The plots in Annex A show subject-related costs against HEFCE-fundable student FTEs for each cost centre, for the 27-8, 28-9 and 29-1 data. The data have been presented in this way to provide the maximum amount of information as clearly and succinctly as possible. The plotted line is the least squares regression of costs against student FTEs, and the coefficient therefore represents the least squares estimate of the costs per student FTE. Also shown for comparison between cost centres are the mean (total subjectrelated costs divided by total ), and both the median and standard deviation of the individual rates associated with each observation. Similar plots and analysis conducted on prior years data are available on the HEFCE web site ( under Learning and Teaching/Funding/Review of price groups and fee assumptions); it should be noted that in previous analyses costs had not been inflated. 8. In order to draw comparisons of costs in different costs centres we have looked at the average (mean) subject-related cost per HEFCE-fundable FTE. Annex B shows the average cost per FTE, together with the number of HEFCE-fundable student FTEs and the total subject-related costs of HEFCE-fundable provision of each cost centre. Annex C shows the calculation of the average cost per FTE for the proposed price groups A, B and C1. 2 GDP deflators have been converted from a financial year basis to an academic year basis. 2

3 Cost ( s) Cost ( s) Cost ( s) Cost ( s) Annex A Plots of subject-related costs against HEFCE-fundable FTE for each cost centre (27-8, 28-9 and 29-1 data combined) Explanatory notes Each plot is a scatter graph of subject-related costs against HEFCE-fundable FTE, taken from the 27-8, 28-9 and 29-1 TRAC(T) returns and HESA student records. The data are displayed in this way to show both the correlation and the spread in any particular cost centre. Also shown is the line of best fit and the equation of that line. Although the gradients of the lines appear similar, in fact they vary considerably since the scales are different for each plot. The coefficient of the equation represents the average based on the least squares regression. Also shown for comparison between cost centres are the mean (total subject-related costs divided by total HEFCE-fundable FTE), and both the median and standard deviation (stdv) of the individual rates associated with each observation. 1 Clinical Medicine 2 Clinical Dentistry 4, 2, 35, 18, 3, 25, 2, 15, 1, 5, y = 14.38x 16, 14, 12, 1, 8, 6, 4, 2, y = 16.84x 5 1, 1,5 2, 2, , mean = median = stdv = 4.64 mean =16.46 median = stdv = Veterinary Science 4 Anatomy and Physiology 3, 12, 25, 1, 2, 15, y = 19.79x 8, 6, y = 8.8x 1, 4, 5, 2, , 1,2 1, , 1,2 mean = median = stdv = 8.8 mean =9.13 median = 9.3 stdv =

4 Cost ( s) Cost ( s) Cost ( s) Cost ( s) Cost ( s) Cost ( s) 5 Nursing and Paramedical Studies 6 Health and Community Studies 1, 12, 9, 8, 7, 6, y = 6.54x 1, 8, y = 7.2x 5, 6, 4, 3, 4, 2, 1, 2, , 1,2 1,4 1, , 1,2 1,4 1,6 mean = 7.6 median = 7.45 stdv = 2.11 mean = 7.2 median = 7.29 stdv = Psychology and Behavioural sciences 8 Pharmacy and Pharmacology 9, 12, 8, 7, 1, 6, 5, 4, 3, y = 7.24x 8, 6, 4, y = 9.67x 2, 1, 2, , , mean = 7.25 median = 6.99 stdv = 1.74 mean = 9.74 median = 9.76 stdv = Biosciences 11 Chemistry 2, 9, 18, 8, 16, 14, 12, 1, 8, 6, 4, y = 9.17x 7, 6, 5, 4, 3, 2, y = 9.74x 2, 1, 5 1, 1,5 2, mean = 9.19 median = 8.86 stdv = 2.13 mean = 9.84 median = 9.81 stdv =

5 Cost ( s) Cost ( s) Cost ( s) Cost ( s) Cost ( s) Cost ( s) 12 Physics 13 Agriculture and Forestry 9, 14, 8, 7, 6, 5, y = 1.5x 12, 1, 8, y = 9.38x 4, 3, 2, 6, 4, 1, 2, , 1,2 1,4 mean = 1.62 median = 1.36 stdv = 2.16 mean = 8.78 median = 7.5 stdv = Earth, Marine and Environmental Sciences 16 General Engineering 16, 16, 14, 14, 12, 1, 8, y = 9.9x 12, 1, 8, y = 9.6x 6, 6, 4, 4, 2, 2, , 1,2 1, , 1,2 1,4 1,6 mean = 1.14 median = 9.7 stdv = 3.4 mean = 1.1 median = 1.7 stdv = Chemical Engineering 18 Mineral, Metallurgy and Materials Engineering 4, 8, 3,5 7, 3, 2,5 2, y = 9.55x 6, 5, 4, y = 1.85x 1,5 3, 1, 2, 5 1, mean = 9.69 median = 9.63 stdv =2.59 mean = 1.82 median = 1.22 stdv =

6 Cost ( s) Cost ( s) Cost ( s) Cost ( s) Cost ( s) Cost ( s) 19 Civil Engineering 2 Electrical, Electronic and Computer Engineering 1, 14, 9, 8, 7, 6, 5, 4, y = 8.74x 12, 1, 8, 6, y = 9.8x 3, 2, 1, 4, 2, , 1, , 1,2 1,4 mean = 8.91 median = 8.99 stdv = 2.33 mean = 9.93 median = 1.3 stdv = Mechanical, Aero and Production Engineering 23 Architecture, Built Environment and Planning 14, 18, 12, 16, 1, 8, y = 9.98x 14, 12, 1, y = 6.89x 6, 8, 4, 2, 6, 4, 2, , 1,2 1,4 5 1, 1,5 2, 2,5 mean = 9.94 median = 9.75 stdv = 2.4 mean = 7.11 median = 7.7 stdv = Mathematics 25 Information Technology, Systems Sciences and Computer Software Engineering 8, 18, 7, 6, 5, 4, 3, y = 6.86x 16, 14, 12, 1, 8, 6, y = 8.26x 2, 4, 1, 2, , 1,2 5 1, 1,5 2, mean = 7.6 median = 7.15 stdv = 2.3 mean = 8.56 median = 8.6 stdv =

7 Cost ( s) Cost ( s) Cost ( s) Cost ( s) Cost ( s) Cost ( s) 26 Catering and Hospitality Management 27 Business and Management Studies 18, 3, 16, 14, 25, 12, 1, 8, y = 7.25x 2, 15, y = 6.64x 6, 1, 4, 2, 5, , 1,2 1,4 5 1, 1,5 2, 2,5 3, 3,5 4, mean = 7.35 median = 6.76 stdv = 2.4 mean = 6.72 median = 6.45 stdv = Geography 29 Social Studies 6, 25, 5, 4, 3, y = 7.2x 2, 15, y = 6.25x 2, 1, 1, 5, , 1,5 2, 2,5 3, 3,5 mean = 7.38 median = 7.2 stdv = 1.81 mean = 6.28 median = 6.24 stdv = Media Studies 31 Humanities 16, 14, y = 7.66x 45, 4, 12, 35, 1, 8, 6, 3, 25, 2, 15, y = 6.42x 4, 1, 2, 5, 5 1, 1,5 2, 2,5 5 1, 1,5 2, 2,5 3, 3,5 4, mean = 7.69 median = 7.41 stdv = 2.5 mean = 6.4 median = 6.35 stdv =

8 Cost ( s) Cost ( s) Cost ( s) Cost ( s) Cost ( s) Cost ( s) 33 Design and Creative Arts 34 Education 8, 14, 7, 6, 12, 1, y = 6.23x 5, 4, 3, y = 8.28x 8, 6, 2, 4, 1, 2, 2, 4, 6, 8, 1, 5 1, 1,5 2, 2,5 mean = 8.38 median = 8.13 stdv = 2.53 mean = 6.67 median = 6.66 stdv = Modern Languages 37 Archaeology 14, 3, 12, 2,5 1, 8, 6, y = 7.21x 2, 1,5 y = 8.38x 4, 1, 2, , 1,2 1,4 1, mean = 7.25 median = 6.88 stdv = 2.3 mean = 8.57 median = 8.52 stdv = Sports Science and Leisure Studies 41 Continuing Education 14, 12, 12, 1, 1, 8, 6, y = 6.61x 8, 6, y = 5.11x 4, 4, 2, 2, 5 1, 1,5 2, 5 1, 1,5 2, 2,5 mean = 6.79 median = 6.84 stdv = 1.93 mean = 6.7 median = 6.35 stdv = 3.4 8

9 Annex B Average subject-related costs per HEFCE-fundable FTE (27-8, 28-9 and 29-1 data combined) Current Price group Proposed Price group Cost centre Total subjectrelated costs of HEFCEfundable provision ( s) HEFCEfundable FTEs Average cost per FTE ( s) A/B A/B 3 Veterinary science 18,355 9, A/B A/B 2 Clinical dentistry 29,646 12, A/B A/B 1 Clinical medicine 1,244,861 83, B B 18 Mineral, metallurgy & materials engineering 92,749 8, B B 12 Physics 293,5 27, B B 14 Earth, marine & environmental sciences 354,2 34, B B 16 General engineering 343,689 34, B B 21 Mechanical, aero & production engineering 48,487 48, B B 2 Electrical, electronic & computer engineering 413,47 41, B B 11 Chemistry 361,928 36, B B 8 Pharmacy & pharmacology 278,547 28, B B 17 Chemical engineering 72,965 7, B B 1 Biosciences 1,132,33 123, B B 4 Anatomy & physiology 174,496 19, B B 19 Civil engineering 24,16 26, B B 13 Agriculture & forestry 12,374 13, C C1 37 Archaeology 75,968 8, Information technology & systems sciences, C C1 25 computer software engineering 968, , C C1 33 Design & creative arts 2,256,8 269, B/C/D C1 3 Media Studies 3 71,92 92, C C2 28 Geography 286,341 38, C C2 26 Catering & hospitality management 155,986 21, C C2 35 Modern languages 531,382 73, C C2 7 Psychology & behavioural sciences 734,57 11, C C2 6 Health & community studies 562,583 78, C C2 23 Architecture, built environment & planning 632,645 89, C C2 5 Nursing & paramedical studies 236,589 33, C C2 24 Mathematics 43,12 6, B/C/D C2 38 Sports science and leisure studies 2 466,832 68, D D 27 Business & management studies 2,69,484 37, C/D C2/D 34 Education 2 77,5 115, D D 31 Humanities & language-based studies 1,523,91 237, D D 29 Social studies 1,95,254 31, D D 41 Continuing education 85,389 14, Not included in the price group calculations 9

10 Annex C Average subject-related costs per FTE for price groups A, B and C1 The average costs for the proposed new price groups A, B and C1 are calculated as the weighted averages of the costs for each constituent cost centre. Proposed price group Total subjectrelated costs of HEFCE-fundable provision ( s) HEFCEfundable FTEs Average cost per FTE ( s) B 4,358,58 451, C1 3,31, , We are unable to use the TRAC(T) data to identify the costs of price group A provision in cost centres 1,2 and 3. We have therefore estimated the costs for price group A. To do this we have assumed that 67 per cent of students in cost centres 1 to 3 are assigned to price group A, with the balance to price group B. We have then deducted costs, reflecting the calculated average price group B costs for the 33 per cent of students assigned to price group B. This calculation is shown in the table below. Calculation of average costs per FTE for price group for A students Total number of students (FTEs) in cost centres 1, 2 and 3 (a) 15,252 Proportion of students in Cost Centres 1 to 3 who are in price group A (b).67 Student numbers (FTEs) in Cost Centres 1 to 3 who are clinical price group A (c)=(a)*(b) 7,519 Student numbers (FTEs) in Cost Centres 1 to 3 who are pre-clinical price group B 34,733 (d)=(a)-(c) Total costs ( s) for cost centres 1, 2 and 3 (e) 1,634,862 Estimated costs ( s) of cost centres1 to 3, price group B students (using average cost per FTE of the rest of price group B) (f)=9.66*(d) 335,454 Estimated costs ( s) of cost centres1 to 3, price group A students (g)=(e)-(f) 1,299,47 Estimated costs per FTE ( s) of price group A students (h)=(g)/(c) Summary of average costs per FTE for price groups A, B and C1 Proposed price group Average cost per FTE ( s) A B 9.66 C