Editing Eleven-plus (section)

=== Scoring ===
{{primary sources section|date=April 2023}}
{{Original research section|date=April 2023}}
[[File:Wechsler.svg|thumb|left|[[Normal distribution]] curve that illustrates [[standard deviations]]. This illustration has the [[Wechsler Intelligence Scale for Children|Wechsler scale]] added, whereby the scores are scaled with a mean of 100 and standard deviation of 15.]]
{| class="wikitable floatright"
|-
! Authority/consortium !! Mean !! Standard Deviation 
|-
| Bishop Wordsworth || 100 || 15
|-
| Chelmsford || 100 || 15
|-
| Dover Grammar School for Boys || 100 || 15
|-
| Folkestone || 100 || 15
|-
| Gloucester || 100 || 15
|-
| Harvey || 100 || 15
|-
| Heckmondwike || 100 || 15
|-
| Henrietta Barnett School || 100 || 15
|-
| Kendrick School || 100 || 15
|-
| Mayfield || 100 || 15
|-
| Reading School || 100 || 15
|-
| Redbridge  || 100 || 15
|-
| The Latymer School || 100 || 15
|-
| Torbay and Devon Consortium || 100 || 15
|-
| West Midland Boys || 100 || 15
|-
| West Midland Girls || 100 || 15
|-
| Buckinghamshire || 100 || 43
|-
| Dame Alice Owen || 106 || 15
|-
| Slough Consortium || 106 || 15
|-
| South West Herts || 106 || 15
|-
| Bexley || 200 || 30
|-
| King Edwards Consortium || 200 || 30
|-
| Warwickshire || 200 || 30
|-
| Wirral Borough Council || 234 || 15
|-
| Altrincham || 315 || 30
|-
| Sale || 318 || 30
|-
| Stretford || 327 || 30
|-
| Urmston || 328 || 30
|}

England has 163 grammar schools 155 of which control their own admissions including the choice of test. (143 Academy Converters, six Foundation schools and six Voluntary aided schools control their own admissions. Admissions for the remaining seven Community Schools and one Voluntary Controlled school are determined by the local authority.;
<ref>DfE Schools Pupils and Characteristics dataset''[https://explore-education-statistics.service.gov.uk/find-statistics/school-pupils-and-their-characteristics School level data]''.  Department for Education.</ref>
<ref>West, Anne and David Wolfe. ''[https://www.semanticscholar.org/paper/Academies%2C-autonomy%2C-equality-and-democratic-the-in-West-Wolfe/6dc5141b4b234ee77dddd4b1dc157515b4b9dc63 Academies, autonomy, equality and democratic accountability: Reforming the fragmented publicly funded school system in England. London Review of Education (2019)]''</ref>

Over 95% of grammar schools now determine their own admissions policies, choosing what tests to set and how to weight each component. Although some form consortia with nearby schools to agree on a common test, there may be as many as 70 different 11+ tests set across the country <ref>CEM set tests for 28 different LAs, consortia and individual schools and claim to have 40% of the market. Extrapolating would suggest there are approx. 70 different sets of tests in total.</ref> meaning it is not possible to refer to ''the eleven plus test'' as a single entity.

Tests are multiple choice. The number of questions varies but the guidance provided by GLA 
<ref>Free familiarisation materials''[https://web.archive.org/web/20230721083539/https://11plus.gl-assessment.co.uk/free-materials/]''.  Granada Learning Assessment (GLA).</ref> shows that full length Maths and English Comprehension tests are both 50 minutes duration and consist of about 50 questions. Verbal Reasoning is 60 minutes containing 80 questions. Non-Verbal Reasoning is 40 minutes broken into four 10-minute separately-timed sections each containing 20 questions. At a rate of one question every 30 seconds, it could be argued that the test is one of speed rather than intelligence.

One mark is awarded for each correct answer. No marks are deducted for incorrect or un-attempted responses.
<ref>Freedom of Information''[https://www.whatdotheyknow.com/request/familiarisation_booklet_accuracy#incoming-1060294]''. Buckinghamshire Grammar Schools.</ref> There are usually five possible answers, one of which is always correct meaning a random guess has a 20% chance of being correct and a strategy of guessing all un-attempted questions in the last few seconds of the exam will, if anything, gain the candidate a few additional marks which may make the difference needed to gain a place.

The actual marks from these tests, referred to as ''raw'' marks, are not disclosed by all schools, and instead parents are given ''Standard Age Scores'' (SAS).  A [[standard score]] shows how well the individual has performed relative to the mean (average) score for the ''population'' although the term ''population'' is open to interpretation. GL Assessment, who set the majority of 11+ tests, say it should be, "a very large, representative sample of students usually across the UK";<ref>GL Assessment ''[https://www.gl-education.com/media/1757/guide-to-standardised-tests.pdf A short guide to standardised tests]''.  013 GL Assessment Limited.</ref> however, grammar schools may ''standardise'' their tests against just those children who apply to them in a given year, as this enables them to match supply to demand.

Test results follow a [[normal distribution]] resulting in the familiar bell curve which reliably predicts how many test takers gain each different score. For example, only 15.866% score more than one [[standard deviation]] above the mean (+1σ generally represented as 115 SAS) as can be seen by adding up the proportions in this graph based on the original provided by M. W. Toews).

By ''standardising'' on just the cohort of applicants, a school with for example, 100 places which regularly gets 800 applications can set a minimum pass mark of 115 which selects approximately 127 applicants filling all of the places and leaving about 27 on the waiting list. The downside of this ''local standardisation'', as it has been called, is parents are frequently unaware that their children are being judged as much by the standard of other applicants as their own abilities.
 
Another issue with the lack of national standards in testing is it prevents any comparison ''between'' schools. Public perception may be that only pupils who are of ''grammar school standard'' are admitted to grammar schools; however, other information such as the DfE league tables<ref>{{cite web|url=https://www.compare-school-performance.service.gov.uk/download-data|title=Download data – GOV.UK – Find and compare schools in England|website=Find and compare schools in England}}</ref><ref>Data source: England_ks4final.csv. Field 39, PTPRIORHI shows the Percentage of pupils at the end of key stage 4 with high prior attainment at the end of key stage 2.</ref> calls into question the existence of any such ''standard''. Competition for places at Sutton Grammar School is extremely fierce with, according to an online forum<ref name = elevenplusexams>{{cite web|url=https://www.elevenplusexams.co.uk/schools/regions/surrey-11-plus|title=11 plus Surrey, Grammar School Admissions|website=www.elevenplusexams.co.uk}}</ref> over 2,500 applicants in 2016. At the other end of the scale, Buckinghamshire council website says, "If your child's STTS is 121 or above, they qualify for grammar school. We expect that about 37% of children will get an STTS of 121 or more."<ref>Marking the Secondary Transfer Test''[https://web.archive.org/web/20230721112048/https://www.buckinghamshire.gov.uk/schools-and-learning/schools-index/school-admissions/school-admissions-guides-policies-and-statistics/guide-to-grammar-schools-and-the-secondary-school-transfer-test-11-plus/marking-the-secondary-transfer-test/]''. Buckinghamshire Council.</ref> Official statistics show 100% of those admitted to Sutton Grammar School have, "high prior attainment at the end of key stage 2", compared to only 44% of those who attend Skegness Grammar School. The Grammar Schools Heads Association's Spring 2017 newsletter<ref>{{Cite web|title=GSHA Newsletter|url=https://schoolsweek.co.uk/wp-content/uploads/2017/02/GSHA_Newsletter_SPRING_2017.pdf|url-status=live|archive-url=https://web.archive.org/web/20181016203312/https://www.schoolsweek.co.uk/wp-content/uploads/2017/02/GSHA_Newsletter_SPRING_2017.pdf |archive-date=16 October 2018 }}</ref><ref>{{Cite web|url=http://schoolsweek.co.uk/wp-content/uploads/2017/02/GSHA_Newsletter_SPRING_2017.pdf|title=The Grammar Schools Heads Association's website went offline in early July 2017 but a copy of their Spring newsletter is still available from Schools Week's website}}</ref> says the government are considering a national selection test which would remove the lack of consistency between different 11+ tests. 
 
Between them, GL (Granada Learning) and CEM (Centre for Evaluation and Monitoring) earn an estimated £2.5m annually<ref>''Coombs v Information Commissioner'', EA/2015/0226, 22 April 2016, p. 11.  CEM told the Information Commissioner they earn £1m from 40% of the market making £2.5m a reasonable estimate for 100% of the market.</ref> from setting and marking the 11+ tests.  Releasing the raw marks would bring some clarity to the admissions process but attempts to do so have generally been unsuccessful. GL have used the fact that they are not covered by Freedom of Information legislation to withhold information<ref name="whatdotheyknow.com">{{cite web|url=https://www.whatdotheyknow.com/request/request_for_report_on_11_exam|title=Request for report on 11+ exam – a Freedom of Information request to Altrincham Grammar School For Boys|date=23 May 2016}}</ref> made for information relating to the 11+ exams used by Altrincham Grammar School for Boys, who stated, "Our examination provider, GL Assessment Limited (GL) is not subject to the Freedom of Information Act 2000 (FOI) as it is not a public body.“, whilst their main rival CEM successfully argued in court<ref>{{Cite web|title=Case no. EA/2015/0226|url=https://informationrights.decisions.tribunals.gov.uk/DBFiles/Decision/i1785/018%20250416%20Coombs%20judgement%20final.pdf|url-status=live|archive-url=https://web.archive.org/web/20200518170621/http://informationrights.decisions.tribunals.gov.uk/DBFiles/Decision/i1785/018%20250416%20Coombs%20judgement%20final.pdf |archive-date=18 May 2020 }}</ref><ref>Coombs v Information Commissioner (ibid), p. 5.</ref> that, "one of the benefits of its 11+ testing is that it is 'tutor proof'” and releasing the raw marks would undermine this unique selling point.

When a [[Standard score#Calculation from raw score|standard score]] is calculated the results is a negative value for any values below the mean. As it would seem very strange to be given a negative score Goldstein and Fogelman (1974)<ref>{{Cite web|title=Age standardisation and seasonal effects in mental testing|url=https://www.bristol.ac.uk/media-library/sites/cmm/migrated/documents/age-standardisation-and-seasonal-effects.pdf|url-status=live|archive-url=https://web.archive.org/web/20150528111147/http://www.bristol.ac.uk/media-library/sites/cmm/migrated/documents/age-standardisation-and-seasonal-effects.pdf |archive-date=28 May 2015 }}</ref> explain, "It is common to 'normalise' the scores by transforming them to give a distribution with a mean of 100 and a standard deviation of 15."<ref>{{Cite web|url=http://www.bristol.ac.uk/media-library/sites/cmm/migrated/documents/age-standardisation-and-seasonal-effects.pdf|title=Goldstein H and Fogelman K, 1974, ''Age standardisation and seasonal effects in mental testing''. Downloaded from /www.bristol.ac.uk}}</ref> Thus a normalised SAS of 100 indicates the mean (average) achievement whilst a score of 130 would be two standard deviations above the mean. A score achieved by only 2.2% of the ''population''. Most, but not all, authorities normalise follow this convention. The following table<ref>{{cite web|url=https://www.whatdotheyknow.com/request/cem_11_test_results_2014_2016|title=CEM 11+ test results 2014 – 2016. – a Freedom of Information request to Durham University|date=6 October 2016}}</ref> showing the normalisation values used by some for 2017 entry (tests taken in 2016).