AU-RSC-20-3965-A

Assessment Regulations

2021-22

Applicable to students in all Stages/Years of programmes

AU-RSC-20-3965-A 2021/22

2

CONTENTS Regulation Summary ............................................................................................................ 3

Related Regulations, Policies, and Guidance ...................................................................... 3

Version Control .................................................................................................................... 3

1 MARKING POLICY ........................................................................................................... 4

Assessment Definitions ........................................................................................................ 4

Assessments and Marking Schemes ................................................................................... 4

Approval and re-use of examination questions .................................................................... 4

Maintaining Standards .......................................................................................................... 4

Moderation and Anonymous Marking ................................................................................... 5

2 WORD COUNTS AND LIMITS .......................................................................................... 5

3 DEFINITIONS OF MARKING PRACTICES ...................................................................... 5

Blind double-marking............................................................................................................ 5

Informed double-marking ..................................................................................................... 6

Moderation ........................................................................................................................... 6

Moderation: Arithmetic checking .......................................................................................... 6

Moderation: Sampling .......................................................................................................... 7

Choosing a sample............................................................................................................... 7

Rounding of marks ............................................................................................................... 7

4 GENERIC MARKING SCALES ......................................................................................... 9

5 LATE SUBMISSION OF WORK AND PENALTIES......................................................... 10

Worked examples:.............................................................................................................. 11

6 RECYCLING OF ASSESSMENT CONTENT .................................................................. 13

Worked examples:.............................................................................................................. 13

7 USE OF TEXT-MATCHING SOFTWARE ....................................................................... 15

Guiding principles ............................................................................................................... 15

Policy .................................................................................................................................. 15

APPENDIX 1: GUIDANCE FOR MODERATION OF MARKS BY MODULE BOARDS ......... 17

AU-RSC-20-3965-A 2021/22

3

Regulation Summary These regulations set out the University’s approach to the marking and moderation of assessment, including: ▪ Baseline University standards and expectations for the availability of assessment

information to students, and the setting of assessment questions.

▪ The University’s approach to inclusive marking and equitable assessment.

▪ Guidance on the use of word counts, using either a single figure or a range.

▪ University definitions and approaches to moderation, including Informed and Blind Double-Marking, Arithmetic Checking, Sampling, and the Rounding of Marks.

▪ A generic marking scale for undergraduate-level assessment, to be used as a reference point for College and Programme-level guidance.

▪ The University’s approach to the use of Turnitin as text-matching software. These regulations also set out procedures for applying penalties to assessments that are submitted beyond the set deadline, or contain content submitted by the same student during a previous assessment. Related Regulations, Policies, and Guidance Supplementary information can also be found in Plagiarism: Notes of Guidance for Markers and Invigilation Procedures for Examinations. Information on Assessing the Work of Students with Specific Learning Difficulties is also available. Further information for candidates can be found in the University’s Examination Regulations for Candidates.

Version Control

Ref No Version Responsible Officer Approved by Approval Date

Effective Date

AU-RSC-18-1860 A Adam Hewitt OLTC Jun 2019 Sep 2019

AU-RSC-19-2814 D Adam Hewitt ULTC Oct 2020 Nov 2020

AU-RSC-20-3965 A Adam Hewitt ULTC Jun 2021 Sep 2021

AU-RSC-20-3965-A 2021/22

4

1 MARKING POLICY Assessment Definitions Summative Assessment: Any formally assessed activity, the outcome of which contributes to the overall module mark and/or module pass requirements. All summative assessment elements will be clearly identified in the Module Specification. Formative Assessment: Formal or informal learning activities that enable a student to receive feedback on progress towards meeting the module learning outcomes. Formative assessment marks do not contribute to the overall module mark and/or module pass requirements.

Assessments and Marking Schemes 1 Colleges should ensure that Programme and Module Specifications (including

assessment weightings) are available to students at the beginning of each Stage of

the programme.

2 Examination and coursework questions should appropriately reflect the Level of the

credit to be assessed, and are expected to differ according to Level, either in the

assessment question set, or in the mark scheme to be used.

3 Changes to published assessments which have a retrospective effect on current

students must not be introduced without full consultation with all students affected.

Approval and re-use of examination questions 1. Specific disciplines may use assessment standard setting techniques which require the

reuse of a proportion of examination questions for each examination paper. However, in

other disciplines, and for other examinations, the re-use of examination questions is poor

practice and has the potential to lead to an unfair student experience. The policy around

re-use of examination questions for each programme should be determined by the

Programme Director and, for Stages contributing to the degree classification, the External

Examiner(s).The policy should be clearly communicated to all those involved in the

programme’s assessment design.

2. Examination papers, initial and resit, together with their marking schemes, must be

formally approved by a Scrutiny Process prior to their issue to students, normally by an

academic panel, the membership of which is defined by the College owning the module.

Maintaining Standards It is essential that Programmes differentiate between expectations of work presentation in the marking of examination scripts and written assignments, and to communicate this clearly in assessment briefs and marking schemes. Spelling, punctuation and grammar accuracy is not expected in examination scripts produced under timed conditions unless this is deemed a core competence to be demonstrated under these circumstances. Minor errors in examination script presentation should not be penalised if the meaning of the student’s

AU-RSC-20-3965-A 2021/22

5

response is clear. This excludes critical spellings such as drug names and medical terminology. Scripts must however be legible to be assessed. Where accuracy of spelling, punctuation and grammar is being assessed in written assignments, assessment briefs should direct students to the inclusive ‘Tools for Learning’ resource. Moderation and Anonymous Marking

4 All formal written examination scripts administered by Student and Academic Services

are to be marked anonymously.

5 A minimum of 20% of final Stage/Master’s written examination scripts will be

moderated.

6 A minimum of 20% of final Stage/Master’s coursework which contributes a significant

percentage of marks to the overall module mark (i.e. 30% or more), or which assess a

crucial aspect of the module, will be moderated.

7 All substantial final Stage projects (i.e. projects carrying 30 credits or more) and

Master’s dissertations will be blind double-marked.

8 Arithmetic checking of examination marks should always occur to ensure that marks

from all questions are included in the total entered on the marksheet or student record.

2 WORD COUNTS AND LIMITS 1 Assessments that specify a word count should express this as either a single figure

and/or a range.

2 Assessments that specify a word count, either as a limit or range, should clearly

articulate what is and is not considered part of the word count.

3 When an assessment is determined to have exceeded the designated maximum word

count by more than 10%, this will be taken into account during the marking process.

4 There is no additional penalty for assessments that fall below a specified word limit or

range.

3 DEFINITIONS OF MARKING PRACTICES All markers should have access to mark schemes and model answers where appropriate.

Blind double-marking Blind double-marking is the marking of an assessment by two separate markers, in which the

second marker cannot see the comments or mark given by the first marker. The two markers

are responsible for marking the work of all candidates in the group independently and the

AU-RSC-20-3965-A 2021/22

6

final mark is normally obtained from an average of the two marks. If there is a large

discrepancy between the two marks (e.g. 6% or more) either the two markers will discuss

and agree the final mark or a relevant member of staff (e.g. Programme Director or module

leader) will mediate a discussion between the two markers to agree a final mark. If no

agreement can be reached a third internal marker will normally mark the work. The third

marker should then act as mediator between the two first markers to agree a mark. If the third

marker is unable to facilitate such an agreement, the third marker will be responsible for

awarding a mark anywhere within the range of marks of the first and second marker.

This type of marking should be used for dissertations and major undergraduate projects.

Informed double-marking Two markers are responsible for marking the work of all candidates consecutively. The

second marker sees the marks and comments of the first marker. If there is a large

discrepancy between the two marks (e.g. 6% or more) either the two markers will discuss

and agree the final mark or a relevant member of staff (e.g. Programme Director or module

leader) will mediate a discussion between the two markers to agree a final mark. If no

agreement can be reached a third internal marker will normally mark the work. The third

marker should then act as mediator between the two first markers to agree a mark. If the third

marker is unable to facilitate such an agreement, the third marker will be responsible for

awarding a mark anywhere within the range of marks of the first and second marker. This

type of marking may result in adjustments to individual marks.

Informed double-marking may be used where the specialist nature of the subject matter

means that blind double-marking would not be practicable or where there are teams of

markers marking a batch of assessments.

Moderation Moderation is the name given to procedures for checking the accuracy and appropriateness

of academic assessment. It usually involves a person ‘new’ to the item being moderated.

That may be a person external to the University or a member of staff who has not been

directly involved in the process to be moderated. Assessment processes which are

moderated at Aston include the preparation of examination papers and marking (e.g. of

examination scripts, essays, oral presentations etc).

Moderation: Arithmetic checking A check is made that all questions have been marked and that the recording and addition of

marks adding to the total mark is correct. This is separate from marking of the work and may

be done by someone who is not a member of academic staff.

This type of check is particularly important when examination papers include multiple choice

questions, short answer questions or require more than three longer questions to be

answered.

AU-RSC-20-3965-A 2021/22

7

Moderation: Sampling A defined sample of work is marked by a member of staff who is not the main marker (or

member of a team of markers) in order to confirm (or not) the mark awarded by the first

marker.

The aim of this type of moderation is to highlight any potential marking problems. For

example marks may be found to be too low or too high either overall or in a particular band

(e.g. 30 – 40%). In the case of a marking team the marking of one member may be found to

be out of line with the others.

If the moderation process identifies concerns about the marking standards of the sample or

has identified a systematic error in marking or marks processing, this should be

communicated to the module leader and/or Programme Director, as appropriate. The module

leader or the Programme Director, as appropriate, will then review the work, consider the

concerns raised, discuss the issue with the marker(s), if appropriate, and respond to the

moderator(s) to indicate what action they intend to take, if any.

Issues identified by this type of moderation may be dealt with by adjusting the marks for the

cohort or a relevant section of the cohort or (in extreme cases) by the moderator taking on

the role of informed second marker and marking all work in the cohort. Agreed marks are

then reached for all candidates following the process outlined under ‘Informed double-

marking’.

Where the proposed action may adjust marks without full second marking of the whole

cohort, this must occur in a systematic and considered way so that all affected work is treated

equally and not just the moderated sample.

A report of moderation should be made available to the Module Board. Evidence of

moderation e.g. checks on sections of the question paper made by the moderator should be

provided. Adjustments resulting from moderation must be considered and confirmed at a

Module Board.

Choosing a sample Samples of work for moderation are selected so as to test the security of standards across

the full marking range and where the candidate has failed. At least 20% of final Stage

examinations and coursework should be sampled. All first markers (e.g. if a team of markers

is involved) should be sampled. A sample should contain at least one exemplar of work

closest to each side of the boundaries of mark bands (e.g. pass/fail, 2.1/2.2,

Merit/Distinction), and all fail marks. A sample may additionally include a few scripts from the

centre of mark bands for benchmarking purposes. This pattern is normally repeated for each

marker sampled. All fail marks should be included in the sample, with particular attention to

marks which may be condoned.

Rounding of marks Rounding of marks means changing fractional marks to the nearest whole number. The

University’s student records system displays marks to two decimal places, and stores marks

to a greater number of decimal places. A module mark of 39.50 for 40% pass Level 4, 5 and

AU-RSC-20-3965-A 2021/22

8

6 mark schemes, and 49.50 for a 50% pass Level 7 mark scheme, will be counted in the

student records system as a pass and the credits will be awarded on the examination board

report. Rounding of marks is therefore unnecessary and should not normally occur, except

where averaging of marks for examination scripts with more than one answer results in a

fractional mark for the assessment as a whole. On such occasions the actual number should

be recorded to two decimal places. Undergraduate Module Boards may choose to moderate

module marks of 39 to either 38 or 40, based on academic judgement of whether all learning

outcomes have been achieved. Postgraduate Module Boards may choose to moderate

module marks of 49 to either 48 or 50, based on academic judgement of whether all learning

outcomes have been achieved.

AU-RSC-20-3965-A 2021/22

9

4 GENERIC MARKING SCALES Suitable for Undergraduate programmes (and qualifying Stages of an integrated Master’s

programmes) - Levels 4, 5 and 6.

DESCRIPTION: Work within the mark ranges below demonstrates the

following characteristics:

MARK RANGE

Unique or insightful work, which is either of publishable quality in a

reputable journal or attains the professional standards expected for the

discipline, or; work which displays a critical awareness of the principles

and practices of the discipline. Thorough comprehension of the

assessment’s requirements, fully realises learning outcomes for the

assessment and develops them far beyond normal expectations.

100 – 80%

Displays an individual perspective which is supported by reasoning or

evidence. Insightful, logical and articulate, demonstrates a

comprehensive coverage of subject matter, engagement with

scholarship and research, very good analytical/creative ability.

Surpasses the intended learning outcomes.

70 – 79%

Sound and well thought out, organised, secure knowledge of subject,

appropriate use of critical references, realises the intended learning

outcomes broadly, well expressed, good analytical/creative skills.

60 – 69%

Displays adequate use of critical method but may be poorly argued.

Adequate or routine knowledge of subject. Evidence is referred to but

there may be inconsistencies in the way it is used. Clear evidence that

learning outcomes are being achieved.

50 – 59%

Competent but largely descriptive in approach. Displays understanding

of subject with some limitations e.g. an element may be missed.

Evidence that learning outcomes are being achieved.

40 – 49%

Evidence that some learning outcomes have been achieved or most

learning outcomes achieved partially. Although work may include brief

signs of comprehension, it contains basic misunderstandings or

misinterpretations, demonstrates limited ability to meet the requirements

of the assessment.

30 – 39%

Brief, irrelevant, confused, incomplete. Does not come close to meeting

the required learning outcomes.

29% and below

The objective of the marking scales is to establish a University-wide reference point which does not

conflict with any guidance currently used in Colleges. There is no intention to replace existing College

marking scales, or to require that each College develops additional College-specific scales in addition

to these. Each College should determine how best to use these marking scales in the context of its

own programmes.

AU-RSC-20-3965-A 2021/22

10

5 LATE SUBMISSION OF WORK AND PENALTIES • This policy applies to all assessments detailed in Module Specifications for which

there is a formal deadline. In the case of assignments that are submitted in more than one format (e.g. identical submissions in physical and electronic formats, or the submission of multiple submission components for the same assessment), the date of submission recorded for lateness penalty purposes will be that of the last piece of work submitted.

• The late submission of group work will be subject to the same lateness penalties as those for an individual piece of work.

• Students may submit work up to five working days after the formal assessment

deadline, but the work will be subject to a penalty. (NB: ‘working days’ excludes submission at weekends and on Bank Holidays or University closed days, when College offices are not open to receive submissions, but it does not preclude submission during vacations. The five-working day submission window will therefore continue to run after the last day of a term.)

• For the purposes of lateness penalties, a working day refers to an exact, twenty-four hour period e.g. an assessment due at 15:00 that is submitted at 18:00 the following working day will be treated as two days late.

• The penalty will be a based on a proportion of the awarded mark.

• The late submission penalty will be 10% of the awarded mark for each working day that the piece of work was submitted after the formal deadline (see Example 1 below).

• There will be a penalty collar at first attempt (or any uncapped attempt), in that the

final mark for the component of assessment after penalty has been imposed will not fall below the pass mark for that component. If the awarded mark for the component was already below the pass mark, prior to late submission penalty, the awarded mark will stand as the final mark. (See Example 2 below).

• Work submitted after the five working day deadline will be accepted by the College office but will be recorded with a mark of zero (0-PN). Such work will require evidence of exceptional circumstances which meet the regulations before an academic mark for it will be considered at the module board.

• Work submitted more than five days late may be passed to markers in order for

students to receive feedback on it, at the discretion of the College. Where this practice

is adopted the mark formally recorded will be zero (0-PN).

• Any exceptions to the late submission policy at module level will be subject to

approval by the relevant College’s Learning & Teaching Committee and will be clearly stated in the appropriate Student Handbook. Exemptions to cover a whole programme should be approved by Regulation Sub-Committee.

• The penalty collar will not apply to second or third attempts, for which a mark capped at the pass mark is formally recorded. Any lateness penalties due will be applied to the mark awarded and may result in a mark that falls below pass level (see Example 3 below).

• Late penalties are considered to be discrete from other penalties, and the late

penalty collar does not prevent a fail mark resulting in cases where another penalty is also applicable. Late penalties should therefore be levied before any other penalty.

AU-RSC-20-3965-A 2021/22

11

Worked examples: Example 1 – Work submitted 3 days late

A piece submitted 3 days late and awarded a mark of 60 incurs a penalty of (10% x 60 =) 6 marks per day of lateness. If the pass mark for this component of assessment is 40, the mark recorded after penalty will therefore be (60 – 18 =) 42.

Example 2 – The penalty collar

a. A piece submitted 3 days late and awarded a mark of 60 incurs a penalty of (10% x 60 =) 6 marks per day of lateness:

▪ where the pass mark for the component is 40, the mark recorded after penalty will be 42. ▪ where the pass mark for the component is 50, the full deduction of 18 marks would result in a fail mark, so the mark actually recorded after penalty will be 50.

b. For a piece submitted 4 days late and awarded a mark of 60, a full deduction would cause a fail whether the pass mark is 40 or 50, so:

▪ where the pass mark for the component is 40, a final mark of 40 will be recorded. ▪ where the pass mark for the component is 50, a final mark of 50 will be recorded.

c. A piece accepted within the 5 working day submission window and awarded a mark of 35 is a fail whether the pass mark is 40 or 50. A final mark of 35 will be recorded.

Example 3 – Work submitted as a second or third attempt

A referred or repeat assessment is submitted one day late and given an academic mark of 50:

▪ where the pass mark is 40, the lateness results in a penalised mark of 45,

but a final (passing) mark of 40 will be recorded.

▪ where the pass mark for the component is 50, a final (failing) mark of 45 will

be recorded.

Example 4 – Late penalties and other penalties in combination

For example:

▪ a group of 3 students submits a group assignment 5 days late (where the

pass mark is 40);

▪ the mark for the group assignment is 80 so the mark is reduced to 40 due to

late submission;

▪ one member of the group should have their mark reduced by half because

of a formal penalty from an Academic Offences Officer so their mark would

be further reduced to 20.

AU-RSC-20-3965-A 2021/22

12

Similarly: ▪ a group of 3 students submits a group assignment 5 days late (where the

pass mark is 40);

▪ the mark for the group assignment is 70, but levying the full penalty would

result in a fail mark of 35, so the mark after late submission penalty is

recorded at 40;

▪ one member of the group should have their mark reduced by half because

of a formal penalty from an Academic Offences Officer so their mark would

be further reduced to 20.

Example 5 – Work submitted late following a University closure day

For example:

▪ an assessment deadline is set for Thursday at noon. A student submits the

assessment on the following Wednesday at 11:00, and the University was

closed on the Monday.

▪ For the purposes of determining lateness penalties, the Friday, Tuesday,

and Wednesday apply. Therefore, the submitted assessment would be

treated as three days late.

AU-RSC-20-3965-A 2021/22

13

6 RECYCLING OF ASSESSMENT CONTENT

1. Previously submitted and marked assessments will form part of credit already

awarded. Students are not allowed to achieve further credit for a piece of work that

has already had credit awarded for it. Where a student uses a piece of work that has

already been assessed, either in whole or in part, for another separate assessment,

this is deemed poor practice and is referred to at Aston University as ‘recycling

material’.

2. Where a student submits a piece of work that contains suspected recycled material

the piece of work will initially be marked without penalty. When it is demonstrated that

there has been recycling of material, marks will then be deducted from the initial

marking in relation to the amount of recycled material the piece of work contains. For

example: if the piece of work contains 10% recycled material, the assessment mark

will be reduced by 10% of the initial mark, if the piece of work contains 25% recycled

material, the assessment will be reduced by 25% and so on. Any proposed deduction

will only be made subject to the approval of the appropriate Academic Offences

Officer.

3. The only allowed exceptions in relation to recycling are:

a. When students have been requested to submit a partial draft or research

proposal/plan at an earlier point in a module which will then be used to directly

inform the final piece of work for that same module or a subsequent related

module.

b. When students are resubmitting a previously failed piece of work for

reassessment.

4. Deductions imposed for the recycling of assessment material are not subject to a

penalty collar, and therefore may lead to an initial pass mark becoming a fail mark following the appropriate deductions being made. In cases where a deduction results in a condonable fail mark, the Board of Examiners will subsequently exercise its discretion as to whether the mark should be condoned (assuming it is the only assessment for the module). In the event of an assessment being submitted late and containing recycled material, the late penalty will be applied first, followed by the deduction for recycled content, calculated on the basis of the initial mark.

Worked examples: Example 1 – Work submitted with 25% recycled content

A piece of work is submitted and given an initial mark of 60, but 25% of the material is then identified as recycled assessment content from previous work. The mark of 60 is therefore reduced by 25% of 60, and a final mark of 45 is recorded.

AU-RSC-20-3965-A 2021/22

14

Example 2 – Work submitted 2 days late with 25% recycled content

A piece of work is submitted and given an initial mark of 60, but the work was

submitted 2 days late and 25% of the material is identified as recycled assessment

content from previous work:

▪ The initial mark of 60 is reduced by 10% of the mark for each day the

assessment is late: (10% x 60 x 2) = 12, giving an interim mark of (60 – 12)

= 48.

▪ The recycling deduction is then applied, based on the original mark, in this

case (25% x 60) = 15, giving a final recorded mark of 48 - 15 = 33

AU-RSC-20-3965-A 2021/22

15

7 USE OF TEXT-MATCHING SOFTWARE

Guiding principles

1.1 Aston University recognises that confidence in the academic output of their students is vital to assure the integrity of the students’ academic achievements. To this end Aston University will use text-matching services such as Turnitin™. It is intended that this service will mainly be used as a supportive and constructive tool.

1.2 Text-matching services will be used by staff at Aston University primarily as a

supportive teaching tool to enable students to understand the principles of good academic practice in relation to referencing and the use of academic texts in their own original work.

1.3 'Students will be provided with formative opportunities to use originality checking

functionality during the early stages of their study.'

1.4 Text-matching services will be used in the detection of poor academic practice and plagiarism.

1.5 The use of text-matching services does not imply that plagiarism is suspected in

every assignment that is electronically submitted through this process.

1.6 The use of text-matching services will not be restricted solely for the purpose of originality checking.

Policy

1.7 Specific consent is given for the University to submit student work to text-matching services through the enrolment process. This consent is listed in the Terms and Conditions of enrolment.

1.8 In giving this consent the student is not waiving their right to ownership of

their original academic work.

1.9 Programme Handbooks will state the use of text-matching services for each specific programme.

1.10 Should the University suspect that plagiarism has taken place they reserve

the right to submit any assignment into the text-matching services process.

1.11 The University may receive requests via text-matching services from staff at another Higher Education Institution to allow them to access a piece of work submitted by an Aston University student in relation to a suspected case of plagiarism. This request must be received in writing (letter or email) from a named individual at the requesting institution. Such requests will normally be agreed to but will be decided on a case by case basis. It is the Executive Dean (or their nominee’s) decision as to whether the assignment will be provided for this

AU-RSC-20-3965-A 2021/22

16

purpose. In replying to such a request the Aston member of staff will ensure that the assignment when sent to the other institution does not breach any data protection guidelines or contain sensitive information. Guidance where needed should be requested from the Director of Student and Academic Services.

AU-RSC-20-3965-A 2021/22

17

APPENDIX 1: GUIDANCE FOR MODERATION OF MARKS BY MODULE BOARDS Context Moderation is a key process in the maintenance of academic standards by checking the accuracy and appropriateness of academic assessment(s). At Aston University, moderation can take place at two points:

1) During the marking process prior to release of provisional marks to students, as detailed in the Assessment Regulations. This should be recorded on the appropriate pro forma and be available for discussion at Module Boards if required.

2) At Module Board level in line with the General Regulations for the Conduct of Boards of Examiners.

This draft guidance is designed to help Module Boards reach decisions to maintain academic quality, with particular relevance where assessments have been affected by the ongoing pandemic. It is not designed to limit or expand the powers of the boards as detailed in the regulations referenced above, nor should moderation at Module Boards be considered as a replacement for robust moderation processes as outlined in point 1 above. Module Boards need to undertake two actions; review of module assessments and review of candidate marks. These two actions need to be taken for all modules, including those which are taken by candidates on programmes accredited or recognised by PSRBs. When implementing this guidance, please pay special attention to section 3.2 in the above, especially: “a to determine and report to Programme Boards the award of marks for all candidates in

the context of overall averages, student performance and particular circumstances affecting the assessment and conduct of the module, and, where appointed, the comments of External Examiners.”

In the current context, particular attention should be given to those assessments where the format was either altered from traditional delivery e.g. from on-campus to remote examination, or altered during the academic year. Module Board Decisions Likely triggers for moderation are:

• An anomalous mark distribution e.g. unusual patterns or skewed distributions.

• A distribution of marks that is inconsistent with previous years (normally against data from the previous three years) with a comparison of means and standard deviations between years.

• A distribution of marks that is inexplicably inconsistent with other modules completed by the same candidates.

• A known issue with an assessment (either identified during the assessment period or when completing the marking). This may include technical issues with delivery of the assessment.

It is difficult to provide guidance for every potential situation and Module Boards should use their collective academic judgement and experience when making decisions. In most cases, comparison of the means, standard deviations and mark distributions should be sufficient to reveal significant differences where moderation would be recommended. There may be

AU-RSC-20-3965-A 2021/22

18

occasions where Module Boards wish to undertake further statistical analysis of the data to inform this decision to determine if moderation is appropriate. Consistent benchmarks should be applied across all modules contributing to a stage.

• Minor differences e.g. fluctuations of 1-2% outside of the “norm” are likely not to require moderation. Above this, Module Boards should apply academic judgement about the need for moderation, taking into consideration changes to the format of assessment etc.

• Larger changes should be considered for moderation unless there is good reason not to e.g. a change in assessment format or deliberate rebalancing of assessments from previous years.

• If assessment means lie outside of the “typical” range for that assessment, then moderation may be appropriate. This is particularly relevant for new assessments / modules.

• If mark distributions are not roughly normally distributed (or have an atypical distribution for the assessment / module / programme) then moderation should be considered providing the dataset is large enough.

Guiding principles for moderation:

• Moderation should be undertaken based on evidence as detailed above.

• Where cohort sizes are too small to use reliable comparison, moderation should only take place based on robust identification of errors and, otherwise, assessments should be remarked.

• Moderation should be done at the level of individual assessments for which potential issues have been identified.

• Moderation should be applied to all candidates e.g. not just those around a boundary; this moderation may not necessarily be uniform across the cohort.

• Moderation should be transparent, and the approach and rationale justified and recorded in the minutes of the Module Board meeting.

Any moderation should always be based on the principle that candidates should only pass a module if they have demonstrated that they have achieved the relevant learning outcomes. Therefore, specific attention to this needs to be paid with respect to candidates on the pass/fail boundary. This should ensure that all candidates awarded a pass have demonstrated that they have met the relevant learning outcomes, and at the same time, those who have failed have clearly demonstrated that they have not met the learning outcomes. Details of the review and any subsequent action should be recorded in the minutes of the Module Board meeting, which should be made available to the relevant Programme Board(s) and for Stages of a programme which contribute to the final award, discussed with the relevant External Examiner(s) and be in accord with any specific Programme Regulations. Guidance Detailed below are some acceptable methods for moderation of marks. This is not an exhaustive list, but non-standard methods should be discussed with an Associate Dean Education or equivalent. A brief rationale for the choice of method should be provided in the minutes of the Module Board. It is likely prudent to model these ahead of the Module Board such that the outcomes can be presented and discussed. In general, consistent approaches should be taken for all modules that affect a particular cohort of students unless there is clear reasoning to do otherwise.

AU-RSC-20-3965-A 2021/22

19

Moderation at Module Board level will normally take the form of scaling. Please note that this scaling need not be linear throughout the module. Principles of scaling:

• The rank order of candidates for the assessment must remain unchanged by scaling.

• Scaling must encompass the full range of marks within the assessment (normally 0-100).

• Scaling should be done with reference to the assessment learning outcomes i.e. a pass mark should not be scaled to a fail without clear evidence of failure (not meeting the relevant learning outcomes).

Suggested methods of scaling:

• Methods routinely used within the discipline e.g. MBChB.

• Simple addition or subtraction of a fixed number of marks to all candidates; no scaled marks should exceed 100 or drop below 0.

• Multiplication of all marks by a particular factor e.g. 0.92; pay close attention to classification borderlines.

• Mapped scaling (see below); this should be linked to the level at which the learning outcomes are achieved.

Determination of scaling factors: Boards should ideally model a number of scenarios and use academic judgement to determine either the fixed number of marks to be added / subtracted or the scaling factor. Ideally, this should be data driven and compared to previous cohort performance in the assessment. The method chosen will be based on the desired outcome e.g. realignment of the mean or altering the shape of the mark distribution. In all cases, the scaled marks should be checked for mean, standard deviation and distribution. Example 1: The subtraction of a number of percentage points equivalent to half the difference in the mean mark for the assessment in question and the highest mean mark in the equivalent assessment in the two preceding academic years. As an example, if the mean of the assessment was 80 and the mean marks of the equivalent assessment in the previous two academic years was 60 and 55, and the raw mark awarded was 90, the adjusted mark would be 80 (90-((80-60)/2)). In this case a collar applied at the pass mark. Example 2: The aggregate mean of the assessment from the previous two years is calculated. The difference between this and the current assessment mean is then calculated and a percentage of this e.g. 40% is used to correct the marks uniformly; if the aggregated difference is -12% (current assessment is lower) then each mark in the current assessment is increased by 4.8% (0.4*12). Mapped scaling: This is a process to non-uniformly scale marks within a cohort. Briefly, where possible, submissions that map to 10, 20, 30, 40, 50, 60, 70, 80 and 90% marks are identified based on the academic quality of the submissions with reference to the marking scheme and Learning Outcomes for the piece of assessment. Note that not all of these marks may be present, but attempts should be made to identify submissions corresponding to as many as possible. The raw mark for the assessment that maps to 40% should be scaled to 40% and the same process

AU-RSC-20-3965-A 2021/22

20

applied to all of the other “benchmarks”. An example below refers to a set of marks which are “bunched” in the 2:1 region:

Original Mark

Scaled Mark

22 20

35 30

44 40

58 50

62 60

66 70

70 80

80 90

Once the benchmarks have been identified, the remaining marks should be scaled linearly between them. Unsuitable assessments In rare cases, if a piece of assessment is deemed to be unsuitable and contains no reliable information, then this assessment, or part of it, may be excluded from, or rescaled within, module calculations provided that Learning Outcomes are still met by candidates who pass the remaining assessments.