Guidance onsubmissionsJune 2005

Ref RAE 03/2005

RAE 03/2005 1

RAE 2008 Guidance on submissions

To Heads of HEFCE-funded higher educationinstitutions

Heads of HEFCW-funded highereducation institutions

Heads of SHEFC-funded higher educationinstitutions

Heads of universities in Northern Ireland

Of interest to those responsible for Research assessment; Research policy;

Planning

Reference RAE 03/2005

Publication date June 2005

Enquiries to Ed Hughes, Davina Madden or Josie Lintontel 0117 931 7267e-mail info@rae.ac.uk

2 RAE 03/2005

Executive summary

Purpose

1. This document specifies the content and datarequirements for submissions in the 2008Research Assessment Exercise (RAE) and guideshigher education institutions on policy andpractical matters in preparing submissions.

Key points

2. In December 2006 the four UK highereducation funding bodies will invite all eligibleUK higher education institutions to makesubmissions in the 2008 RAE. Submissions willcontain a common set of data comprising:

• information on staff in post on the censusdate, 31 October 2007

• details of publications and other forms ofassessable output which they have producedduring the publication period (1 January 2001to 31 December 2007)

• data about research students and researchincome and a written commentary relating tothe assessment period (1 January 2001 to 31July 2007).

Submissions will be assessed in the course of2008. The results, which the higher educationfunding bodies intend to use in calculatingresearch grants from academic year 2009-10, willbe published in December 2008.

Action required

3. This document is for information and to guideinstitutions in collecting data for inclusion inRAE 2008 submissions. No action is required byhigher education institutions at this stage.

RAE 03/2005 3

Introduction4. This document sets out the administrativearrangements and data requirements forsubmissions to the 2008 Research AssessmentExercise (RAE). It should be read in conjunctionwith the statements of working methods andassessment criteria (panel criteria) of the 15 mainpanels and 67 sub-panels, which we will publishon the RAE web-site in draft form forconsultation in July 2005 and in final form earlyin 2006. This document and the panel criteriawill describe comprehensively the datarequirements and how panels will use the data intheir assessments. We may issue supplements tothis guidance at later dates to clarify points ofdetail on the data definitions that could assisthigher education institutions (HEIs) in preparingsubmissions, but such supplements will notrequest any new items of data.

5. In December 2006 we will formally inviteeligible HEIs to make submissions for assessment.A technical manual on how to make submissionswill accompany that invitation.

6. This document contains the following parts:

a. Part 1 Overview of the 2008 RAE

i. Purpose

ii. General principles

iii. Key changes since RAE 2001 ‘Guidanceon submissions’

iv. Assessment method, units of assessmentand assessment panels

v. Timing and publication of results and ofsubmissions

vi. Code of practice on preparingsubmissions and selecting staff

vii. Method of submission

viii. Data verification and access to researchoutputs

ix. Data protection

b. Part 2 Form and content of submissions

i. Multiple submissions

ii. Joint submissions

iii. Interdisciplinary research, arrangementsfor cross referral and specialist advice

iv. Content of submissions

c. Part 3 Data requirements and definitions

i. General definitions

ii. Section 1 Staff details (RA0 and RA1)

iii. Section 2 Research outputs (RA2)

iv. Section 3 Research students andstudentships (RA3a and RA3b)

v. Section 4 Research income (RA4)

vi. Section 5 Research environment andesteem (RA5a), individual staffcircumstances, (RA5b) and Category Cstaff circumstances (RA5c)

d. Annexes

Annex A Quality profiles and definitionsof quality levels

Annex B Definition of research for the RAE

Annex C Units of assessment

Annex D Timetable

Annex E Summary of data requirements(table)

Annex F Standard data analyses

Annex G Guidance from the EqualityChallenge Unit on drawing up acode of practice on preparingRAE submissions

Annex H Related and forthcomingpublications

Annex I Glossary of terms

Annex J List of abbreviations

7. Enquiries should be addressed to the RAEteam and should be routed wherever possiblethrough the individual HEI’s designated RAEcontact.

4 RAE 03/2005

Purpose8. The RAE is conducted jointly by the HigherEducation Funding Council for England(HEFCE), the Scottish Higher Education FundingCouncil (SHEFC), the Higher Education FundingCouncil for Wales (HEFCW) and the Departmentfor Employment and Learning, Northern Ireland(DEL). The RAE is managed by the RAE team,based at HEFCE, on behalf of the four UK highereducation funding bodies. In this document, ‘we’refers to the RAE team.

9. The primary purpose of RAE 2008, the fourthsuch national exercise since 1992, is to producequality profiles (see paragraph 30 and Annex A)for each submission of research activity made byinstitutions. The four higher education fundingbodies intend to use the quality profiles todetermine their grant for research to theinstitutions which they fund with effect from2009-10. Any HEI in the UK that is eligible toreceive research funding from one of these bodiesis eligible to participate.

10. The definition of research for RAE 2008 is atAnnex B. It is virtually the same as that used inRAE 2001: we have made minor amendments tothe phrasing of the definition used in 2001 thatclarify but do not change its meaning.

General principles11. The following principles govern the conductof the RAE. They set the framework in which theRAE team coordinates the exercise and in whichthe 15 main panels and 67 sub-panels will deploytheir collective professional judgement to draftcriteria for assessment and to assess submissions(see RAE 01/2005 ‘Guidance to panels’).

12. Equity. All types of research and all forms ofresearch output shall be assessed on a fair andequal basis. Panels have been instructed to adoptassessment processes and criteria that enable themto recognise and treat on an equal footingexcellence in research across the spectrum ofapplied, practice-based and basic/strategicresearch, wherever that research is conducted.

13. Diversity. Submissions to RAE 2008 willreflect the diversity of excellent research conducted

across the UK higher education sector. We aredeveloping processes for the 2008 RAE that aresufficiently flexible to encourage that diversity.Accordingly, we have instructed panels to defineappropriate criteria for identifying excellence indifferent forms of research endeavour, whileattaching no greater weight to one form overanother; and to make provision to recognise thediversity of evidence for excellent research.

14. Equality. HEIs are encouraged to submit thework of all their excellent researchers, includingthose whose volume of research output has beenlimited for reasons covered by equal opportunitiesguidelines. To comply with equal opportunitiesmonitoring legislation, we will require HEIs toconfirm that they have developed, adopted anddocumented an appropriate internal code ofpractice in preparing submissions and selectingstaff for inclusion in RAE submissions (seeparagraphs 35-38). We have already briefed panelson all current and pending equal opportunitieslegislation that will affect RAE 2008 (see RAE02/2005 ‘Equality briefing for panel chairs,members and secretaries’) and the panel criteriawill take account of this.

15. Expert review is central to the RAE.Widespread confidence in discipline-based expertreview founded upon academic judgements wasupheld in consultations following Sir GarethRoberts’ Research Assessment (RA) review. Tomaintain this confidence, we have appointedpanels of experts who are currently or haverecently been active in high quality research.While these experts will draw on appropriatequantitative indicators to support theirprofessional assessment of RAE submissions,expert review remains paramount.

16. Clarity. We aim to make all writtendocuments and statements about the RAE clearand consistent. This principle guides thepreparation of both written briefing documentsfrom the RAE team and of panel criteriastatements.

17. Consistency. Assessments made in the RAEshould be consistent across cognate areas.Following consultation on proposals for theconduct of RAE 2008, we have grouped under

Part 1: Overview of the 2008 Research Assessment Exercise

RAE 03/2005 5

the leadership of each of 15 main panels, a groupof broadly cognate sub-panels for units ofassessment that share broadly similar researchapproaches. Main panels are working with sub-panels to define and agree criteria and workingmethods that are appropriate to the unit ofassessment and provide for consistent approaches.Main panels will also ensure consistentapplication of the quality standards in the qualityprofiles awarded to submissions.

18. Continuity. The RAE has developed throughan evolutionary process, building on and learningfrom previous RAEs. With every successiveexercise a balance has to be struck betweencontinuity and development. Enhancements toRAE 2008 follow from extensive review andconsultation, and have been adopted where it isjudged they can bring demonstrable improvementswhich outweigh the cost of implementing them.

19. Credibility. As was demonstrated throughresponses to the RA review and the fundingbodies’ subsequent consultation on the conduct ofthe RAE, the fundamental methodology, formatand processes employed in the exercise – that is,discipline-based expert review founded uponacademic judgements – are credible to those beingassessed. We will maintain this credibility throughdue process, by upholding the integrity of theassessment process through careful guidance topanels and by assuring the integrity of data insubmissions through data verification.

20. Efficiency. The cost and burden of the RAEshould be the minimum possible to deliver arobust and defensible process. Previous RAEs havebeen highly cost-effective given the value of publicfunds distributed through their ratings (includingthe estimated cost to HEIs). For example, weestimated the costs of the 1996 RAE in Englandto be some 0.8 per cent of the value of publicresearch funding subsequently allocated withreference to its results. We will continue to weighthe burden on institutions against the need toensure accountability in disbursing public funds.

21. Neutrality. The RAE exists to measure thequality of research in HEIs. It should carry outthat function without distorting the activity thatit measures and it should not encourage ordiscourage any particular type of activity or

behaviour other than providing a general stimulusto the improvement of research quality overall.

22. Transparency. The credibility of the RAE isreinforced by transparency about the processthrough which decisions are made. This principleis central to the funding bodies’ work. It hasalready been applied throughout the RA reviewand the funding bodies’ subsequent consultation.Decisions and decision-making processes will beexplained openly.

Key changes since RAE 2001‘Guidance on submissions’ 23. The data requirements and definitions buildon those for the 2001 RAE but do differ, so HEIsshould not rely on their knowledge of the datarequirements in previous years. Amongst thechanges, we draw particular attention to:

a. The abolition of Category A* staff, whichwas introduced for the 2001 RAE.

b. The abolition of the distinction we made inprevious RAEs between general funds, NHSfunds and specific funds in reporting sourcesof salary for selected staff.

c. The withdrawal of the facility in the 2001RAE to submit, alongside the research outputof selected staff, research outputs producedby their supervised research students orresearch assistants.

d. The end of the publication period on 31December 2007, which is one month after thesubmission deadline (30 November 2007) andfive months after the end of the assessmentperiod (31 July 2007) to which the other dataabout research students, income, environmentand esteem must relate. This change recognisesthe fact that many research outputs that arepublished in print and other media bear onlythe year of publication and not the month.Therefore, it would not be possible to verifythat an item bearing a 2007 year ofpublication had in fact been published by anydate earlier in the year than 31 December.Paragraphs 44 and 45 explain thearrangements for HEIs to submit researchoutputs whose publication is pending between30 November and 31 December 2007.

6 RAE 03/2005

e. Our intention to collect certain categories ofresearch outputs in electronic formatalongside submissions (see paragraph 96).

f. The combining of forms RA5 and RA6 inthe 2001 RAE into one form, RA5a thatdescribes the research environment andevidence of esteem. The panels will specifythe format of RA5a in their criteriastatements (see paragraph 154 to 158).

Furthermore, we are no longer collectinginformation on staff who have not been selectedfor assessment for the RAE. However, for the2007-08 (and subsequent) Higher EducationStatistics Agency (HESA) individualised staffrecords, there will be a continuing requirementfor institutions to indicate whether each eligiblemember of staff was selected for assessment in the2008 RAE, and to assign non-selected staff tounits of assessment. This is necessary so that theHE funding bodies can fulfil their obligations forequal opportunities monitoring under the RaceRelations (Amendment) Act 2000 (see paragraphs35 to 38).

Assessment method, units ofassessment and assessment panels 24. Institutions will make submissions by 30 November 2007. Each submission willcontain, in summary:

• information on staff in post on the censusdate, 31 October 2007

• details of publications and other forms ofassessable output which they have producedduring the publication period (1 January2001 to 31 December 2007)

• data about research students and researchincome and a written commentary relating tothe assessment period (1 January 2001 to 31 July 2007).

25. Sixty-seven sub-panels will conduct a detailedassessment of submissions within 67 units ofassessment (UOAs). They will work under theguidance of 15 main panels. For each submissionassessed each sub-panel will provide a provisionalquality profile to the main panel for endorsement.

26. The UOAs are listed in Annex C. SomeUOAs have been reconfigured since the 2001RAE as described in RAE 03/2004 ‘Units ofassessment and recruitment of panel members’. Adescription of each UOA will be published in thepanel criteria statements. Each will indicate thekey subject areas the UOA covers, but will notgive an exhaustive account of the disciplines itencompasses. HEIs should refer to the UOAdescriptions when deciding in which UOAs tomake submissions.

27. As we indicated in RAE 01/2004 ‘Initialdecisions by the UK funding bodies’, we havetried to ensure that the membership of the mainand sub-panels comprises individuals who haveexperience in conducting, managing and assessinghigh quality research, as well as experts who arewell-equipped to participate in the assessment ofapplied and practice-based research from apractitioner, business or other user perspective.The process for appointing panel members isdescribed in RAE 03/2004 ‘Units of assessmentand recruitment of panel members’ and themembership of each panel is atwww.rae.ac.uk/panels. The panel criteriastatements detail how they will assess a broadrange of research, including applied researchrelevant to users in industry, commerce and thepublic sector.

28. As with previous RAEs, the assessmentprocess is based on expert review: panels will usetheir professional judgement to form a view aboutthe quality profile of the research described ineach submission, taking into account all theevidence presented. Annex A describes in moredetail how panels will formulate quality profiles.

29. Panels will assess submissions betweenJanuary and November 2008. A timetable for theexercise is at Annex D.

Timing and publication of resultsand of submissions 30. A quality profile for each submission will bepublished in December 2008. It will profile theproportions of research activity judged to meeteach of four ‘starred’ quality levels, in multiples of5 per cent. The profile for a submission that

RAE 03/2005 7

contains no research that meets or exceeds the onestar threshold will be 100 per cent unclassified. Ifa submission contains no work that meets thedefinition of research for the RAE, it will not beawarded a quality profile. Annex A describes howwe will formulate and present quality profiles,including the rounding methodology.

31. Alongside the quality profile, the fundingbodies will publish the full-time equivalent (FTE)number of staff included in each submission(submitted staff ). Unlike previous RAEs, we willnot collect any information about staff who havenot been included in submissions at eitherinstitutional level or at UOA level (non-submittedstaff ). Therefore, panels will only receive forassessment information about submitted staff. Inreaching their judgements, panels will not takeaccount of any information about non-submittedstaff.

32. Reports and feedback from the exercise willbe available early in 2009. In 2007 we willpublish information about the form of reportsand feedback. The feedback is expected tocomprise:

a. A published report by each main panelconfirming its working methods and giving abrief account of its observations about thestate of research (strengths, weaknesses andvitality of activity) in the areas falling withinits remit.

b. Feedback on each submission summarisingthe reason for the quality profile awardedwith reference to the published criteria of thesub-panel that assessed it. We expect to sendthis feedback only to the head of theinstitution concerned. In the case of jointsubmissions, we will provide this feedbackconfidentially to the heads of all of theinstitutions involved.

33. We will also publish on the internet thoseparts of submissions that contain factual data andtextual information about research activity. Weanticipate publishing submissions in the spring of2009. We will include the names of selected staffand the listings of research output but removepersonal and contractual details, and details of the

future research plans of institutions. We willpresent aggregate data on research students,research assistants supervised, and income. Thedata collection software will include a facility forHEIs to mark as confidential the names of anystaff or other data that should be omitted fromthe published data for reasons of security.

34. The results of the RAE are not subject toappeal. The funding bodies have taken legaladvice about the absence of a formal appealsprocess and it has been determined that thisabsence does not make the RAE process lessrobust. There are two main types of appealprocess that could be developed:

• either a process for appealing against thevalidity of the actual quality profile awardedin a specific case

• an appeal based on the procedures employedto reach the final outcome.

We consider that developing a system for appealagainst the validity of quality profiles awardedwould not be cost-effective as it would effectivelyconstitute a ‘shadow’ panel system. Developing asystem for appeals on procedural grounds mirrorsthe judicial review process.

Code of practice on preparingsubmissions and selecting staff 35. Compliance with equal opportunitieslegislation is an obligation for HEIs. For theirpart the four UK higher education funding bodieshave a statutory obligation as public bodies topositively promote a range of equalities issues,currently in respect of race, and quite probably bythe time of the 2008 RAE in respect of genderand disability. The RAE team is assisting thefunding bodies in meeting their statutoryobligations in a number of ways.

36. First, we are ensuring that all panels arebriefed on current and pending equalopportunities legislation that will affect RAE2008. We have instructed panel members to takeaccount of equalities issues that may have abearing on the volume of research undertaken andpublished by submitted researchers in the 2008RAE. These issues are detailed in RAE 02/2005

8 RAE 03/2005

‘Equality briefing for panel chairs, members andsecretaries’. Panel criteria will explain how thepanels will take account of these issues in assessingsubmissions.

37. Second, we are encouraging HEIs to submitthe work of all their excellent researchers,including those whose volume of research outputhas been limited for reasons covered by equalopportunities guidelines. HEIs will be invited toflag the contributions of such colleagues and touse a confidential section of the submission todescribe the circumstances affecting theircontribution to the submission (see paragraphs159 to 162).

38. Third, we will require the head of each HEImaking an RAE submission to confirm that theHEI has, in preparing its submissions andselecting staff for inclusion, developed, adoptedand documented an appropriate internal code ofpractice, which attends to all relevant equalopportunities legislation in force on thesubmission date. The requirement for a code ofpractice will both satisfy the funding bodies’ equalopportunities obligations to monitor the effects oftheir policies, and support HEIs in meeting theirown legal obligations. Clearly, managerialdecisions on which staff to submit are at thediscretion of HEIs but they will need to bedefensible. In collaboration with the EqualityChallenge Unit (ECU) we have developedguidance to HEIs in drawing up a code ofpractice that frames their decision-makingprocesses in the context of the principle ofequality of opportunity and all relevantlegislation. This is at Annex G. HEIs are notrequired to submit their code alongside their RAEsubmissions, but we may require them to submitit as part of the data verification proceduresdescribed in paragraphs 41-47. We will alsoextract and pass to HESA the following data fromRA1 to enable verification: for each individualsubmitted as research active in categories A and B,their UOA, HESA staff identifier code and dateof birth. Hence, HEIs should ensure thatdecisions documented in line with their code ofpractice are consistent with relevant parts of theirannual individualised staff return to HESA. The2007-08 (and subsequent) HESA individualised

staff return will require HEIs to return the UOAfor all academic staff, including those not selectedfor inclusion in RAE 2008.

Method of submission39. For the 2008 RAE, we are developing specialpurpose software to collect submissions fromHEIs, on behalf of the four funding bodies. Thisdata collection system will differ from that used inthe last two RAEs: it will be a web-basedapplication using a database hosted at HEFCEand will be the only way HEIs can make asubmission to the RAE. A pilot version of thesystem will be made available to HEIs by August2006. The final version will be available byDecember 2006.

40. Data entry will not be limited to direct entryon screen but will also allow HEIs to import datain various file formats, including XML files (theXML schemes are available at www.rae.ac.uk). Allcommunications between the web server andclient machines in institutions will be encrypted.Access to the database before the submission datewill not be permitted to anyone other thanauthorised personnel within each HEI and toHEFCE’s system administrators.

Data verification and access toresearch outputs41. All information provided by HEIs insubmissions to the 2008 RAE must be capable ofverification. We will check a proportion ofsubmissions from each institution as a matter ofcourse. Panel members will also be asked to drawattention to any data that they would like us toverify and this data will be investigated. HEIsshould therefore be able to provide detailedjustification for all information submitted.

42. Where possible, submission data will becompared with other available datasets includingHESA returns, other surveys conducted by thefour funding bodies, and information held by theResearch Councils. As the basis of return for somedatasets may differ, an exact match might not bepossible. However, where there appear to bemajor differences between submission data andother returns, we will investigate further. We will

RAE 03/2005 9

issue a fuller statement on the verificationarrangements for RAE data, including action wewill take when serious discrepancies are found, ata later date.

43. We remind HEIs that they must be able tomake available, or arrange access to, any item ofoutput cited,1 especially those that have notalready been collected in electronic format withthe submission (see paragraph 96). Outputs maybe requested for verification as well as for panelsto assess them.

44. HEIs should note that research outputs areonly eligible for submission if their actual date ofpublication or dissemination is within thepublication period. Where it is unclear, we willrequire HEIs to submit evidence of the actualdate of publication or public dissemination ofresearch outputs, particularly where an item bearsa date which is outside the limits of thepublication period (1 January 2001 to 31 December 2007).

45. For verification, we may require institutionsto physically submit any output listed in asubmission that is pending publication betweenthe submission date and the end of thepublication period (that is, 30 November 2007 to31 December 2007). If an HEI cannot makeavailable a requested item of output or provideevidence of its publication within the publicationperiod, that item will be removed from thesubmission and excluded from assessment. Therewill be no opportunity to submit a substituteitem.

46. Where an HEI is unable to verify any pieceof information contained in its submission, thatinformation will be excluded from assessment.The funding bodies will consider what furtheraction to take in any case where seriousdiscrepancies are found.

47. We are conscious of the potential additionalworkload verification may cause HEIs and aim tominimise this. To this end, data checking andverification will normally be conducted bycorrespondence with the RAE team. If HEIs wishto make it available, our verification team mayalso seek to rely on any relevant internal auditwork that an HEI has undertaken in preparing orsubmitting RAE data. However, members of theRAE team and other officers of the fundingbodies reserve the right to visit HEIs to verifysubmission information.

Data protection48. We will collect, store and process all RAEinformation from HEIs in accordance with theData Protection Act 1998. As stated in paragraph38, we will extract and pass some raw data toHESA to enable data verification for equalopportunities monitoring purposes. We will alsopublish parts of submissions on the internet (asdescribed in paragraph 33).

10 RAE 03/2005

1 In the case of non-text outputs, such as performances, we will require evidence that the outputwas disseminated in the public domain.

49. Each HEI eligible to receive funding fromone of the four UK higher education fundingbodies may make a submission in any of the 67UOAs listed at Annex C. Normally there shouldbe only one submission per UOA per institutionand only exceptionally will this be waived. Suchexceptions are set out in paragraphs 50, 51 and54d. A submission comprises data organised in sixforms (forms RA0-RA5). These are described inPart 3.

Multiple submissions50. The funding bodies will require any HEI thatwishes to make more than one submission(multiple submissions) in the same UOA torequest prior permission from the RAE manager.We will invite applications to make multiplesubmissions in December 2006 and responses willbe required (through the data collection software)by March 2007. Aside from the exceptionsmentioned in paragraphs 51 and 54d, a separateapplication will be required for each UOA inwhich the HEI wishes to make multiplesubmissions. Multiple submissions are more likelyto be permitted where the HEI contains two ormore substantially distinct research units that fallwithin the scope of a single UOA. Applicationswill be judged by the RAE manager, inconsultation with the relevant main and sub-panelchairs, taking account of both:

• the degree of academic distinctivenessbetween the proposed submissions

• the structural distinction between theproposed submissions.

We will only grant permission where a convincingcase satisfying both criteria is made.Administrative convenience will not be a factor.

51. HEIs that merged between 1 January 2001and 30 June 2005 should normally make onesubmission only to each UOA in which they wishto be awarded a quality profile. Where HEIsmerge after 1 July 2005 they may seek permissionto make two separate submissions in all of theUOAs in which they wish to submit, if forexample they anticipate difficulty in achievingacademic cohesion between the merger date and

the submission date. Permission is unlikely to begranted to such HEIs to make separatesubmissions only in selected UOAs.

Joint submissions52. Joint submissions to one UOA by two ormore UK institutions, of research they havedeveloped or undertaken collaboratively, areencouraged where this is the most appropriateway of describing the research. The method forjoint submissions is described in paragraph 53and is driven by two considerations:

a. Panels should receive joint submissions in theform of a unified entity, enabling them toassess a joint submission in the same way assubmissions from single institutions.

b. The RAE team must be able to verify data ina joint submission through the HEI to whichthe data relates.

53. Purely for administrative purposes, one HEIneeds to be identified as the lead in terms ofmanagement and data security of the jointsubmission. Two elements of the RAE data (RA2:outputs and RA5a: research environment andesteem) will be submitted by the lead HEI onbehalf of all the other HEIs in the jointsubmission. Each HEI involved in the jointsubmission will submit separate RAE data informs RA0, RA1, RA3a, RA3b, RA4, RA5b andRA5c (staff, student and income data, andinformation about individual staff circumstancesand Category C staff ). In line with thesesubmission arrangements, the data collectionsoftware will include the facility for HEIsinvolved in joint submissions to give view andedit permissions to the other HEIs involved in therelevant UOAs. In order for panels to be able tojudge the joint submission like a singlesubmission, the RAE team will aggregate theseparate data so that panels receive and assess it asa coherent whole.

54. The following rules apply:

a. Panels will assess the joint submission as theywould a single submission and the outcomewill be a single quality profile. The qualityprofile for a joint submission will list the

Part 2: Form and content of submissions

RAE 03/2005 11

HEIs involved in alphabetical order,irrespective of which HEI took theadministrative lead in making thesubmission.

b. Panels will provide confidential feedback onjoint submissions to the heads of all the HEIsconcerned; but the panels and the RAE teamwill not comment on the specificcontribution by an individual HEI to theoverall quality profile.

c. In line with a general RAE rule that noindividual may be submitted as Category Aresearch active in more than one submissionunless they hold a fractional employmentcontract with more than one HEI (seeparagraph 86), no individual can besubmitted in a joint submission and in asubmission from one HEI unless they holdtwo separate employment contracts with twodifferent HEIs.

d. Institutions involved in a joint submissionthat wish to make a separate submission inthe same UOA would normally be permittedto do so, subject to meeting the conditionsdescribed in paragraph 50.

55. The funding bodies will distribute researchfunding in respect of RAE 2008 results in linewith each one’s individual funding formula.Assuming that, as in former years, the FTE ofstaff submitted will be an element in thoseformulae, the funding bodies envisage using theFTE of staff submitted by each HEI involved in ajoint submission, unless the HEIs involvedpropose a different, agreed percentage split offunding at the time of submission.

56. Further guidance on the technical procedurefor making joint submissions will accompany theinvitation to make submissions.

Interdisciplinary research,arrangements for cross-referral andspecialist advice 57. There have been concerns that the assessmentof interdisciplinary research has presentedchallenges in previous RAEs. In view of these, wewill cross-refer parts of submissions and enhance

our arrangements for using specialist advisors toensure interdisciplinary research is identified andassessed by those competent to do so.

58. An HEI may request that parts of thesubmissions it makes to one UOA are cross-referred to other relevant sub-panels. Parts ofsubmissions may range from a body of workrelating to a submitted researcher (that is, all ofthe research output listed against that researcher)to all of the research output and textualcommentary relating to one or more researchgroup. Neither entire submissions nor singleoutputs may be cross-referred, although singleoutputs may be referred to specialist advisers.

59. Sub-panels may also request cross-referral ofparts of submissions on the same grounds, evenwhen the HEIs have not done so. In all cases, theRAE manager will consider the request and advicefrom the relevant main and sub-panel chairs.Where it is thought that cross-referral will enhanceassessment, the relevant parts will be cross-referredto all of the sub-panels concerned for advice.Although advice will be sought only on the cross-referred parts, the entire submission will be madeavailable to the receiving panel so that it can judgethe cross-referred part in the context of the wholesubmission. Advice will be sought and given onthe basis of the assessment criteria for the UOA towhich the work was originally submitted. Theoriginal sub-panel will retain responsibility for thequality profile awarded.

60. Sub-panels may request that parts ofsubmissions, including but not limited tointerdisciplinary research, are referred to specialistadvisers if they believe this will enhance theassessment process. This includes instances whereHEIs identify single or multiple research outputs asbeing outcomes of interdisciplinary research. TheRAE team has retained a database of individualswho were nominated as specialist advisers throughthe nominations process described in RAE03/2004 ‘Units of assessment and recruitment ofpanel members’. When we carry out the survey ofsubmission intentions in March 2007 we will alsoask HEIs to identify submissions in which theyintend to submit substantial bodies ofinterdisciplinary research. Responses to the survey

12 RAE 03/2005

will not be binding; but forewarning will help theRAE team to review and augment this database ofadvisors before the main assessment phase of theexercise begins in January 2008.

Content of submissions 61. Each submission will contain the core datadetailed in paragraphs a-i following. (The RAprefix refers to the research assessment form inwhich the data will be collected.) For detaileddefinitions of the data required in each RA form,see Part 3.

a. Overall staff summary (RA0): summaryinformation on research active staff selected(FTE and headcount) and related academicsupport staff (FTE) in the unit of assessment.The data collection software will populatesome of RA0 using the data HEIs enter inRA1.

b. Research active individuals (RA1): detailedinformation on individuals selected by theinstitution for inclusion as research active.

c. Research output (RA2): up to four items (orfewer if designated for a particular UOA byits sub-panel) of research output producedduring the publication period (1 January2001 to 31 December 2007) by eachindividual named as research active and inpost on the census date (31 October 2007).

d. Research students (RA3a): numbers of full-time and part-time postgraduate researchstudents and degrees awarded.

e. Research studentships (RA3b): numbers ofpostgraduate research studentships andsource of funding.

f. External research income (RA4): amountsand sources of external funding.

g. Textual description (RA5a): includinginformation about the research environmentand indicators of esteem.

h. Individual staff circumstances (RA5b).

i. Category C staff circumstances (RA5c).

62. We have considered whether we couldreasonably avoid collecting any of this datathrough the RAE and to rely instead on data thathas already been collected for other purposes.However, reconfiguring the units of assessmentsince the 2001 RAE means that we cannot rely onother data returns such as research student data orincome data submitted to other surveysconducted by the funding bodies or HESA.

63. Panels may request in their criteria statementsspecific, further information where this isreasonable, justifiable and explicit. Conscious ofthe potential data collection burden on HEIs, wewill ensure that any panel’s requests will notrequire HEIs to submit any new types of data.We will publish other information that panelsmay reasonably require HEIs to supply, and astructure for RA5a which may differ by UOA, inthe panels’ criteria statements.

64. When we circulate submissions to sub-panels, we will circulate standard analyses (listedin Annex F) of the quantitative data and anyadditional data analyses specified in each sub-panel’s criteria statement. Sub-panels’ criteriastatements will indicate how any additionalanalyses will be used in each UOA.

RAE 03/2005 13

General definitions 65. The following general definitions applythroughout Part 3:

a. ‘UOA’ means one of the 67 subject Units ofAssessment defined for the 2008 RAE whichare listed in Annex C.

b. ‘Submission’ means a complete set of formsRA0 to RA5c returned by an HEI in any ofthe 67 UOAs.

c. ‘Department’ means the staff included in asubmission to one of the 67 UOAsrecognised by the RAE, and, by extension,their work and the structures which supportit. RAE departments are often not identifiedwith a single administrative unit within anHEI, or in the case of joint submissions,across HEIs.

d. ‘Census date’ means the date determining theaffiliation of research staff to a particularinstitution. Staff may be submitted in theRAE by the institution by which they areemployed on this date (or in the case ofCategory C staff by the institution that is thefocus of their research), regardless of previousor forthcoming changes in their employmentstatus. The census date is 31 October 2007.

e. ‘Assessment period’ means the period from 1January 2001 to 31 July 2007 and is theperiod to which the research described insubmissions, including data about researchstudents and research income and the textualcommentary must relate.

f. ‘Publication period’ means the period duringwhich research outputs must be placed in thepublic domain (or in the case of confidentialoutputs, lodged with the sponsor) if they areto qualify for assessment in RAE 2008. Thepublication period runs from 1 January 2001to 31 December 2007 for all UOAs.

g. ‘FTE’ means full-time equivalent:

i. For staff, it refers to the extent of amember of staff ’s contracted duties ascompared to those of a typical full-timemember of staff in the same category.

The length of time in the year for whichthe individual was employed and therelative proportion of total contractedtime spent on research are irrelevant inreporting staff FTE.

ii. For students, it refers to the amount ofstudy undertaken in the year ofprogramme of study compared to a full-time student with the same qualificationaim studying for a full year.

FTEs should be expressed to two decimalplaces, such as 0.67, apart from thecontracted FTE of Category A staff, forwhich the minimum FTE that may bereported is 0.2 (see paragraph 75g).

h. ‘Selected staff ’ refers to the named staffincluded in RAE submissions by HEIs inaccordance with their own internal code ofpractice on preparing submissions andselecting staff for inclusion. Other staff maybe eligible for inclusion, that is, satisfy thedata definitions and requirements, but HEIsare not required to select for inclusion alltheir eligible staff.

i. ‘Returned’ refers to any data included in anyof forms RA0 to RA5c.

66. The following sections contain datarequirements, data definitions and notes aboutreturning data on:

• staff (section 1)

• research outputs (section 2)

• research students and studentships (section 3)

• research income (section 4)

• research environment, esteem, and individualstaff circumstances (section 5).

Part 3: Data requirements and definitions

14 RAE 03/2005

Part 3 Section 1 Staff details RA0and RA1

Staff summary data requirements: formRA0

67. The following summary data are required inRA0:

a. The FTE number and headcount on thecensus date (31 October 2007) of CategoryA individuals whom the HEI has submittedas research active and headcount of CategoryB, C and D individuals. The data collectionsoftware will populate some of RA0 using thedata HEIs enter in RA1. Each HEI maydecide which individuals to select as researchactive, according to its internal code ofpractice on preparing submissions andselecting staff for inclusion in them.

b. The FTE number on 31 July 2007 ofresearch assistants (see paragraph 70 and 71).There is no requirement to distinguishbetween postgraduate and postdoctoralresearch assistants. Only one total is required.

c. The FTE number on 31 July 2007 ofresearch technicians, scientific officers, andexperimental officers (one total) and of otherstaff devoted to research (a separate total).

Staff summary data definitions and notes

68. The definitions of staff categories A to D are:

a. Category A: academic staff in post and onthe payroll of the submitting institution onthe census date. Eligible Category Aacademic staff must be employed under acontract of employment with the HEI on thecensus date. Their contract must list researchand/or teaching as their primary function.

b. Category B: academic staff who held acontract with the institution after 1 January2001 and who left the institution (ortransferred into a department returned to adifferent UOA) after that date and before thecensus date, and who otherwise would havebeen eligible for inclusion as Category A.

c. Category C: independent investigators activein research who do not meet the definition

for Category A staff but whose research onthe census date is clearly and demonstrablyfocussed in the department that returnsthem.

d. Category D: independent investigators whomet the definition for Category C staffduring the period 1 January 2001 to 31October 2007 but not on the census date.

These definitions are repeated, for ease ofreference, in paragraph 76.

69. HEIs should only list research assistants,research technicians, scientific officers, experimentalofficers and other staff devoted to research if theyare clearly associated with research active staffreturned to that UOA in any of the Categories A-D. For example, research assistants funded onresearch council grants may be listed only if aprincipal or co-investigator on that grant is selectedas research active in that UOA. Research assistants,technicians and others who are only associated withnon-selected staff should be excluded.

Research assistants

70. Research assistants are individuals who are onthe payroll of and hold a contract of employmentwith the institution. They are academic staffwhose primary employment function is defined as‘research only’. They are employed to carry outanother individual’s research programme ratherthan as independent investigators in their ownright (except in the circumstances described inparagraph 79). They are usually funded fromresearch grants or contracts from ResearchCouncils, charities, the European Commission(EC) or other overseas sources, industry, or othercommercial enterprises, but they may also befunded from the institution’s own funds.Individuals who meet this definition but aredescribed in HEIs’ grading structures assomething other than research assistant (forexample research associate, assistant researcher)may be returned as research assistants providedthey meet this definition.

71. Where a research assistant is registered for ahigher degree they can be returned as a researchstudent (see Part 3 Section 3) or as a researchassistant or fractionally as both, with a maximumFTE across both roles of 1.0.

RAE 03/2005 15

Other research support staff

72. The terms technician, scientific officer andexperimental officer are generic terms and will needto be related to local practice. Where an individualwithin these groups has a split role betweenteaching and research, only the FTE devoted toresearch should be recorded. Where this is a veryminor part of their job description (less than 0.2 FTE) they should be omitted entirely.

73. Research support staff may include, forexample, computing staff and others who aredevoted wholly or largely to the support ofresearch. They may not include departmentalsecretarial or administrative staff except wheretheir contract of employment designates their roleas wholly supporting research.

Research active staff data requirements(form RA1)

74. Research active staff must be listed in one ofthe categories A to D (see paragraph 76 fordefinitions). Each HEI may decide whichindividuals to select as research active, inaccordance with its internal code of practice.

75. The following data are required on full andpart-time staff in the categories A to D whom theHEI has selected as research active:

a. HESA staff identifier (Category A and B staffonly). This is for equal opportunitiesmonitoring purposes.

b. Staff reference code: a code determined bythe HEI.

c. Surname.

d. Initials.

e. Date of birth.

f. Category of staff (A, B, C or D) on thecensus date.

g. Contracted FTE on the census date(Category A staff only). The minimum FTEthat may be reported is 0.2.

h. Whether the individual is a research fellow, asdefined in paragraph 80, (Category A and Cstaff only) and their source of funding (thedata collection software will include a list of

possible sources consistent with those definedby HESA).

i. Whether the individual has been in post forthe entire period from 1 January 2001 to 31 October 2007 (Category A only). ForCategory C, whether the individual’s researchhas been demonstrably focussed in thedepartment for the entire period from 1 January 2001 to 31 October 2007.

j. Year of entry into the academic profession.For the RAE we define this as the year anindividual gains an academic post in an HEIthat renders the individual eligible to bereturned as Category A staff. This is not amandatory field, but only throughcompletion of this field will an HEI be ableto identify early career researchers (seeparagraph 160d).

k. Dates (day, month and year) of starting andleaving employment as academic staff at theinstitution, if between 1 January 2001 and31 October 2007 in either case (Category Aand B only).

l. If the individual is on a fixed-term contract,the start and end dates (day, month and year)of the contract (Category A and B only).Staff on rolling contracts or a series ofrenewable fixed-term contracts will beregarded as fixed-term for this purpose,although institutions may wish to drawattention to their use of rolling contracts inthe textual part of their submissions,especially where a fixed-term contract has anexpiry date soon after the census date.

m. Details of any change of status between 1 January 2001 and 31 October 2007(Category A and C staff only), with the dateof the change. Only the most recent changeshould be recorded if the individual haschanged status more than once. Changes ofstatus are:

i. From Category A to Category C withinthe same institution.

ii. Released on unpaid leave or secondment,still on leave/secondment on the censusdate, and contracted to return within twoyears.

16 RAE 03/2005

n. The numbers (FTEs) of research assistantsand research students supervised on thecensus date. All research students supervisedshould be included, as well as those registeredat different HEIs. Jointly supervised studentsand assistants should be split, either accordingto the agreed division of responsibility, or inproportion to the number of supervisors.

o. Any research groups that the individualbelongs to, where relevant and up to amaximum of four. This is not a mandatoryfield. Some sub-panels may ask HEIs todescribe research groups in RA5a but neitherthe presence nor absence of research groups isassumed.

Research active staff data definitions

76. The definitions of staff categories A to D are:

a. Category A: academic staff in post and onthe payroll of the submitting institution onthe census date. Eligible Category Aacademic staff must be employed under acontract of employment with the HEI on thecensus date. Their contract must list researchand/or teaching as their primary function.

b. Category B: academic staff who held acontract with the institution after 1 January2001 and who left the institution (ortransferred into a department returned to adifferent UOA) after that date and before thecensus date, and who otherwise would havebeen eligible for inclusion as Category A.

c. Category C: independent investigators activein research who do not meet the definitionfor Category A staff but whose research onthe census date is clearly and demonstrablyfocussed in the department that returnsthem.

d. Category D: independent investigators whomet the definition for Category C staffduring the period 1 January 2001 to 31 October 2007 but not on the census date.

Notes on returning research active staff

Category A and B staff

77. If they satisfy the criteria in paragraph 76athen the following are eligible as Category A staff:

a. Staff who hold institutional/NHS jointappointments (that is, staff on ‘A+B’contracts). They should be returned asCategory A with an FTE less than 1.0,reflecting their contract of employment withthe institution.

b. Eligible research assistants (see paragraph 79).

c. Eligible research fellows (see paragraph 80).

d. Pensioned staff who continue in salariedemployment contracted to carry outacademic duties.

e. Academic staff who are on unpaid leave ofabsence or secondment on the census dateand are contracted to return to normal dutiesup to two years (inclusive) from the start oftheir period of absence, provided that anystaff recruited specifically to cover theirduties are not also listed as Category A.

f. Staff absent from their ‘home’ institution onthe terms described in sub-paragraph 75m,but working as contracted academic staff atanother UK higher education institution onthe census date. These staff may be returnedby either or both institutions. In such a casethe individual and both institutionsconcerned should agree how the return is tobe made. Their total FTE may not exceedtheir contracted FTE with their mainemployer.

78. Category A staff who are employed by anHEI and based in a discrete department or unitoutside the UK are eligible staff if the HEIdemonstrates in RA5a a clear and currentconnection on the census date with researchundertaken by the submitting department basedin the UK. Staff whose connection cannot bedemonstrated to the satisfaction of the RAEmanager, as advised by the relevant panel, will bediscounted from the assessment and removedfrom the RAE database.

RAE 03/2005 17

Research assistants

79. Research assistants are not eligible to be listedas research active staff unless, exceptionally, theyfulfil the criteria in this paragraph. Researchassistants eligible to be listed as research activestaff must be named as principal investigator on aresearch grant or significant piece of researchwork and satisfy the definition for Category Aacademic staff in paragraph 76a. Researchassistants must not be listed as Category A or Bresearch staff purely on the basis that they haveone or more research outputs to their name.Research assistants listed as Category A researchstaff may not also be included in the FTE countof research assistants in RA0 and RA1.

Research fellows

80. Research active academic staff should be listedas research fellows only if they hold a specificfellowship award on the basis of their own researchrecord or research proposals. The fellowship awardmust be to a named individual in recognition ofindependent research they have undertaken orproposed, must include a significant element ofexternal funding and must follow a process ofexpert review (including competitive review)involving an input from outside the institution.Such fellowships include Research Council fellows(senior, advanced or postdoctoral) and RoyalSociety research fellows and professors.

81. Staff on an HEI-funded or awardedfellowship, even with external referees involved inthe selection process, may not be listed as aresearch fellow for RAE purposes.

82. Research fellows who receive all of theirsalary funding directly from their sponsoringbodies may not be returned as Category A or Bstaff, unless so sponsored while on unpaid leave(see sub-paragraph 75m). However they areeligible for inclusion as Category C or D.

Casual and hourly paid staff

83. Casual staff, individuals employed underconsultancy contracts and individuals paid inresponse to a claim or invoice submitted by themas payment for fees or services, without a contractof employment are ineligible as Category A staff.

Individuals who receive payment automaticallythrough the HEI’s normal payroll, net of tax andnational insurance contributions, and who enjoysimilar employment rights to other employees (forexample, annual leave, sick pay and pensionentitlements) qualify as having a salaried contract,regardless of whether payment is calculated on anhourly basis.

84. Teaching assistants such as foreign languageassistants/lectors who are not employed asacademic staff cannot be included as researchactive members of staff.

Other notes on Category A and B staff

85. Other than individuals on secondment onthe terms described in paragraphs 75m and 77f,an individual may only be returned as researchactive Category A by more than one HEI if she orhe has a contract with and receives a salary frommore than one HEI. In such cases, the followingadditional conditions apply:

a. The two HEIs must ensure that the FTEvalue of the individual sums to no more thanthe lower of 1.0 or the individual’s totalcontracted FTE duties. If any individual isreturned in submissions with a contractedFTE that sums to more than 1.0, the RAEteam will rectify this through verification andwill apportion the FTE to each HEI pro-ratato the individual’s contracted FTE at eachHEI.

b. The number of research assistants and thenumber of research students supervised bythe individual returned by each institutionshould relate to each single institution, not toboth.

c. The same research works need not be cited inboth submissions.

d. Only grants/contracts held by the individualthat are administered through the submittingHEI may be returned by that HEI.

86. No individual may be returned as active on31 October 2007 in more than one UOA, exceptas described in paragraph 85 or, in exceptionalcircumstances, where an individual’s research hasundergone a significant shift in subject focus

18 RAE 03/2005

within the period 1 January 2001 to 31 October2007. Such individuals may be returned asCategory A in one UOA at 31 October 2007with a start date later than 1 January 2001 and asCategory B, up to that date, in another UOA ateither the same or a different institution. Wherean individual holds a joint appointment acrosstwo or more departments within the sameinstitution, HEIs must decide on one UOA inwhich to return the individual. They should notethe joint appointment in all relevant submissionsto other UOAs, using form RA5a.

87. Staff employed directly by the FederalUniversity of Wales may be returned in theuniversity’s submissions as Category A or B, or inthose of its constituent institutions as Category Cor D. However, the same member of staff maynot be returned as Category A by both theFederal University and any of its constituentinstitutions.

88. The following are eligible as Category B staff:

a. Staff who died in post before the census date.

b. Staff who left the HEI before the census date.

c. Staff who transferred to a differentdepartment in the same HEI before thecensus date.

Category C and D staff

89. The following are eligible as Category C staffor Category D staff:

a. Individuals whose research at the census date(or for Category D staff, during the period 1January 2001 to 31 October 2007 but not atthe census date) is clearly and demonstrablyfocussed in the department, but who areexcluded from Category A (or B) becausethey do (did) not have a contract with theinstitution, do (did) not receive a salary fromit, or are (were) not academic staff. Thedepartment or institution must be (havebeen) the focus of the individual’s researchactivity. Where they meet these criteria, thefollowing are eligible:

i. Vice-chancellors or principals.

ii. Librarians.

iii. Staff in university museums.

iv. Staff employed directly by the FederalUniversity of Wales but returned in thesubmissions of its constituent colleges.(See paragraph 87 for instructions onhow such staff should be returned insubmissions made by the FederalUniversity.)

v. Staff in ‘embedded’ research council units.

vi. Retired staff who are still active inresearch.

vii. Research active academic staff recruitedspecifically to cover the duties of staff onleave of absence or secondment, returnedin accordance with sub-paragraph 77e.

90. For each individual returned in Category C,an entry in RA5c (Category C staff circumstances)will be required giving evidence that their researchis clearly and demonstrably focussed in thedepartment (see also paragraph 163).

91. The following are ineligible as Category C orCategory D staff:

a. Visiting professors, fellows and lecturers ifthey normally carry out their research outsidethe department.

b. Short-term visiting staff (less than 12consecutive months in the period 1 January2001 to 31 October 2007). Reference may bemade to such individuals in RA5a.

Part 3 Section 2 Research outputs(RA2)

Data requirements

92. Details are required of up to four items ofresearch output, (or fewer if a UOA’s sub-paneldesignates fewer in its criteria statement) that eachCategory A or C staff member has produced andthat have been brought into the public domain2

during the publication period (1 January 2001 to31 December 2007).

93. For each output cited, enough informationshould be given to enable the RAE team andpanels to determine precisely what is being listed,

RAE 03/2005 19

2 Confidential outputs must have been lodged with the body to which they are confidential in thepublication period.

whether it is a product of sole or multipleauthorship or production, in what physical formit exists and where it may be found. Thefollowing are required for each output:

a. Output number: sequentially from one tono higher than four for each individualreturned. This number is for administrativeconvenience of referencing only. The datacollection software will also enable HEIs toenter an output reference code, determinedby the HEI.

b. Year of output: the year in which the outputbecame publicly available. For books andchapters in books, the year is required; forother forms of output, months or dates willbe required.

c. Type of output: all forms of publicly availableassessable output will be acceptable as willitems listed as confidential reports (seeparagraph 98). A detailed indicative list ofoutput types will be provided in the RAE datacollection software and is likely to includepatent/published patent applications, software,internet publications, performances,compositions, designs, artefacts and exhibitionsas well as the print media detailed below.

d. Title of the output: if the output has no titlea description is required. (See also subparagraphs 93g and 93h for requirements onchapters and journal articles.)

e. Co-authors: names of first, second and thirdco-authors and whether they are internal orexternal to the HEI; and the number ofadditional co-authors.

f. For books: title, number of pages, publisher,year published, ISBN number.

g. For chapters in books (including othershort works such as contributions tocollections of essays published in book form):title of chapter, page numbers of chapter, titleof book, name(s) of editor(s), publisher, yearof publication, ISBN number.

h. For articles in journals: title of article, pagenumbers of article, title of journal, volumenumber, month and year of publication,ISSN number.

i. For conference contributions: name ofconference/published proceedings, number ofpages, month and year published (or in thecase of non-text based material, date ofconference and medium of output).

j. Other outputs:

i. In the case of public exhibitions: title orbrief description, the gallery where theexhibition was held, the number of piecesexhibited and the opening and closingdates of the exhibition.

ii. For performances: title, place, date(s) andbrief details of the performance.

iii. For all other outputs: title or briefdescription and date and place at whichoutput was made publicly available. Inthe case of internet publications or webcontent, the URL. Forthcoming journalor other articles that are electronicallyavailable before the planned publicationdate on publishers’ or authors’ web-sitesor institutional repositories should betreated as internet publications and thefacility described in paragraph 94 may beused to provide other factual details.Panels, in their statements of criteria, mayspecify further required details.

k. Whether the output is the outcome ofinterdisciplinary research (so panels may ifnecessary identify suitable specialist advisers).

l. If appropriate, the research group to whichthe research output is assigned. This is not amandatory field and neither the presence norabsence of research group is assumed (see alsoparagraph 75o).

94. Brief, additional information may also begiven in RA2 ‘other relevant details’ to identifyrelevant, factual circumstances concerning anyoutput. It need not be supplied in every case. Itmay be, for example:

• to identify a keynote address to a conference

• to identify an invited conference paperespecially where the perceived status of theconference is high

20 RAE 03/2005

RAE 03/2005 21

• to indicate the significance or impact of anapplied research outcome

• to identify the research content or author’scontribution in edited works, translations, orco-authored works.

In the case of a non-text output, it may be usedto give further information on the whereabouts ofa work or to note that a photographic, electronicor other record exists. It may not be used tovolunteer opinions about the relative quality of anoutput. See panel criteria statements for furtherguidance, including the word limit for this text,which, unless otherwise stated, will be 300 words.

95. For research outputs produced in a languageother than English or Welsh, a short abstract inEnglish is required in the other relevant detailsfield describing the content and nature of thework. Panels will use this to identify appropriatespecialist advisers to whom the work may bereferred. The abstracts themselves will not formthe basis for assessment. This requirement iswaived for outputs submitted to any of UOAs 51to 57 if the output is produced in any of thelanguages in the remit of the UOA.

96. We will also require HEIs to submit somecategories of output in electronic format withtheir submissions. These categories are likely tobe: journal articles, chapters in books andpublished conference proceedings. The method ofsubmission may involve HEIs depositing items ona protected web-site or giving access toinstitutional repositories of publications. Detailswill accompany the pilot version of the datacollection software in August 2006.

Data definitions and notes

97. Research outputs may be any form ofpublicly available assessable output. All itemseligible for submission must be publicly availableby 31 December 2007. Confidential outputsmust be lodged with the body to whom they areconfidential by 31 December 2007. An itemexpected to be published after 31 December 2007should not be submitted, even if it has beenaccepted for publication. HEIs may list an outputwhose publication is pending if they are satisfiedthat it will enter the public domain before the end

of December 2007. Such items will be flagged insubmissions, and where only some of therequirements in paragraph 93 can be supplied, wewill require full details to be submitted by 31 January 2008. HEIs may have to physicallysubmit any output so flagged for verificationpurposes. If an HEI cannot make it available, theitem will be removed from the submission,discounted from assessment, and no substitutewill be accepted.

98. Confidential reports include any itemproduced for and lodged, in the publicationperiod, with a company, government body orother research sponsor(s), but which has not beenpublished because of its commercial or othersensitivity. A confidential report may only besubmitted if the HEI has prior permission thatthe output may be made available for assessment.HEIs will confirm permission has been securedwhen they make submissions. If the RAE teamrequests a confidential report for assessment theHEI must make it available. All panel members,advisers, observers and others involved in theassessment process are bound by a confidentialityagreement. Therefore, HEIs may submitconfidential reports without compromising anyduty of confidentiality upon them. There may bemain or sub-panel members who HEIs believewould have a commercial conflict of interest inassessing confidential reports. HEIs will berequired to name such individuals when makingsubmissions.

99. HEIs may list any output produced byselected Category A and Category C staff duringthe publication period regardless of where thosestaff were employed during the period. Referencemay be made in RA5a to research outputs otherthan those listed in RA2.

100. For publications the date of appearance willgenerally be that indicated by its imprint. If thedate of imprint differs from the date of actualpublication, and one of these dates lies outside thepublication period, the variance must be noted inthe data collection software. Examples could bejournal volumes relating to a particular year in asequence but actually published in a differentcalendar year. In such circumstances institutions

may be required to provide evidence of the actualdate of publication for data verification purposes,such as a letter from the publisher. In all cases thedate for determining whether or not an outputwas produced within the publication period andhence is eligible for submission will be the date atwhich it became publicly available. For webcontent and electronic corpora, HEIs will need tomaintain proof of the date at which the itembecame publicly available and of its content atthat date, for example a scanned or physicalprintout, date stamped.

101. Edited works including editions of textsand translations may be included if they embodyresearch as defined in Annex B. Systematicreviews may also be included where they meet thiscondition. Editorships of journals and otheractivities associated with the dissemination ofresearch findings should not be listed as outputon RA2. (See panel criteria statements for furtherguidance.)

102. Co-authored/co-produced outputs shouldnot be counted pro-rata. The names of up tothree co-authors/co-producers may be recordedand where there are more than three, the numberof others will be required. Any co-authors or co-producers who have not been members of staff,research assistants or research students, at theinstitution during the publication period shouldbe noted as such (see paragraph 93e). In principle,where two or more co-authors or co-producers ofan output are returned as research active (in eitherthe same or different departments and UOAs) anyor all of these may list the same output. However,panels’ criteria statements will give more detailson whether and how they would like HEIs todescribe or account for instances where the sameco-authored output is listed more than once inthe same submission.

Ineligible outputs

103. Theses, dissertations or other itemssubmitted for a research degree including doctoraltheses may not be listed. Other assessablepublished items based on research carried out fora research degree may be listed. HEIs may useRA5a to alert panels where a research degree has

been conferred on a submitted staff member inthe assessment period.

104. HEIs may not list as the output of aCategory A or Category C staff member anyoutput produced by a research assistant orresearch student whom they supervised.

Part 3 Section 3 Research studentsand studentships (RA3a and RA3b)

Research student data requirements

105. The following data is required abouthome/EC and overseas research students who aresupervised by any of the submitted research activestaff in Category A to D who are listed in RA1:

a. Headcount of students on research-basedhigher degrees for whom the year ofprogramme of study falling on 31 July ineach of 2001, 2002, 2003, 2004, 2005,2006 and 2007 is full-time.

b. Headcount and FTE of students onresearch-based higher degrees for whom theyear of programme of study falling on 31 July in each of 2001, 2002, 2003, 2004,2005, 2006 and 2007 is part-time.

c. Total number of research doctoral degreesawarded (the date of award approval ratherthan ceremony – see paragraph 122) in eachcalendar year 2001, 2002, 2003, 2004, 2005,2006 and 2007 (partial year to 31 July).

d. Total number of research masters degreesawarded (the date of award approval ratherthan ceremony – see paragraph 122) in eachcalendar year 2001, 2002, 2003, 2004, 2005,2006 and 2007 (partial year to 31 July).

Research student data definitions andnotes

Research-based higher degree

106. A research-based higher degree is oneawarded primarily on the basis of a substantialthesis (or equivalent) submitted by the studentand resulting from the student’s original research.Regulations on examiners should stipulate thatthe student’s work is examined by at least two

22 RAE 03/2005

examiners, individually appointed for the student,one of whom is external. Students registered for aspecialist doctoral degree should only be includedin the submission if their degree is awarded onthe basis described in this paragraph and if theysatisfy the ensuing criteria.

107. Students undertaking taught mastersprogrammes and taught doctoral programmes areexcluded.

108. Students who have completed theirresearch work and are writing up their thesis (orequivalent) are excluded. Students are consideredto be writing up when their research work iscomplete and they will not be undertakingadditional research, even if they still receive asmall amount of supervision and/or still haveaccess to other facilities at the institution.Students registered for research qualificationsawarded primarily on the basis of published worksshould also be excluded, unless they undertakeresearch at the institution.

109. Students should be actively supervised by asubmitted member of staff. Students employed byoutside research organisations and based outsidethe department for the majority of their studyshould be excluded. Students engaged on forms ofwork-based learning such as Knowledge TransferPartnerships (KTP, formerly the TeachingCompany Scheme, TCS), who are registered for aresearch-based higher degree can be includedprovided they are based in the department for themajority of their study.

Registration

110. Student headcounts and FTEs may only berecorded if the student registers for the year ofprogramme of study. For the purposes of the RAE,registered students are those who have a bindingundertaking to pay a fee to the institution fortuition and supervision of research for aprogramme of study (whether or not they pay thefee themselves). Acceptance of a place does not initself qualify a student to be counted in the RAE.If a programme is run jointly by two or moreinstitutions, only the institution that receives thefee should include the student. If more than oneinstitution receives a fee from a student, the

institution that receives the largest portion shouldinclude the student. If the institutions receive thesame amount, they must ensure that only oneincludes the student. In some cases the institutionin which the student is based collects the fee onbehalf of another institution. Where this is thecase, if the fee is transferred in full3 to anotherinstitution, the institution to which the fee istransferred should include the student.

111. Research students should be returned bythe institution where they are registered, except inthe case of submissions from institutions in Walesand Scotland when research students should beincluded under the institution where they arestudying. (This difference reflects differencesbetween the countries in the way students arecurrently counted for funding purposes.) Astudent can only be registered at one institutionfor any individual qualification.

112. In some cases students may be registered tostudy for a research degree at an institution whichdoes not have, or until recently did not have, thepower to award its own higher degrees at theappropriate level. Another institution mayvalidate the programme and award the degrees ofsuch students. In these cases the students shouldbe counted by the institution where they areregistered. The validating institution may notcount these students.

Year of programme of study

113. A student’s first year of programme ofstudy starts when they first begin studyingtowards the qualification. Subsequent years ofprogramme of study start on or near theanniversary of this date.

Full-time equivalence

114. For the RAE, a student’s FTE refers to theamount of study undertaken in the year ofprogramme of study compared to a full-timestudent with the same qualification aim studyingfor a full year. A student is said to be full-time ifthey are normally required to attend the institution,or elsewhere, for a total of at least 24 weeks withinthe year of programme of study; and during thattime is normally expected to study an average of at

RAE 03/2005 23

3 If the fee itself is not transferred but is used to offset payments between institutions in respect of the student,this should be treated as if the fee had been transferred.

least 21 hours per week. The FTE should be theFTE for the year of programme of study falling onthe relevant date and not an estimate of thestudent’s activity on the date. So a studentundertaking full-time study should be returned as1.0 FTE; a student studying part-time should bereturned with an FTE that reflects the amount oftime they study compared to a full-time student.

115. The FTE for a part-time programme iscalculated by comparison with an equivalent full-time programme. When viewed as a whole, thetotal FTE for a part-time programme shouldequal the total FTE of the equivalent full-timeprogramme. If an equivalent full-time programmedoes not exist, a reasonable academic judgementshould be made of the load relative to a full-timestudent.

116. In some cases a student may not study for afull year in their final year of programme of study(for example, the second year of programme of a15-month MPhil programme). Here the studentshould be recorded as part-time in their final year,with an FTE reflecting the proportion of a fullyear for which they have studied. This should berecorded even if the student is not activelypursuing studies on 31 July in their final year.

117. Category A staff who are registered for aresearch degree may be included as part-timestudents. Their total FTE (as Category A staff andresearch student) may not exceed 1.0.

118. Periods of suspension need to be accountedfor when calculating both the FTE and the yearof programme of study. For example a studentwhose registration is suspended for three monthson ill health grounds in the first year of a threeyear full-time programme of study should becounted as follows: year one: 0.75, year two: 1.0,year three: 1.0, year four: 0.25.

119. Research students should be returned tothe UOA in which their supervisor is returned.Where a student is supervised by more than onemember of staff and these supervisors are returnedin more than one UOA, the student may bereturned split either according to the agreeddivision of responsibility, or in proportion to thenumber of supervisors.

120. The arrangements for recording researchassistants who are registered for a higher degreeare in paragraph 71.

121. Where a department contains students ondegrees not covered by the definition at paragraph106 but whose presence is considered significant,attention may be drawn to these in RA5a.

Degrees awarded

122. The award should be recorded when theinstitution’s senate, or other body or personempowered to approve the award, formallyapproves the award.

123. A degree awarded may be returned in linewith paragraph 119, that is, in one UOA orfractionally in more than one UOA, split eitheraccording to the agreed division of responsibilityor in proportion to the number of supervisors.

Research studentships data requirements

124. Data is required on the number of newstudentships awarded for research-based higherdegrees and held in the department in eachcalendar year 2001, 2002, 2003, 2004, 2005,2006 and 2007 (partial calendar year to 31 July in2007). Studentships must be held by studentsreturned in RA3a. Numbers of studentshipsshould be listed against each of the followingsources:

a. OST Research Councils et al – includesstudentships awarded by the ResearchCouncils covered by the Office of Scienceand Technology (OST), the former Artsand Humanities Research Board, theScottish Executive, DEL and theDepartment of Agriculture and RuralDevelopment Northern Ireland. It excludesresearch assistantships funded by theResearch Councils.

b. UK-based charities – studentships awardedby charitable foundations and charitabletrusts etc, based in the UK and registeredwith the Charities Commission, or bythose recognised as charities by the InlandRevenue in Scotland.

24 RAE 03/2005

c. UK central government – includesstudentships awarded by the BritishCouncil and NHS R&D.

d. UK local authorities, health and hospitalauthorities.

e. UK industry, commerce and publiccorporations.

f. Institutional self-funded – includesstudentships funded by institutionswhether from grant or endowment incomeor from other sources.

g. Overseas.

h. Other – funded from any other sourceincluding self-funded by the student, withthe source specified.

The number of new studentships should be listedagainst each source. A studentship sponsored bymore than one source should be listed fractionallyagainst each one.

Research studentships data definitionsand notes

125. Studentships are only eligible if the fee paidis at or above the studentship fee level paid in therelevant year by UK Research Councils onresearch council funded projects. Studentships forpartial fees below these amounts, for maintenanceonly, or bench fees only (charged formaterials/consumables used by postgraduatestudents) or college fees, may not be included.

126. Studentships for part-time students areonly eligible if the fee paid is at or above the levelsdetailed in paragraph 125, pro-rata to the full-time equivalent.

127. The number of studentships held bypostgraduate students registered for researchdegrees should be recorded. Each new studentshipshould be counted once only, in the calendar yearin which it first started (or partial calendar year,to 31 July in 2007). An extension to the durationof a studentship, or a change in its source offunding is not regarded as new for the purpose ofthis return. If a student holds a studentship frommore than one source in the first year thestudentship should be listed fractionally againsteach source.

128. If a student is jointly supervised by staffbeing returned to more than one UOA, thestudentship should be returned fractionally, in thesame way as the student (see paragraph 119).

129. The number of new studentships allocatedby HEIs through Research Council doctoraltraining accounts (DTAs) should be returnedunder ‘OST Research Councils et al’.

Part 3 Section 4 Research Income(RA4)

Data requirements

130. Data are required on external researchincome received in each financial year between 1 January 2001 and 31 July 2007. For 2001 a partyear return is required (1 January to 31 July 2001).

131. External research income data should belisted against the following sources:

a. OST Research Councils et al – includes allresearch grant and contract income fromresearch councils covered by the OST, theformer Arts and Humanities ResearchBoard, the British Academy and the RoyalSociety. Research Council income inrespect of the value of time spent byresearchers on Research Council facilitiesshould be listed in a separate line (seeparagraphs 150 and 151).

b. Joint Infrastructure Fund (JIF) – theproportion of grants awarded from the JIFprovided from OST or Wellcome Trustfunds may be included. The proportion ofJIF grants provided by the highereducation funding bodies may not beincluded. Such awards may, however, bereferred to in the textual part ofsubmissions

c. Joint Research Equipment Initiative (JREI)– grants awarded under this initiativeshould be returned broken down into theportions provided by Research Councils,industrial sources and funding bodies.

d. UK-based charities – research grants andcontract income from all charitablefoundations and charitable trusts etc, based

RAE 03/2005 25

in the UK and registered with the CharitiesCommission, or from those recognised ascharities by the Inland Revenue inScotland.

e. UK central government bodies, bodiesgoverned by devolved administrations,local, health and hospital authorities –includes all research grants and researchcontract income from UK centralgovernment bodies, UK local authoritiesand UK health and hospital authorities,including NHS R&D, except the ResearchCouncils and UK public corporations. Thistherefore includes governmentdepartments, Northern Irelanddepartments, the Scottish Executive, theWelsh Assembly Government and allorganisations financed from centralgovernment funds. Regional developmentagency (RDA) income should be returnedhere in a separate line. Income from non-departmental public bodies (NDPBs)except the four higher education fundingbodies should be returned under thisheading. It excludes NHS funding inrespect of clinical or non-clinical teaching.Some panels may specify in their criteriathat NHS R&D funding should bereported as a sub-heading under ‘UKcentral government bodies’ (see paragraph153).

f. UK industry, commerce and publiccorporations – includes all research grantand contract income from industrial andcommercial companies operating in theUK. Public corporations (defined aspublicly owned trading bodies, usuallystatutory corporations, with a substantialdegree of financial independence) in theUK include nationalised industries andbodies such as the Ordnance Survey.

g. EU government bodies – includes allresearch grant and contract income from allgovernment bodies operating in theEuropean Union (EU), including theEuropean Commission (EC) but excludingbodies in the UK. Socrates Erasmus, EU

Lingua programme, EU Leonardoprogramme, Trans-European MobilityProgramme for University Studies(TEMPUS), European Social Fund (ESF)and similar grants are excluded from thisreturn, apart from where a portion of agrant from one of these sources wasgranted for research and spent on research:that portion may be returned. Where aninstitution is acting as a co-ordinator for anEC award it should only include the fundsthat will be directly spent in thatinstitution as one of the contractors for theaward. In respect of countries joining theEU during the assessment period, incomeshould only be included from the date ofaccession onwards; income received beforethe date of accession should be returnedunder ‘other overseas’.

h. EU other – includes all research grants andcontract income from all non-governmentbodies operating in the EU, excluding theUK. In respect of those countries joiningthe EU during the assessment period,income should only be included from thedate of accession onwards – incomereceived prior to the date of accessionshould be returned under ‘other overseas’.

i. Other overseas – includes all research grantand contract income from bodies operatingoutside the EU (with the source specified:‘Other overseas’ will be a specified source).

j. Other sources – includes all research grantsand contract income not covered by theheadings described in this paragraph. HEIsmay specify the source if they wish or listas ‘Other’. Income from another HEI,where that other HEI is the ultimate sourceof funding should be included.

Data definitions and notes

132. The HESA definition of research incomeapplies: that is income in respect of externallysponsored research carried out by the institution(or its subsidiary) and for which directly relatedexpenditure has been incurred (including recoveryof indirect costs). The actual research income

26 RAE 03/2005

from external grants or contracts for whichexpenditure has been incurred in each financialyear should be recorded, rather than the initialvalue of the grants or contracts. Where a grant orcontract award is for a project lasting longer thanone accounting period, the amount returned foreach year should be the amount recorded in thefinancial statements for that year, not the fullamount of the award.

133. We have excluded certain sources ofincome as detailed in paragraphs 142 to 143 onthe basis that we expect grant income to havebeen won through competitive peer review. Thisexpectation does not apply in the case of contractincome or income from commissioned research.

134. Grants or contracts for which income isincluded in a submission must be associated withat least one member of staff who is submitted asresearch active in Categories A-D. Subject to this,all income in respect of research, as defined forthe RAE (see Annex B), carried out at theinstitution and for which directly relatedexpenditure has been incurred, should bereturned.

135. We would normally expect consultancyincome to be excluded since consultancy isnormally concerned with applying existingknowledge. However, consultancy income may beincluded if the work it funds meets the RAEdefinition of research, irrespective of the nature ofthe contract or invoicing arrangement.

136. Income should be stated at full valueincluding any recovery of indirect costs whetherretained by the institution or department, or atthe disposal of an individual in a department.Amounts in respect of costs for existing staff orfacilities within research grants or contracts maybe included.

137. External research income administeredthrough the institution’s own accounts should beincluded. So should income administered throughthe accounts of a subsidiary company, whereverthat income was consolidated in the institution’saudited financial statements. Other income,including income received directly by individualsworking in the department, may not be returned

but may be referred to in the textual part of thesubmission. Similarly, gifts of equipment, andfunding for building work not paid as part of thegrant or contract for a specified research project,may also be cited in the textual part of asubmission and should be excluded from RA4.

138. Research grant or contract income forbuilding projects may be returned. Such income,whatever its source, should be returned under‘other’, identifying the source and adding‘(building)’. However, general building grants ordonations are not eligible and any part of abuilding grant that relates to teaching or othernon-research use may not be included.

139. Where an institution has income fromendowments made for research purposes, and hasincurred directly related expenditure on researchwith which at least one member of staff submittedas research active is associated, this may bereturned under ‘other’, identified as endowments.

140. Research income from KTPs is eligible,apart from any portion in respect of studentshipsor tuition fees (see paragraph 141).

Ineligible income

141. Income from studentships or tuition fees isineligible, including money from a research grantor contract applied to support a research student,whether or not this formed part of the originalgrant or contract or was decided subsequently.

142. Except for JREI, grants awarded underfunding body special initiatives are ineligible.Where a grant scheme is co-funded by one ormore of the funding bodies and an externalpartner body, the element of the grant attributableto the funding body(ies), as specified in the grantaward letter, is ineligible.

143. Science Research Investment Fund incomeand similar formula-based schemes funded by theHE funding bodies are ineligible.

144. Gifts of any kind for research purposes forwhich the institution did not incur qualifyingdirect expenditure, including all gifts ofequipment, are ineligible: reference to them maybe made in the textual parts of submissions.

RAE 03/2005 27

Other notes

145. Where a grant or contract is held, or thework conducted, across more than one UOA itshould be divided between submissions todifferent UOAs according to the way the incomehas been used. Research projects which are fundedfrom several sources should have their incomeallocated under respective headings to reflect theactual source of the income.

146. Where a project is undertaken by severalinstitutions or organisations and one institutionacts as the ‘lead institution’, only income receivedand spent by the institution making the return,and relating to the work actually beingundertaken by it, should be included. The leadinstitution should not therefore include anyportion of income in a particular year passed onto other institutions or organisations. Otherinstitutions who receive a portion of a grant orcontract as described in this paragraph may returnthe income against its original source as listed inparagraph 131.

147. Researchers will on occasion move betweeninstitutions during the progress of a grant orcontract. If the original institution continues toreceive income from the grant/contract then thisshould be returned by that institution alone. Ifthe old grant/contract is cancelled and a new oneis issued to the new institution, then the incomefrom each grant/contract should be cited by theappropriate institution.

148. If an individual moves institution butcontinues to work on a grant/contract held jointlywith other members of staff at the originalinstitution, the income may be cited only by theinstitution which received it. The ruling ongrants/contracts held by researchers at more thanone institution applies only to awards that arespecifically allocated at the start to individuals atmore than one institution, not in the case ofindividuals who move institution after the awardis announced.

149. Where Scottish HEIs hold small biomedicaland clinical research grants and contracts incomefrom the Chief Scientist Office at the ScottishExecutive Health Department, they may be

included under ‘OST Research Councils et al’, assuch awards are made in competition against apool of research funds. Biomedical and clinicalresearch grants awarded by the Wales Office ofResearch and Development for Health and SocialCare (WORD) on the basis of peer review may beincluded under ‘OST Research Councils et al’.Any grants or contracts awarded by WORD for aspecific study in its capacity as an agency of theWelsh Assembly Government should be returnedunder ‘UK government’.

150. The estimated value of Research Councilfacility time allocated through peer review andused by researchers in university departments maybe returned under ‘OST Research Councils et al’(see paragraph 131a) and will be identified in aseparate line. The lead Research Councilresponsible for access and funding of a namedfacility will supply data on the value of theseallocations to the HEIs concerned and to theRAE team. We expect that the Research Councilswill supply data to HEIs relating to the RAEassessment period 1 January 2001 to 31 July 2007and this will be made available to HEIs insufficient time for it to be included in theirsubmissions. We will consult with the ResearchCouncils on these arrangements and will issuefurther guidance, including confirmation of theprovisional dates for the supply of data to HEIs.

151. The value of Council for the CentralLaboratory of the Research Councils (CCLRC)facility development grants, which are allocatedthrough peer review to researchers in universitydepartments but accounted for by CCLRC, mayalso be returned as a separate line under ‘OSTResearch Councils et al’ (see paragraph 131a).CCLRC will supply data on the value of thesegrants to the HEIs concerned and to the RAEteam. Further arrangements, includingconfirmation of the provisional dates for thesupply of data to HEIs will be issued in duecourse.

152. Where, exceptionally, VAT is payable on allor part of a research grant or contract at the pointwhere the money is paid to the HEI, this shouldnot be included as part of the institution’sresearch income.

28 RAE 03/2005

153. Some sub-panels may wish HEIs to reportresearch income by research group. In this case,the sub-panel’s published criteria statement willspecify this and the data collection software willinclude a facility to report income by group.Some sub-panels may also wish HEIs to reportNHS R&D research income as a separate sub-heading under ‘UK government central bodies’.In this case, the sub-panel’s published criteriastatement will specify this and the data collectionsoftware will enable it.

Part 3 Section 5 Researchenvironment and esteem (RA5a)154. Information is required describing theresearch environment and evidence of esteem inthe department. Sub-panels will specify in theircriteria statements a structure and minimumcontent for RA5a to enable them to assesssubmissions against their published criteria.

155. Detailed guidance on the requirements forthe content of RA5a will be specified in panelcriteria. This will invite departments to explaintheir submission in terms of their researchenvironment and organisation, including whererelevant:

• information about the HEI’s strategicinvestment in the UOA

• their strategies for promoting anddeveloping research staff, particularly thosenew to research

• their strategies in relation to collaborativeresearch with academic and non-academicbodies and with overseas HEIs

• their strategies in relation tointerdisciplinary research

• the significance of their research on a rangeof academic and other audiences,including, where appropriate, practitioners,businesses and other users.

RA5a will also invite HEIs to describe wheremarks of esteem in research have been conferredupon the department, or individuals in thedepartment, during the assessment period.

156. RA5a will have a length limit. We willdecide this limit in the context of the criteriastatements drawn up by panels. We do not expectit to be less than the combined limit for RA5 andRA6 in the 2001 RAE (that is, four A4 sides forsubmissions of up to six FTE staff, eight A4 sidesfor submissions of between six and 20 FTE staff,and one additional A4 side for each additional setof 20 FTEs up to a maximum of 14 sides). Wewill publish the limit alongside the panel’s finalpublished criteria statements.

157. Where sub-panels request them, we willmake available forms RA5 and RA6 from HEIssubmissions in the 2001 RAE, describing researchplans and strategies.

158. Panel criteria statements will describe howthey will use the information in form RA5a inassessing submissions.

Individual staff circumstances RA5b

159. Panels will use the information suppliedconfidentially in RA5b in assessing submissionsagainst their published criteria. No informationcontained in RA5b will be published.

160. Information is required describing anyindividual staff circumstances that havesignificantly adversely affected their contributionto the submission. Such circumstances mightinclude, but are not restricted to:

a. Matters covered by legislation includingmaternity leave.

b. Part-time working.

c. Engagement on long-term projects.

d. Early career researcher status – defined ingeneral for RAE 2008 as entering theacademic profession from 1 January 2001to 31 October 2007 (see paragraph 75 j).Panel criteria statements should be referredto for subject specific guidance.

e. Prolonged absence (more than six monthsconsecutively from 1 January 2001 to 31 October 2007) due to:

i. Secondment.

ii. Career break.

iii. Ill health or injury.

RAE 03/2005 29

161. HEIs will be required to indicate in RA1those individuals whose circumstances havehindered their contribution to the researchdescribed in the submission significantly and whothey therefore describe in RA5b. HEIs will needto provide sufficient, explicit information inRA5b about how the individual’s particularcircumstances have adversely affected theircontribution to enable panels to apply theirpublished criteria. HEIs need not describecircumstances (for example, a disability) that havehad no adverse effect on an individual’s capacityto undertake research.

162. RA5b will have a length limit perindividual. We will decide this in the context ofthe criteria statements drawn up by the panelsand publish it with the panels’ final publishedcriteria statements.

Category C staff circumstances RA5c

163. For each individual returned as CategoryC, information is required in RA5cdemonstrating that their research is clearly anddemonstrably focussed in the department. Sub-panel criteria statements will give examples of thetypes of evidence to be supplied. If a sub-panel isnot convinced by the evidence provided, it maytake account of this in assessing that individual’scontribution to the department’s research. LikeRA5b, RA5c will have a length limit perindividual which we will publish alongside thepanels’ final published criteria statements. Theinformation in RA5c will be published (seeparagraph 33).

30 RAE 03/2005

Notes to Tables 1 and 2

1. Sub-panels will use their professionaljudgement to form a view about the qualityprofile of the research activity described in eachsubmission, taking into account all the evidencepresented. Their recommendations will beendorsed by the main panel in consultation withthe sub-panel.

2. ‘World-leading’ quality denotes an absolutestandard of quality in each unit of assessment.

3. ‘World leading’, ‘internationally’ and‘nationally’ in this context refer to qualitystandards. They do not refer to the nature or

geographical scope of particular subjects, nor tothe locus of research nor its place ofdissemination, for example, in the case of‘nationally’, to work that is disseminated in theUnited Kingdom of Great Britain and NorthernIreland.

4. The profile for a submission that contains noresearch which meets the one star threshold willbe 100 per cent unclassified. A submission thatcontains no research (that is, no work that meetsthe definition of research for the RAE research)will not be awarded a quality profile.

Annex AQuality profiles and definitions of quality levels

RAE 03/2005 31

Table 1 Sample quality profile*

Unit of FTE Category A Percentage of research activity in the submissionassessment A staff submitted judged to meet the standard for:

for assessment

four star three star two star one star unclassified

University X 50 15 25 40 15 5

University Y 20 0 5 40 45 10

* The figures are for fictional universities. They do not indicate expected proportions.

Table 2 Definitions of quality levels

Four star Quality that is world-leading in terms of originality, significance and rigour.

Three star Quality that is internationally excellent in terms of originality, significance and rigour but which

nonetheless falls short of the highest standards of excellence.

Two star Quality that is recognised internationally in terms of originality, significance and rigour.

One star Quality that is recognised nationally in terms of originality, significance and rigour.

Unclassified Quality that falls below the standard of nationally recognised work. Or work which does not meet the

published definition of research for the purposes of this assessment.

Notes to Figure 1

1. Panels are required to consider all thecomponents of the submission when reaching anoverall quality profile. The components equate tothe different data collected in the RAE, namelysubmitted staff information (RA1), research outputs(RA2), research student data (RA3), researchincome (RA4), supporting statement on researchenvironment and esteem indicators (RA5a).

2. These different components will be assessedunder three overarching elements: researchoutputs, research environment, and esteemindicators. Research outputs (RA2) will always beassessed as one of these three elements.

3. Main panels will decide whether thecomponents of submissions other than researchoutputs (RA3, 4 and 5) will be assessed under the‘Research environment’ or ‘Esteem indicators’element. For example, a panel may consider thatresearch income contributes to researchenvironment, or that it is a measure of esteem inits subject area. Similarly research studentnumbers, research student completions and

research studentships may either be part of theresearch environment or an indicator of esteem.Main panels will explain in their statements ofcriteria and working methods their reasoning forassigning components of the submission to aparticular element.

4. Main panels will allocate a percentage weightingto each of three elements – research outputs,research environment and esteem indicators –which will indicate the extent to which the differentelements will contribute to the overall qualityprofile of a submission. Given the primacy of expertreview in the process, the weighting allocated toresearch outputs must be at least 50 per cent of theoverall quality profile: some main panels mayreasonably decide that research outputs should beweighted more highly. Main panels must allocate asignificant weighting to each of the other aspects,environment and esteem, as they see fit, but sincethe quality profile will be defined in multiples of 5 per cent, the minimum weighting in either casewill be 5 per cent. Main panels must define theirreasoning in every case in their criteria statements.

32 RAE 03/2005

The percentage weightings to the three elements are illustrative. Panels should allocate these. The minimum weighting

for the research outputs profile is 50%. In this example the overall quality profile shows 15% of research activity is at

4* level. This is made up of 70% x 10 (research outputs), 20% x 20 (research environment) and 10% x 30 (esteem

indicators), rounded as described in paragraphs 8-11 below.

Figure 1 Building a quality profile

Overall quality profile

Quality level 4* 3* 2* 1* u/c

% of research 15 25 30 20 10activity

eg 20% (Minimum 5%) eg 10% (Minimum 5%)eg 70% (Minimum 50%)

The overall quality profilecomprises the aggregateof the weighted profilesproduced for researchoutputs, researchenvironment and esteemindicators

Research outputs

4* 3* 2* 1* u/c

10 25 40 15 10

Research environment

4* 3* 2* 1* u/c

20 30 15 20 15

Esteem indicators

4* 3* 2* 1* u/c

30 25 10 20 15

RAE 03/2005 33

5. Sub-panels will assess research outputs anddevelop a quality profile for this element. Sub-panels will also assess the evidence within thecomponents of the submission assigned to theresearch environment and esteem indicatorselements and draw up a quality profile for each.

6. Sub-panels will sum the three weighted qualityprofiles to develop an overall quality profile forthe submission. They will use the roundingmethodology described in paragraphs 8-10 of thisAnnex to round the overall quality profile. Overallquality profiles will be published in steps of 5 percent.

7. Sub-panels will finally confirm that, in theirexpert judgement, the overall profile is a fairreflection of the research activity in that submission,and that their assessment has taken account of allthe different components of the submission.

Rounding

8. All sub-panels will adopt a cumulativerounding methodology to ensure that the overallquality profile for any submission will alwaysround to 100 per cent and to avoid the unfairconsequences that simple rounding can produce.They will first sum the weighted quality profilesfor outputs, environment and esteem and thenadopt a cumulative rounding methodology.

Worked example

9. Using the example in Figure 1, first calculatethe initial overall profile, that is, the sum of theweighted profiles for outputs, environment andesteem.

4* 3* 2* 1* u/c

Outputs 10 25 40 15 10

Environment 20 30 15 20 15

Esteem 30 25 10 20 15

Weighted

70% 7 17.5 28 10.5 7

20% 4 6 3 4 3

10% 3 2.5 1 2 1.5

Initial profile 14 26 32 16.5 11.5

10. Cumulative rounding works in three stages:

a. The initial profile is:

4* 3* 2* 1* u/c

14 26 32 16.5 11.5

b. Stage 1: Calculate the cumulative totals (forexample the cumulative total at 3* or betteris 26+14=40)

4* 3* or 2* or 1* or u/c or

better better better better

14 40 72 88.5 100

c. Stage 2: Round these to the nearest 5 percent, (rounding up if the percentage ends inexactly 2.5 or 7.5)

4* 3* or 2* or 1* or u/c or

better better better better

15 40 70 90 100

d. Stage 3: Find the differences betweensuccessive cells to give the rounded profile.So, for example, the percentage allocated to2* is the difference between the cumulativetotal at 2* or better, minus the cumulativetotal at 3* or better (70-40 =30).

4* 3* 2* 1* u/c

15 25 30 20 10

11. Cumulating the totals other way (roundingdown if the percentage ends in exactly 2.5 or 7.5)gives exactly the same answer.

(Changes in phrasing from the definition used forthe 2001 RAE are in bold.)

‘Research’ for the purpose of the RAE is to beunderstood as original investigation undertaken inorder to gain knowledge and understanding. Itincludes work of direct relevance to the needs ofcommerce, industry, and to the public andvoluntary sectors; scholarship*; the invention andgeneration of ideas, images, performances,artefacts including design, where these lead tonew or substantially improved insights; and theuse of existing knowledge in experimentaldevelopment to produce new or substantiallyimproved materials, devices, products andprocesses, including design and construction. Itexcludes routine testing and routine analysis ofmaterials, components and processes such as forthe maintenance of national standards, as distinctfrom the development of new analyticaltechniques. It also excludes the development ofteaching materials that do not embody originalresearch.

* Scholarship for the RAE is defined as the creation,development and maintenance of the intellectualinfrastructure of subjects and disciplines, in forms suchas dictionaries, scholarly editions, catalogues andcontributions to major research databases.

Annex BDefinition of research for the RAE

34 RAE 03/2005

RAE 2008 Units of assessment and main panels

Main panel UOA UOA name

A 1 Cardiovascular Medicine

2 Cancer Studies

3 Infection and Immunology

4 Other Hospital Based Clinical Subjects

5 Other Laboratory Based Clinical Subjects

B 6 Epidemiology and Public Health

7 Health Services Research

8 Primary Care and Other Community Based Clinical Subjects

9 Psychiatry, Neuroscience and Clinical Psychology

C 10 Dentistry

11 Nursing and Midwifery

12 Allied Health Professions and Studies

13 Pharmacy

D 14 Biological Sciences

15 Pre-clinical and Human Biological Sciences

16 Agriculture, Veterinary and Food Science

E 17 Earth Systems and Environmental Sciences

18 Chemistry

19 Physics

F 20 Pure Mathematics

21 Applied Mathematics

22 Statistics and Operational Research

23 Computer Science and Informatics

G 24 Electrical and Electronic Engineering

25 General Engineering and Mineral & Mining Engineering

26 Chemical Engineering

27 Civil Engineering

28 Mechanical, Aeronautical and Manufacturing Engineering

29 Metallurgy and Materials

H 30 Architecture and the Built Environment

31 Town and Country Planning

32 Geography and Environmental Studies

33 Archaeology

Annex CRAE 2008 Units of assessment

RAE 03/2005 35

Main panel UOA UOA name

I 34 Economics and Econometrics

35 Accounting and Finance

36 Business and Management Studies

37 Library and Information Management

J 38 Law

39 Politics and International Studies

40 Social Work and Social Policy & Administration

41 Sociology

42 Anthropology

43 Development Studies

K 44 Psychology

45 Education

46 Sports-Related Studies

L 47 American Studies and Anglophone Area Studies

48 Middle Eastern and African Studies

49 Asian Studies

50 European Studies

M 51 Russian, Slavonic and East European Languages

52 French

53 German, Dutch and Scandinavian Languages

54 Italian

55 Iberian and Latin American Languages

56 Celtic Studies

57 English Language and Literature

58 Linguistics

N 59 Classics, Ancient History, Byzantine and Modern Greek Studies

60 Philosophy

61 Theology, Divinity and Religious Studies

62 History

O 63 Art and Design

64 History of Art, Architecture and Design

65 Drama, Dance and Performing Arts

66 Communication, Cultural and Media Studies

67 Music

36 RAE 03/2005

January 2005 Guidance to panels issued

May 2005 Panel membership published

June 2005 Guidance on submissions issued

Summer 2005 Draft criteria and working methods of main panels and sub-panels issued forconsultation (16 July to 19 September 2005)

January 2006* Final criteria and working methods of main panels and sub-panels issued

August 2006 Issue pilot software and draft manuals

December 2006 Issue final software, final manuals and invitation to HEIs to make submissions

Spring 2007 Survey of HEIs submission intentions

31 July 2007 End of assessment period for research income and research student data

31 October 2007 Census date

30 November 2007 Closing date for submissions

31 December 2007 End of publication period (cut-off point for publication of research outputs)

December 2008 Results published

* In the timetable at Annex E of ‘RAE 01/2005 Guidance to panels’, the RAE team published its intention topublish final panel criteria and working methods by the end of 2005. However, in view of our decision to extend bytwo weeks the period of consultation on draft criteria to enable HEIs and subject associations to make consideredresponses, we now expect to publish final versions in January 2006.

Timetable of panel meetings

Round one January-March 2005 Criteria-setting (drafting)

Round two March-May 2005 Criteria-setting (drafting)

Round three October/November 2005 Criteria-setting (final)

Round four June/July 2007 Consideration of submission intentions

Rounds five to eight January-November 2008 Assessment phase

Each round of meetings will comprise one meeting of each main panel and sub-panel normally lastingone or two days. Sub-panels normally meet before main panels, except in round one when main panelsmet first.

Annex DTimetable for RAE 2008

RAE 03/2005 37

This Annex provides a summary of the data requirements and points to relevant paragraphs in thisdocument that describe the data requirements and definitions.

Summary of data requirements RA Form Paragraphs

FTE of submitted category A research active staff RA0 65g.i, 67a

FTE of research assistants RA0 67b, 69-71

FTE of research support staff (technicians, scientific officers, RA0 67c, 69, 72-73experimental officers, other support staff devoted to research)

Contractual and other data on submitted research active in RA1 74-91dividuals by category A-D

Designation of Category A and C staff as research fellow as RA1 75h, 80-82defined for RAE 2008

Data about a maximum of 4 research outputs produced by each RA2 92-104submitted Category A and C staff individual

Headcount of full-time research students in each year 2001 to 2007 RA3a 105a, 106-121

Headcount and FTE of part-time research students in each year RA3a 65g.ii, 105b, 106-1212001 to 2007

Number of research doctoral degrees awarded in each year RA3a 105c, 122 -1232001 to 2007

Number of research masters degrees awarded in each year RA3a 105d, 122-1232001 to 2007

Number of new studentships, by source of funding, awarded for RA3b 124-129research-based higher degrees in each year 2001 to 2007

External research income expended in each financial year RA4 130-1532001 to 2007, by source

Information describing the research environment and evidence of esteem RA5a 154-158*

Information about individual staff circumstances that have adversely RA5b 159-162affected their contribution to submissions

Evidence for the inclusion of Category C staff – that their research is RA5c 163focussed in the submitting department

* Panel statements of criteria and working methods will specify a structure for RA5a

Annex E Summary of data requirements

38 RAE 03/2005

Standard analyses provided to panels for each submission

1. Total number of research-active* staff by category (headcount)

2. Total number of research fellows

3. Total number of research assistants who are supervised by research-active staff

4. Research assistants supervised per research-active staff

5. Total number of research students who are supervised by research-active staff

6. Research students supervised per research-active staff

7. Total number of outputs listed in RA2 for assessment

8. Numbers of staff with 0, 1, 2, 3 or 4 research outputs (five separate totals)

9. Total number of studentships (listed by sponsor)

10. Studentships (listed by sponsor) per research-active staff

11. Studentships per research student

12. Total research income (listed by source of income)

13. Research income (listed by source of income) per research-active staff

Standard analyses provided to panels for each UOA

14. Total number of research-active staff

15. Total number of staff by category (headcount)

16. Average number of research assistants supervised per research-active staff

17. Average number of research students supervised per research-active staff

18. Total number of outputs listed

19. Average number of research degrees awarded per annum (by each category: doctoral and masters)

20. Average number of studentships (listed by sponsor) per research-active staff

21. Average number of studentships (listed by sponsor) per research student

22. Average research income (listed by source of income) per research-active staff.

* In all cases ‘research-active’ refers to those selected for submission

Numbers of staff, assistants and students will be expressed as full-time equivalents in the analyses, unlessotherwise specified.

In all cases, analyses per research-active staff will be presented as two separate totals: one total perCategory A staff and one total per Category A staff + Category C staff.

Annex FStandard data analyses

RAE 03/2005 39

Introduction1. It is recognised that many institutions havealready been working towards making RAE 2008submissions, although obviously we do not expectfinal decisions have been made. The purpose ofthis guidance is to aid institutions in ensuring thatthe maximum number of staff who are conductingexcellent research have their work included insubmissions. The intention is that it shall be usedto inform each institution’s internal equality RAEcode of practice, which should also refer to thatinstitution’s existing equal opportunities policies.4

The legislative context

Funding bodies’ legal responsibilities

2. Under the Race Relations (Amendment) Act(RRAA), the higher education funding bodies inEngland, Wales and Scotland have a statutoryduty to monitor the HE sector for any adverseand differential impact of a funding bodysponsored policy or service on a minority ethniccommunity. They also have a duty to monitor, byracial group, the numbers of staff in the HEIsthey are responsible for, and to publish annually,as far as possible, the results of that monitoring.As part of their responsibility under the RRAA,they also need to monitor, by racial group, thestaff submitted to the RAE5.

3. As similar duties will be in force for disabilityand are expected for gender before RAE 2008 isactually carried out, the funding bodies inEngland, Wales and Scotland will have similarobligations to monitor in those areas, includingmonitoring the RAE returns.

HEIs’ legal responsibilities

4. HEIs likewise have a general duty and specificduties under the RRAA. As well as having a dutyto monitor submissions to the RAE by racial

group, institutions’ internal selection processeswill have to be assessed for their impact ondifferent racial groups. Institutions will havesimilar responsibilities in respect of gender anddisability before the RAE in 2008.

5. In addition, under the Fixed-term and Part-time Employees Regulations, fixed-term and part-time employees have the right not to be treated byan employer any less favourably than theemployer treats a comparable ‘permanent’employee.

Summary of legislation

6. A summary of the equal opportunitieslegislation with which institutions have to complygenerally, and which they should take intoaccount when preparing their RAE 2008submissions is included in this Annex. ‘Equalitybriefing for panel chairs, members and secretaries’(RAE 02/2005) advised members of RAE panelsabout this legislation and instructed them todevelop working methods and assessment criteriathat encourage HEIs to submit the work of all oftheir excellent researchers, including those whosevolume of research output may have been limitedfor reasons covered by equal opportunitieslegislation. Panel statements of criteria andworking methods give guidance for submittinginstitutions about each panel’s treatment ofdifferent individual circumstances.

RAE requirement for a code of practice

7. It is a requirement of the RAE that eachsubmitting institution establishes a code ofpractice. Clearly managerial decisions on which oftheir staff to submit are at the discretion of HEIsbut they need to be defensible. The fundingbodies will require the head of each HEI makingan RAE submission to confirm that the HEI has– in preparing its submissions and selecting staff

Annex GGuidance from the Equality Challenge Unit on drawing up a code of practice on preparing RAE submissions

40 RAE 03/2005

4 In the guidance that follows reference is made to the legal frameworks in England, Scotland and Wales. It isexpected, however, that institutions in Northern Ireland will also produce codes of practice of the kind set out below.

5 HEIs in Northern Ireland and the Department of Education and Learning Northern Ireland have responsibilitiesunder the Northern Ireland Act 1998HEFCE, HEFCW and SHEFC as well as all English, Welsh and Scottish HEIs have responsibilities under the RaceRelations (Amendment) Act 2000All higher education funding bodies and HEIs will have responsibilities under the Disability Discrimination Act 2005.

for inclusion – developed, adopted anddocumented an appropriate internal code ofpractice, which attends to all relevant equalopportunities legislation in force on thesubmission date. HEIs may be required to submittheir code of practice for verification but they willnot be routinely required to submit their codewith their RAE submission.

8. Given the increasingly stringent requirementsof the law and the risk of challenge, which mark asignificant change from the context in whichprevious RAEs were carried out, it is inevitablethat HEIs would have to establish formal andtransparent policies and procedures for handlingthe RAE in order to discharge their own legalresponsibilities. The RAE code of practicerequired by the funding bodies, so that their legalobligations can be met, thus provides a commonframework for the development of policies andprocedural statements within the exercise, butdoes not generate obligations beyond those thatHEIs will in any case need to address.

9. We have developed the following guidancewith the ECU. It is intended to assist HEIs indrawing up a code of practice that frames theirdecision-making processes in relation to the 2008RAE in the context of the principle of equality ofopportunity and all relevant legislation. For thoseHEIs that do not already have in place an equalopportunities code governing their RAEpreparations, it offers suggestions on proceduresthey might consider including.

Guidance to institutions

The basic principles

10. Each institution’s code of practice shouldmake all the processes concerned with selection ofstaff for inclusion in RAE 2008 submissionstransparent. It is essential that practice in respectof submissions is consistent across the institution.As decisions regarding the exercise will be made atdifferent levels within institutions, the code ofpractice should set out the principles to beapplied to all aspects/stages of the processes.Responsibilities should be clearly defined and theoperating criteria and terms of reference for

individuals, committees, advisory groups and anyother bodies concerned with the RAE should bemade readily available to all individuals andgroups concerned. Also, we recommend that theyare published on the staff intranet and that theexistence of the code is well publicised throughoutthe institution, with copies being made freelyavailable.

Senior management

11. It will commonly be the case that majorresponsibility for conducting aspects of the RAEprocess and making key decisions will be carriedout by existing designated senior officers of theinstitution (such as pro-vice chancellors, deansand heads of school). The procedures andprocesses for the selection of the designated seniormember of staff (even if, in this case, it is a seniorofficer such as the pro-vice chancellor [research])should be described in the code of practice.

12. The equal opportunities training that thisdesignated person will undertake, or hasundertaken, or the level of understanding of theissues they will be required to attain should beincluded in the code.

13. Clear definitions of each person’s role withinthe process must be provided.

Committees

14. The structural differences within HEIs meanthat the method of developing submissions will notbe uniform across the sector. Where a committee orcommittees have designated RAE responsibilities,whether it is at departmental, faculty, UOA orcentral level, these should be detailed in the code ofpractice, including, for each committee:

• how the committee has been formed

• its membership

• the definition of its position within theadvisory or decision-making process

• the steps taken to ensure that members arewell informed about their own and theinstitution’s legal obligations regarding equalopportunities.

RAE 03/2005 41

15. The following details should be providedabout its mode of operation:

• the criteria that it will use in carrying out itsfunctions

• the method by which these criteria arecommunicated

• the method and timescale in which feedbackwill be provided in respect of the decisionsmade.

16. Where faculties/UOAs consider reports fromdepartmental RAE committees or designated staff,the parameters of the discussion must be clear,and records must be kept. When individualperformance is discussed and the individual isabsent, committees should be made fully aware ofall the facts relating to the individual.

Guidelines for UOA coordinators,faculties and departments

17. The code of practice should includeguidelines for UOA coordinators, faculties anddepartments on selection and submission. Theguidance should include definitions of theinstitution’s policies and procedures for:

• appointing an external advisor, if required

• the transparent selection of staff who areresponsible for deciding which work is to besubmitted

• selecting staff for submission

• communicating the process to staff

• offering feedback.

18. In the light of these guidelines, each unit ofassessment should produce a statement of intentgiving information about how it will carry out itsselection and submission process. This mayrequire processes to be established at facultyand/or departmental level. Staff should beconsulted on the statement.

19. Where decisions are made by a single UOAleader, the code should set out a procedure forchecking for consistency or verifying that thesedecisions are in keeping with the institution’spolicy or benchmark for selection.

Impact assessment

20. Each HEI should also provide an equalityprofile – in terms of disability, gender andethnicity – of staff who are eligible for submissionand indicate those who are submitted and thosewho are not. If any prima facie imbalance isfound relative to the total potential pool, then theHEI may be required to account for it. HEIs mayconsider undertaking impact assessments at lowerlevels, for example if they perceive imbalance inparticular UOAs.

Appeals

21. The code of practice should include detailson how appeals or complaints will be handled ifthey are on the grounds of potentialdiscrimination. These could be a personalcharacteristic (such as gender or disability) or awork pattern/absence that has not been fullytaken into account. This could be done throughreference to the institution’s existing complaintreporting mechanisms, if they are thought to beappropriate for this purpose.

Personal circumstances

22. Main panels and sub-panels have producedguidance on how they will deal with individualcircumstances that might have had an effect on anindividual’s contribution to a submission. Whilethere will be some variation between UOAs, theinstitution’s code of practice must clearly drawattention to the fact that each panel’s guidancedocument has an equal opportunities statement.

23. In common with the panel guidance, theinstitutional code of practice must also includethe standard list of circumstances that theinstitution and the panels will take into account.These are:

• absence due to maternity/adoption leave

• women returning to part-time work aftermaternity leave in the period 1 January2001 to 31 October 2007

• adoptive parents returning to part-time workafter adoptive leave in the period 1 January2001 to 31 October 2007

42 RAE 03/2005

• part-time work

• disability, including temporary incapacitythat lasts for at least 12 months. Institutionsare advised that the Disability DiscriminationAct 2005 will cover people with cancer,Multiple Sclerosis and HIV from the point ofdiagnosis from October 2006

• young staff who entered the professionduring the period 1 January 2001 to 31October 2007

• mature staff who were new entrants to theprofession during the period 1 January 2001to 31 October 2007

• absence due to ill-health or injury.

24. When making submissions to UOAs that arelaboratory-based, HEIs should be mindful ofhealth and safety restrictions imposed on pregnantand breastfeeding women which may haveprevented them from undertaking some types ofresearch during the period 1 January 2001 to 31October 2007. Similar consideration should begiven to the restrictions that pregnancy andnursing might place on fieldwork.

25. Institutions will also need to develop policiesin respect of absences for sabbaticals, career breaksand secondments.

26. The code of practice must define whichabsences, and for what duration, will beaccounted for in submissions, and theinstitutional approach to them. This must bestandard across all departments.

Fixed-term and part-time staff

27. In the light of the Fixed-term and Part-timeRegulations, consideration will also need to begiven to the way in which the institution candemonstrate the implementation of equality ofopportunity for those on fixed-term and part-timecontracts.

28. The code of practice should therefore includea statement about how the institution supports itsfixed-term and part-time staff, including contractresearch staff.

Joint submissions

29. HEIs making joint submissions may wish tomake their code of practice available tocollaborating HEIs. In any case, they shouldensure that joint decision-making across HEIsdoes not compromise their adherence to theirrespective codes of practice.

Mock exercises

30. HEIs that conduct mock RAEs mightconsider using the mock exercise as anopportunity to apply their draft code and refine itthereafter.

Further information31. Further information, including the text oflegislation, can be accessed through the EqualityChallenge Unit’s web-site at www.ecu.ac.uk.Specific queries should, however, be addressed tothe RAE team in the first instance.

RAE 03/2005 43

Direct Indirect discriminationdiscrimination (The wording in respect of indirect discrimination is taken

directly from the relevant legislation.)

Age Forthcoming in 2006

Disability Occurs when a disabled person is treated less favourably for a reason relating to his/herDisability disability than someone of whom that reason does not or would not apply.DiscriminationAct 1995

Race* A person is treated less 1. A person discriminates against another if he applies to Race Relations favourably than other that other a requirement or condition which he applies orAct 1976 people on racial grounds. would apply equally to persons not of the same racial group

as that other but:

(i) which is such that the proportion of persons of the sameracial group as that other who can comply with it isconsiderably smaller than the proportion of persons not ofthat racial group who can comply with it; and

(ii) which he cannot show to be justifiable irrespective of thecolour, race, nationality or ethnic or national origins of theperson to whom it is applied; and

(iii) which is to the detriment of that other because hecannot comply with it.

2. A person discriminates against another if he applies aprovision, criterion or practice which he applies or wouldapply equally to persons not of the same race or ethnic ornational origins to that other person:

i) which puts or would put persons of the same race orethnic or national origins as that other at a particulardisadvantage when compared with other persons,

ii) which puts that other at that disadvantage, and

iii) which he cannot show to be a proportionate means ofachieving a legitimate aim.

Religion or A person is treated less A applies to B a provision, criterion or practice which he belief favourably than other applies or would apply equally to persons not of the same Employment people on grounds of religion or belief as B, but:Equality his or her religion or i) which puts or would put persons of the same religion or (Religion or belief. belief as B at a particular disadvantage when compared with Belief ) other persons,Regulations

ii) which puts B at that disadvantage, and2003iii) which A cannot show to be a proportionate means ofachieving a legitimate aim.

Table 3 Summary of equality legislation

44 RAE 03/2005

Sex** A man or woman is 1. A person discriminates against a woman [or a man] if he Sex discriminated against applies to her [or him] a requirement or condition which he Discrimination when he or she is treated applies or would apply equally to a man [or woman] but:Act 1975 less favourably than a i) which is such that the proportion of women [or men]

woman or man would who can comply with it is considerably smaller than the be, respectively. proportion of men [or women] who can comply with it, and

ii) which he cannot show to be justifiable irrespective of thesex of the person to whom it is applied, and

iii) which is to her [or his] detriment because she [or he]cannot comply with it.

2. A person discriminates against a woman [or a man] if heapplies to her [or him] a provision, criterion or practicewhich he applies or would apply equally to a man [or awoman], but

i) which is such that it would be to the detriment of aconsiderably larger proportion of women than of men, [ormen than of women] and

ii) which he cannot show to be justifiable irrespective of thesex of the person to whom it is applied, and

iii) which is to her [or his] detriment.

Sexual A person is treated less A applies to B a provision, criterion or practice which he orientation favourably than other applies or would apply equally to persons not of the same

people on grounds of sexual orientation as B, but:his or her, actual or i) which puts or would put persons of the same sexual perceived, sexual orientation as B at a particular disadvantage whenorientation. compared with other persons,

ii) which puts B at that disadvantage, and

iii) which A cannot show to be a proportionate means ofachieving a legitimate aim.

* The Race Relations Act 1976 was amended by the EU Race Directive in 2003. This has resulted in an anomaly inthat the amendment refers to persons of the same race or ethnic or national origins, while the original Act, which hasnot been repealed, refers to persons of the same colour or nationality.

** Wording has been added, denoted by square brackets, which does not appear in the Act. This is to make explicitthe fact that the Act covers men as well as women.

Table 3 (continued)

RAE 03/2005 45

EmploymentEquality(SexualOrientation)Regulations2003EmploymentEquality(SexualOrientation)Regulations(NorthernIreland) 2003

The following publications give relevant further information about the 2008 RAE: titles of forthcomingpublications are provisional. All are available (or will be when published) at www.rae.ac.uk:

• RAE 01/2004 ‘Initial decisions by the UK funding bodies’

• RAE 03/2004 ‘Units of assessment and recruitment of panel members’

• RAE 01/2005 ‘Guidance to panels’

• RAE 02/2005 ‘Equality briefing for panel chairs, members and secretaries’

• Panel members for the 2008 RAE (web only)

• RAE data collection – submission software XML schemes (web only)

• draft statements of main and sub-panel criteria and working methods (to be published indraft form – web only – by July 2005)

• final statements of main and sub-panel criteria and working methods (to be published byJanuary 2006)

• guidance on completion of RAE submission software (to be published by August 2006).

Annex H Related and forthcoming publications

46 RAE 03/2005

Publication period The period during which research outputs must be placed in the publicdomain (or in the case of confidential outputs, lodged with the sponsor) ifthey are to qualify for assessment in RAE 2008. The publication periodruns from 1 January 2001 to 31 December 2007.

Census date The date determining the affiliation of research staff to a particularinstitution. (Category A staff may be submitted to the RAE by the employinginstitution on this date, regardless of previous or forthcoming changes in theiremployment status.) The census date will be 31 October 2007.

Department The staff included in a submission to one of the 67 discrete units ofassessment recognised by the RAE, and, by extension, their work and thestructures which support it. RAE departments are often not identified witha single administrative unit within an HEI.

Eligible staff Academic staff whose primary employment function is research, teaching orboth and who can be shown to have undertaken significant autonomousresearch, or otherwise to have made a significant independent researchcontribution to the research output of a unit or department.

Expert review Assessment of outputs by experts in the discipline. Most will be activeresearchers but, in some cases, research users who are not currently activeresearchers but are expert users will participate in the assessment.

Financial year For RAE purposes a financial year is 1 August to 31 July (not 1 April to 31 March).

Funding bodies The four UK funding bodies for higher education: the Higher EducationFunding Councils for England, Scotland and Wales, and the Departmentfor Employment and Learning, Northern Ireland.

Joint submission The collaborative work of two or more institutions in a single subject area,submitted in the form of a unified entity to a single unit of assessment.

Multiple submission More than one, discrete submission made by one HEI to one unit ofassessment. Each of the discrete submissions describes an academically andstructurally distinct body of research.

Panel Generic term covering both main panels and sub-panels in RAE 2008.

Quality profile The results of each submission’s assessment – replacing the rating used inprevious RAEs. The quality profile shows the proportion of overall researchactivity described in a submission that meets each of four defined levels ofquality (one, two, three and four star) and the proportion that is unclassified.

Research activity The totality of the research and research-related activities reported in asubmission. Research activity includes the conduct, management anddissemination of research.

Research output The outcome of a research process, presented in the public domain (or inthe case of confidential outputs, lodged with the sponsor).

Submission The complete set of information provided to the RAE by a departmentwithin a unit of assessment.

Unit of assessment (UOA) One of 67 discipline areas to which RAE submissions may be made byinstitutions.

Annex IGlossary of terms

RAE 03/2005 47

FTE Full-time equivalent

HE Higher education

HEI Higher education institution

HEFCE Higher Education Funding Council for England

HESA Higher Education Statistics Agency

OST Office of Science and Technology

RA Research Assessment

RA review Sir Gareth Roberts’ Research Assessment review

RAE Research Assessment Exercise

u/c Unclassified (quality level)

UOA Unit of assessment

48 RAE 03/2005

Annex JList of abbreviations

Northavon House, Coldharbour Lane, BRISTOL BS16 1QD

www.rae.ac.uk