Selection for medical schools Chris McManus University College London ‘Selection in Medicine’, ASME, London, 28th April 2010 Selection for medical schools
Selection for medical schools
Chris McManusUniversity College London
‘Selection in Medicine’, ASME, London, 28th April 2010
Selection for medical schools
Selection for medical schools
• Academic
– A-levels
– Aptitude tests (BMAT, UKCAT, GAMSAT, etc)
• Non-academic
– Personality
– Motivation
• Widening access
– Complex issue
Selection for medical schools
• A-levels:– Valid predictor of outcome in medical school and
postgraduate assessments
• Westminster clinical entry cohort (1975-82)
• 1991 cohort study
• UCL entrants 2004
– Valid across all universities for all subjects(HESA)
The 1991 Cohort Study
EntrantsOct 1991
EntrantsOct 1992
N ~ 3300N ~ 590092%
ApplicantsOct-Dec 1990
Questionnaire
Leverhulme/DoH
N ~ 270060%
Final Year1996
Questionnaire
Final Year1997
Questionnaire
DoH
N ~ 270055%
PRHOsAug 1996-Jul 1997
Questionnaire
PRHOsAug 1997-Jul 1998
Questionnaire
NTPMDE
N ~ 240060%
SHO/SpRGPs
Dec 2002
Questionnaire
LondonDeanery
MRCP~1997-99
N = 92531%
March2009
N ~ 270061%
LondonDeanery
Questionnaire
Part 1mark at first attempt
80.075.0
70.065.0
60.055.0
50.045.0
40.035.0
30.025.0
20.015.0
10.0
100
80
60
40
20
0
Std. Dev = 11.34
Mean = 47.7
N = 925.00
MRCPmark
MRCP(UK) markat first attempt
5.55.04.54.03.53.02.52.01.51.0
Pa
rt1
ma
rka
tfirs
ta
tte
mp
t
75
70
65
60
55
50
45
40
35
30
25
20
MRCP(UK) markat first attempt
EEE DDD CCC BBB AAA
A-level grade
A-levels
BasicMed Sci
BSc Class
MRCPmark
BSc Taken
GCSE
AH5
MRCP(UK) markat first attempt
A-levels
BasicMed Sci
BSc Class
MRCPmark
BSc Taken
GCSE
AH5
MRCP(UK) markat first attempt
AH5 =High-level intelligencetest for university levelstudents
Selection for medical schools
• A-levels:– Of course no longer usable in medicine …
UCAS 2003-5
All UCASapplicants
MedicalSchoolapplicants
MedicalSchoolentrants
• High A-level grades require:
• Intellectual ability
• Motivation / drive / study habits / learningstyles / personality (conscientiousness)
• Content knowledge (e.g. biology /chemistry)
• Which are the key components for predictingmedical school outcome?
• Probably not intellectual ability
– Necessary but not sufficient
A-levels as predictors
Selection for medical schools
• “Aptitude” tests– BMAT
– UKCAT
– GAMSAT
– MSAT
– MCAT
Selection for medical schools
• What is meant by ‘aptitude’?
• Tests are of two fundamental types:
– Achievement: Knowledge/content– A-levels
– BMAT: Section 2: Scientific knowledge and applications;
– MCAT biological sciences / physical sciences subtests
– GAMSAT: III: Reasoning in Biological and Physical Science
– Ability/Aptitude: Essentially IQ tests, although that term islittle used:
– UKCAT: Verbal reasoning/ Quantitative reasoning / Abstract reasoning/ Decision analysis
– BMAT: Section 1: Aptitudes and skills
– MCAT: Verbal reasoning subtest
– GAMSAT: I - Reasoning in the Humanities and Social Sciences
Selection for medical schools
• Do ability/aptitude tests predict?
(1973)
Commissioned byCommittee of Vice-Chancellorsand Principals
The ISPIUA project: Investigationinto Supplementary PredictiveInformation for University Admission
Test of Academic Aptitude (TAA):Mathematics (M) and Verbal (V)
October 1967: 27,315 students inthe 4th term of the Sixth Form
If ISPIUA’s Test ofAcademic Aptitudedidn’t predict, doUKCAT and BMAT?
• BMAT– 2003 onwards, previously MVAT
– ~ 12,000 entrants to medical school
• UKCAT– 2006 onwards
– ~ 24,000 entrants to medical school
• BMAT – Medical Education 2009, n=1002
• UKCAT – Lynch et al Med Ed 2010 n=341
• Why are there no proper validity/outcomestudies?
Do UKCAT and BMAT predict?
Achievement tests
• MCAT
• BMAT
Academic Medicine, 2007
• High A-level grades require:• Intellectual ability• Motivation / drive / study habits / learning
styles / personality (conscientiousness)• Content knowledge (e.g. biology /
chemistry)• Which are the key components for predicting
medical school outcome?• Not clear
– Content – intuitively reasonable– Personality & Motivation – may also matter
• Why A-level chemistry?
A-levels as predictors
Widening access
• Discussion of the issues is not popular
• People do differ, and some of that variationmay well affect the brain
• Socio-economic group (SEG) is complex
– Income
– Education
– Cultural inheritance
– Status
• “Poor” or “working class” make goodheadlines but bad science and bad policy
The Guardian, 26th April 2010
Dr Harry Isenberg (1916-2010)