HEFCE Report 00/10

ISR 1998-99 derived statistics for funding allocations and monitoring in FECs

Respond by 31 March 2000

Executive summary

Purpose

1. This document describes how we compare data returned in the Higher Education in Further Education: Students Statistics survey 1998 (HEIFES98) with data supplied on the Further Education Funding Council’s July 1998-99 Individualised Student Record (ISR). The technical supplement describes the methods used, and gives details of their weaknesses.
2. We are comparing the two data sources to check data returned on HEIFES98, including forecast registrations and non-completions, as this information formed the basis for funding allocations in 1999-2000. We believe this exercise will help to improve the quality both of data supplied to the ISR, and of responses to the HEIFES survey.
3. The reconstruction of HEIFES98 from the ISR may also provide the basis for calculating premiums based on the following:
4. • full-time mature undergraduate students
   • students receiving some tuition-fee remittance
   • the neighbourhood types of students.
5. The mature premium will be used in the funding allocations for 2000-01. Both the neighbourhood types and the tuition-fee data are provided because it has not yet been decided which will be used for the premium for students with a disadvantaged background. The tuition-fee data are not yet of adequate coverage for funding, but may be used in future. The FEFC ISR 2000-01 may contain an additional field for HE students which will allow us to calculate our disability premium directly from the data set. We await the outcome of the FEFC consultation document (FEFC 99/51) which will determine whether or not such a field is added. At present we shall base the further education colleges’ (FECs) disability premium on the FEFC allocation of funds for additional support.

6. Key points

7. Where we identify significant differences in the data, we shall write to the institutions concerned and ask for a response.
8. If the data remain unreconciled, we may audit both sets of data to arrive at agreed figures. Any amendments to data following this exercise may result in retrospective adjustments to funding.
9. All institutions are invited to comment on the methods described in the technical supplement, and to contact us if they are concerned about the suitability for publication of the franchised-out data.

10. Action required

11. Where we require a response it should be sent to Anthony Ryan no later than 31 March 2000.

12. Selection of institutions asked to respond

13. As this is the first year of this comparison, only a few institutions will be asked to respond. Where there was doubt about whether or not a student would be included in the HEIFES98 population, or which category he should be in, the methods used in the comparison were such that as few colleges as possible would meet the selection criteria given below. We shall ask for a response from institutions if either of the following are true:

14. a. The total difference in the standard resource calculated using HEIFES98 and FEFC ISR14 (July 1998-99) data exceeds £1,000,000.

    b. The difference in the standard resource is more than 70 per cent.

15. Standard resource was chosen as the basis of selection because:

16. a. It is calculated at the highest level of aggregation used in operating our funding process.

    b. It shows major changes in resource while remaining neutral to variations in other institutions' data.

17. We shall conduct similar exercises in future years. We hope that as institutions improve the quality of their data we may be able to add a criterion based on forecast numbers.
18. An exercise for higher education institutions (HEIs), using the HESA student record and the Higher Education Students Early Statistics Survey (HESES), has been operating for several years and has led to improvements both in the matching of the two data sets and in their quality. Holdback was used to set the criteria for HEIs and may be used for further education colleges in future. We could not use it this year because it is based on funding contracts between HEFCE and the colleges, few of which were in place in 1998-99.
19. Our data set was extracted in mid-January 2000. We shall write to those colleges who did not submit a valid ISR14 return before the extraction, setting out the actions we intend to take to ensure adequate data on which to base our funding. Some colleges have re-submitted their ISR14 return since the extraction; we shall send them re-calculated tables as soon as possible. However, we cannot change the 31 March 2000 deadline for responses.

20. Responses required

21. We shall write separately to the institutions from which we require a response.
22. In such cases and in the first instance, we would like explanations of differences in the following three areas:

23. a. Columns 4 and 4a of the main student tables, 1a, 2a and 3a.

    b. The proportions of students across fee levels and fee sources, Tables 1b, 2b and 3b.

    c. Part-time load factors. These are calculated from Table 3a.

24. Responses should contain both estimates of the number of students involved and reasons for the differences.
25. If differences between the data cannot be reconciled, we may audit both sets of data. If the data are amended, we may make retrospective adjustments to funding.
26. All institutions are invited to comment on the methods described in the technical supplement, and to suggest how they can be improved.
27. We believe this exercise helps to improve the quality both of data supplied to the FEFC ISR, and of responses to the HEIFES survey.

28. Information supplied

29. When we received signed-off ISR14 data from the FEFC we performed the comparison. We are now writing to institutions indicating whether a response is required, and enclosing the following information:

30. a. A copy of the HEIFES98 return, including all amendments made during the funding process.

    b. HEIFES98 as re-created using data supplied to the FEFC ISR14, and the coding methods described in the technical supplement.

    c. The number of mature home and EC fundable full-time undergraduates.

    d. The number of young, home and EC fundable full-time undergraduates receiving some tuition-fee remittance.

    e. The numbers of young, English domiciled, home and EC fundable, full-time undergraduates in each of 160 neighbourhood types.

    f. The institutions teaching indirectly funded students and the number of such students.

    g. Summary information comparing HEIFES98 to ISR14 data. This includes a table showing the quality of the fields that are compulsory for HE students only. The quality of these HE specific fields is given in terms of the proportion of HEIFES students for which these fields are filled, and the extent of mismatching between the HE specific fields and the rest of the ISR14.

    h. A summary of the FEFC student records excluded from the HEIFES re-creation.

    i. A list of the qualification titles that are not recognised as HE on the qualifications database but which the college indicated to be HE. We shall include the number of students with each qualification title.

31. Table 4 of HEIFES, Island students, has not been included as it was not used in funding and there is no intention to publish these data.
32. All the above data can be supplied electronically on request. An individualised record is also available on request. It contains the following additional fields:

33. a. A flag showing whether or not the student is counted in HEIFES98 and, if not, why the student was excluded.

    b. The HEIFES classification of the student.

    c. The classification of the student’s neighbourhood type used to calculate the geodemographic premium. This field is completed for a restricted group of students only.

34. This information is available to all institutions, but we shall give priority to institutions from which we are expecting a response. To obtain any of the above, please contact:

35. Anthony Ryan
    tel 0117 931 7297
    e-mail isr_heifes_stats@hefce.ac.uk.

36. After we have given the HEIs an opportunity to check their data, we plan to publish information on the number of students taught at FECs but funded indirectly through the HEIs. A list of institutions that are teaching indirectly funded students registered at each HEI will be published.

37. Deadline for responses

38. Responses should arrive no later than 31 March 2000 and should be sent to:

39. Anthony Ryan
    Analytical Services Group
    HEFCE
    Northavon House
    Coldharbour Lane
    Bristol BS16 1QD

Technical supplement

1. Purpose

2. This supplement to HEFCE 00/10 describes the methods used to compare HEIFES and FEFC ISR data. It also gives details of known discrepancies between the data sources.
3. This supplement is aimed at expert readers with in-depth knowledge of the data. Readers are advised to have a copy of the 1998-99 ISR institution support manual and HEIFES98 (HEFCE 98/53) to hand when using this document. We also used version 12.3 of the qualifications database whose guidance notes are provided with the database software.

4. Fields used in comparisons

5. Only certain fields, detailed below, were used to generate the comparison between the HEIFES and ISR14 data. Throughout this supplement, fields taken from ISR14 are shown using the numbers given below. Fields from the student data set part of ISR14 take the prefix ST_; those from the qualification aim data set have the prefix QA_; and those from the qualifications database use the same names as described in the guidance notes.

6. Field	Description	HEFCE Name
   AWARD_BOD1	Primary awarding body	AWARD_BO
   AWARD_BOD2	Secondary awarding body	AWARD_B2
   ENG_LEVEL	Level of study (FE or HE) England specific	ENG_LEVE
   NAT_NVQ_LV	NVQ level	NVQ_LV
   QUAL_TITLE	Qualification title	QUAL_TIT
   QUAL_TYPE	Qualification type	QUAL_TYP
   SUPERCLASS SUERCLAS2 SUPERCLAS3	A classification system to enable users with no IT skills to retrieve qualifications accurately	SUPERCL1 to SUPERCL3
   	Full or reduced record – derived by the FEFC	ST_FULLR
   Q01	Row number relating to each student – derived from Q01	QA_SEQNO
   Q02	Qualification aim reference	QA_AIM_R
   Q07A	Annual fees indicator	QA_TUIT_
   Q07B	Amount of tuition fees received or expected from the student	QA_TU34
   Q08	Reason for partial or full non-payment of tuition fees	QA_NON_P
   Q09	Major source of tuition fees	QA_TU13
   Q10	FEFC or HEFCE funding	QA_FEHE_
   Q11	Major source of funding other than tuition fees or FEFC/HEFCE funding	QA_OTHER
   Q13	Outward collaboration provision arrangements	QA_FRANC
   Q15	Guided learning hours	QA_GUIDE
   Q16	Start date	QA_ST_DA
   Q17	Expected end date	QA_EQA_EXP_E
   Q18	Actual end date	QA_EN_DA
   Q19	Completion status	QA_COMP_
   QHE01	Major source of tuition fees – HE students	QA_FEE_S
   QHE02	Year of programme	QA_PROGY
   S01	FEFC institution code – derived from S01	ST_INST
   S01	Student reference	ST_REF
   S04	Date of birth	ST_DOB
   S06	Home postcode	ST_POSTC
   S07	Country of domicile	ST_DOMIC
   S14A	Annual fees indicator	ST_TUIT_
   S14B	Amount of tuition fees received or expected from the student	ST_TU20
   S15	Reason for partial or full non-payment of tuition fees	ST_NON_P
   S16	Major source of tuition fees	ST_FEE_S
   SHE01	Highest qualification on entry	ST_QUAL_
   SHE05	New entrant to HE	ST_NEWHE
   SHE09	Type of programme year	ST_PYTYP
   SHE10	Mode applicable to HEIFES/HESES	ST_MHESE
   SHE11	Level applicable to HEIFES/HESES	ST_LHESE
   SHE12	Completion of year of study	ST_COMPY
   SHE13	Student FTE	ST_FTE

7. The date of birth (S04) and start date (Q16) were used to calculate the number of mature home and EC fundable full-time undergraduates. The home postcode (S06) field was used to determine the neighbourhood type of full-time and sandwich, home and EC fundable undergraduates in the HEIFES Column 4 population, who were young on entry and had an English postcode. The QHE01 field was used to determine those students who received no award or financial backing.

8. Differences between HEIFES and FEFC ISR data

9. Some data returned in HEIFES cannot be re-created exactly using data supplied to the FEFC ISR because of differences in definition. In such cases, reasonable approximations have been made. Listed below are areas where there may be some uncertainty about the correspondence of FEFC ISR14 records to HEIFES cells. Where possible, we have indicated the likely effects of the uncertainties.
10. In some cases coding assumptions may have caused records to be wrongly rejected from the comparison, or wrongly assigned to a HEIFES cell within the comparison. If institutions believe this may have been the case, they can obtain full details of the assignment of each record, including rejected records and reasons for the rejection, from Anthony Ryan at the HEFCE. Individual records can be identified using the following fields:
11. • the student data set reference (S01)
    • the qualification aim data set reference (Q01)
    • the qualification reference code (Q02).
12. The differences are grouped as follows:

13. a. Those that could affect the selection criteria, that is, affect Column 4 of the main tables.

    b. Those that do not affect Column 4 of the main tables but could lead to differences in the other columns.

    c. Areas of uncertainty in completing HEIFES98.

    Differences critical to selection criteria

14. The details listed below are similar to those in HEIFES98 Annex M, but where the coverage and quality of the HE specific fields was poor we had to use other fields in the re-creation. In general we excluded students where the HE fields had a specific exclusion code. We then classified the remaining students according to both the HE specific field and the non-HE specific fields. We used the HE specific field provided it was not contradicted by the other fields in ISR14.

15. Students excluded from HEIFES

16. The following table gives details of students excluded from HEIFES98 and the exclusion code used (HEFEXCL)

17. Reason for exclusion	Exclusion codes
    Specifically excluded by HE level field	SHE11=99. HEFEXCL=2
    Qualifications database conflicts with HE level field which was not coded as ‘99’	HEFEXCL=4
    Classed as FE or non-recognised HE by the qualifications database/insufficient information given on the qualifications database	HEFEXCL=8
    First part of a non-standard academic year	HEFEXCL=16
    Specifically excluded by HE mode field	SHE10=99. HEFEXCL=32
    Specifically excluded by HE completion field	SHE12=9. HEFEXCL=64
    Less than 3 per cent FTE	HEFEXCL=128
    Unable to assign eligibility for core funding and residential status	S07 not completed. HEFEXCL=256
    Unable to assign price group because of insufficient superclass information	HEFEXCL=512
    Not active in academic year. Started after 31 July 1999 or left before 1 August 1998.	HEFEXCL=1024
    Classified by non-HE fields as non-completion and left before 1 November 1998	HEFEXCL=2048
    Incoming exchange students (only ERASMUS identifiable)	Q11=017 or Q12=017. HEFEXCL=4096

    Students on courses not leading to a recognised HE qualification

18. Students are included in HEIFES if their qualification aim is a recognised HE qualification. If the HE level field (SHE11) was completed and a student was classified as not in the HEIFES population, the student was excluded. The qualifications database was used to decide whether a qualification was a recognised HE one or not. Where the qualifications database gave inadequate information, or where it showed the student was not on a recognised HE qualification in contradiction to the SHE11 field; the student was excluded.

19. Students with very low FTEs

20. Students whose FTE for the year of programme of study is less than 3 per cent are excluded from HEIFES. The HE FTE field (SHE13) was used to calculate the FTE of students on some higher degrees. However, generally this field was inadequately completed and we decided to base the FTE calculations, where possible, on the qualifications database, the rest of ISR14 and the table in Annex E of HEIFES98. Expected course lengths were calculated from the aim reference, start date and expected end date in the qualifications section (Q02, Q16 and Q17), and the qualification type and awarding body (QUAL_TYPE and AWARD_BOD) in the qualification database. The start and expected end dates were also used to work out whether a student was on a standard academic year or not, and if not, whether they were commencing, midway through or finishing their course. The FTEs, the year of programme of study and the expected length of study within that year were then calculated, where necessary, from these derived variables.

21. No countable years of study

22. A student who had no FTE for the year of programme of study recorded within the academic year was excluded from HEIFES. In the FEFC ISR the FTE for a year of programme of study is reported in the academic year in which the student finishes the year of programme of study. Therefore a student deemed to be in the first year of a non-standard academic year was excluded from the comparison.

23. Differences not affecting selection criteria

    Level of study

24. The HE level field (SHE11) was used to determine the level of study if it was completed and was not contradicted by information of the qualifications database. In other cases the level of study was calculated from the qualifications database using the type of qualification (qual_type) and the awarding body (award_bod) fields.

25. Long courses

26. Only those courses that were specified as ‘long’ in the HE level field (SHE11) were counted as long courses, provided there was consistency between the HE level field and the level determined by the qualifications database.

27. Mode of study

28. The definition of full-time given in HEIFES98 is based on students attending an institution for at least 24 weeks and during that time undertaking at least 21 hours of study per week. This is not equivalent to the guided learning hours used by the FEFC. Those students that were identified in the HE mode field (SHE10) as not being in the HEIFES population were excluded.
29. If the SHE10 field was not completed, or was contradicted by the non-HE specific fields, then we estimated the number of weeks a student attended from the level of study, the expected length of study and the year of programme of study. Similarly the number of hours of study per week was estimated from the guided hours field (Q15), the expected length of study and the year of programme of study. These estimates were based on general assumptions, the purpose of which was to verify that students with a resultant mode of full-time were likely to meet the HEIFES full-time criteria. It was not possible to identify sandwich year-out students unless the HE mode field (SHE10) was completed, nor was it possible to identify undergraduate students on international courses where study at the institution lasted less than 10 weeks.
30. All students were assumed to be part-time unless they were specified as ‘full-time and sandwich’ in the HE mode field (SHE10), and this was not contradicted by the non-HE specific fields, or they had all of the following characteristics:

31. a. At least 24 weeks attendance in the year of programme of study.

    b. Average guided learning hours of 16 or more per week. The 16 guided hours was assumed to equate to the 21 hours criterion once self-study was included.

    c. Their normal fees for the year were greater than or equal to the minimum requirement or their fees were waived for any of the following reasons:

    i. in receipt of state benefit

    ii. asylum seeker receiving state benefit

    iii. in receipt of job seekers allowance

    iv. unwaged dependent of any of the above

    v. a reason consistent with local college policy

    vi. other funding, for example modern apprenticeship.

    We hope to collect standard fees information on HE students in the 2001-2002 ISR data set.

32. For this comparison it was assumed that, for courses lasting less than one year, the duration of the course and the attendance time are equivalent. This could cause some students to be coded as full-time when they should have been coded as part-time.

33. Eligibility for core funding and residential status

34. Those students whose country of domicile (S07) was within the European Community were classed as home and EC students. Of these, those who received FEFC or HEFCE funding (Q10) were classed as home funded and those that were ‘eligible for HEFCE funding but funding not claimed’ were classed as home independently funded. All other home and EC students were classed as home non-fundable. Students whose country of domicile was outside the EC were classed as island and overseas students.

35. Indirectly funded students

36. Students are only recorded as indirectly funded (franchised out) in the ISR if more than 50 per cent of their qualification is delivered by another institution. In order to convert this to the current academic year we assumed that the student was franchised out for at least part of each academic year. We were unable to identify students who were wholly franchised out, it is hoped that this will be remedied in the 2001-2002 data set when the proportion of the year of programme of study for which a student is franchised out may be collected for all HE students.

37. Student load

38. Student load recorded on HEIFES98 should relate to the years of programme of study being counted. This may not be consistent with the FTE calculated from the FEFC ISR, particularly where we have had to rely on the non-HE specific fields. An outline of the FTE calculations is given in paragraph 11 above.

39. New entrants

40. Figures shown in Column 5 of the re-creation may not accurately reflect the numbers of new entrants returned on HEIFES. The following students were classed as ‘not new entrants’:

41. a. Those whose HE new entrant field (SHE05) was completed and coded as ‘in their first year on this qualification aim and previously attended an HE course in the UK’ or ‘not in the first year of this qualification aim’.

    b. Those not in the first year of their programme of study. This may be a variable derived by the comparison programme (see paragraph 11) or the HE specific field (QHE02).

    Fee eligibility

42. Students were classified as new or existing students on the basis of their year of programme of study, and their highest qualification on entry. Where possible the source of tuition fees was calculated from the HE tuition fee field (QHE01). If that was not completed the non-HE specific fields (S16, Q09 and S07) were used.

43. Allocation to price groups

44. HEIFES98 states that price groups are defined in terms of Superclass II, but that where this is not appropriate students should be allocated to the most appropriate price group, provided that the student moves into a price group attracting lower funding. In the comparison programme we therefore considered all three levels of superclass, mapped each of these onto price groups and took the lowest price group of the three. Where students had not been allocated to any level of superclass we were unable to assign a price group and the student was excluded from the re-creation. This may lead to a substantial underestimate of the recreated HEIFES98 population. The lack of superclass information is due to the inadequate coverage of the qualifications database. As this is the first year of this comparison the thresholds for eliciting a response are high. However, we hope that in future the coverage of the qualifications database will improve.
45. Some qualifications contain study in two distinct subject areas, and in HEIFES the student FTE should have been split between the relevant price groups. We were unable to do this in the re-creation where whole students were allocated to price groups. It may be necessary to identify the proportion of study in each superclass group in future ISRs.
46. Students on qualified teacher status (QTS) courses were identified as those whose major source of funding other than tuition fees and FEFC/HEFCE funding (Q11) was coded as ’Other HM government departments and public bodies including EC’. We were unable to identify students on initial teacher training (ITT) and so all QTS students were allocated to the INSET(QTS) price group. As a result, the numbers in the INSET (QTS) price group may be inflated. We hope to be able to use the qualifications database to improve the identification of ITT and INSET (QTS) courses in future comparisons.
47. The mapping of superclass codes to price groups is given in the table below.

48. Areas of uncertainty in completing HEIFES98

    Forecasts of countable years of programme of study and non-completions

49. HEIFES98 required institutions to provide forecasts of countable years of programme of study between 2 November 1998 and 31 July 1999. Forecasts are by their nature inexact, and institutions with a high proportion of students following non-standard academic years may not be able to predict countable years of programme of study accurately.
50. Forecasts of non-completions are also inexact. However, it is expected that these figures will, when considered as a whole, be an accurate reflection of actual non-completions.

51. Price group	Superclass II
    B (Laboratory-based science, engineering and technology)	PB, PE, RA, RC – RF, RH, SA, SB, SK, TL, TM, XH – XL, XP – XR, XT, YC – YE
    C (Other high cost subjects with a studio, laboratory or fieldwork element)	CA – CH, CY, DC, FK (excluding FK.343, FK.335, FK.349), J, L, M, NA – NH, PA, PC, PD, PF – PQ, RB, RG, SC – SJ, SL – SP, TA – TK, W, XA – XF, XM, XN, XS, YA, YB
    D (All other subjects)	A, B, CX, CZ, DA, DB, DD, DE, E, FB, FC, FJ, FK.343, FK.335, FK.349, FL,G, H, KA – KD, KH, NK – NN, Q, V, Z
    Psychology	PK
    Media studies	KE – KG
    INSET(QTS)	Q10=9 and Q11=10

    Additional data derived from the FEFC ISR

    Count of mature entrants

52. Full-time, home and EC fundable undergraduates included in the HEIFES98 Column 4 population were counted as mature students if a comparison of start date (Q16) and date of birth (S04) revealed that they were aged 25 or older on entry.

53. Counts of young students in neighbourhood types

54. We enclose a table showing the number of HEIFES98 Column 4, young, English domiciled, home and EC fundable, full-time undergraduates in each of 160 neighbourhood types. An unclassified category is included because it was not always possible to determine the neighbourhood type. Students are ‘young’ if they are under 25 on entry. For a description of the allocation of postcodes (S06) to the neighbourhood type please see ‘The influence of neighbourhood type on participation in higher education’ (HEFCE, April 1997).

55. Count of young students receiving some tuition fee remittance

56. Those HEIFES98 Column 4, young, home and EC fundable, full-time undergraduates with some award or financial backing for tuition fees (QHE01 ¬= 01) were counted as students receiving some tuition-fee remittance.