South African medical schools: Current state of selection criteria and

RESEARCH

76 January 2016, Vol. 106, No. 1

Selection procedures in most medical schools are

necessary because substantially more applications

are received than the number of places available.

Medical schools have a duty to ensure that students

admitted to their programmes have the potential

to successfully complete their studies and display competent and

ethical behaviour in practice.

[1]

In South Africa (SA), there is also an

imperative to ensure that adequate numbers of quality candidates are

attracted and selected, not only to ensure their academic success, but

also to provide future healthcare practitioners who are fit to practise

in the local society, including rural and currently under-served areas

in the country. Selection instruments must therefore be aligned with

social accountability objectives, so that widening access is granted to

those from a disadvantaged background and to ensure access across

the entire demographic profile of SA. At a meeting of representatives

of the South African Committee of Medical Deans, a decision was

made to employ collective learning from current practices across

medical schools in SA to point the way forward.

A variety of indicators, tests, procedures and methods are used to

select candidates for places in medical school, which include both

academic proficiency and non-academic characteristics. Although it

has been argued that certain core competencies important for future

doctors should be considered in the selection process, this does not

necessarily predict academic success,

[2]

nor does it assure that socially

accountable practitioners are produced. At the same time, there is

uncertainty as to whether the widely accepted view that academic

success at high school is the best predictor of academic success at

medical studies holds true.

[3]

Selection procedures at SA medical schools currently employ pooled

data from academic performance tests indicating cognitive ability,

comprising a combination of the National Senior Certificate (NSC)

results in compulsory subjects and the National Benchmark Tests

(NBT), and non-academic performance indicators. The latter include

performance in extracurricular activities (leadership, sport, cultural,

community engagement and service), measures of dis advantage such as

family income, schooling and rural origin, personal reports (biographical

South African medical schools: Current state of selection

criteria and medical students’ demographic prole

L J van der Merwe,

MB ChB, MMedSc, PhD; G J van Zyl,

MB ChB, MFamMed, MBA, Dipl Community Health, Dipl Health

Administration, PhD; A St Clair Gibson,

MB ChB, PhD, MD; M Viljoen,

BComm, MBA; J E Iputo,

MB ChB, PhD; M Mammen,

MSc,

PhD, FAIMER; W Chitha,

MB ChB, MPH Health Economics, AMDP; A M Perez,

DBS, DHSM, MDent; N Hartman,

BA, BSocSc Hons,

MSocSc, PhD; S Fonn,

MB BCh, FFCH, PhD; L Green-ompson,

MB BCh, FCA (SA); O A Ayo-Ysuf,

BDS, MSc (Odont), DHSM, MPH,

PhD; G C Botha,

MA; D Manning,

BSc Hons, MEd, PhD; S J Botha,

BSc Hons, MSc, PhD; R Hi,

MB ChB, MMed, PhD;

P Retief,

BA Hons, MA, DLitt; B B van Heerden,

MB ChB, MSc, MMed; J Volmink,

MB ChB, DCH, FRCP (Edin), MPH, DPhil

Faculty of Health Sciences, University of the Free State, Bloemfontein, South Africa

Faculty of Health Sciences, Walter Sisulu University, Mthatha, Eastern Cape, South Africa

Faculty of Health Sciences, University of Cape Town, Cape Town, South Africa

Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, South Africa

Sefako Makgatho Health Sciences University, Ga-Rankuwa, Gauteng, South Africa

Faculty of Health Sciences, University of Pretoria, Pretoria, South Africa

College of Health Sciences, University of KwaZulu-Natal, Howard College, Durban, South Africa

Faculty of Medicine and Health Sciences, Stellenbosch University, Tygerberg, Cape Town, South Africa

Corresponding author: L van der Merwe (merwelj@ufs.ac.za)

Background. Selection of medical students at South African (SA) medical schools must promote equitable and fair access to students from

all population groups, while ensuring optimal student throughput and success, and training future healthcare practitioners who will fulfil

the needs of the local society. In keeping with international practices, a variety of academic and non-academic measures are used to select

applicants for medical training programmes in SA medical schools.

Objectives. To provide an overview of the selection procedures used by all eight medical schools in SA, and the student demographics

(race and gender) at these medical schools, and to determine to what extent collective practices are achieving the goals of student diversity

and inclusivity.

Methods. A retrospective, quantitative, descriptive study design was used. All eight medical schools in SA provided information regarding

selection criteria, selection procedures, and student demographics (race and gender). Descriptive analysis of data was done by calculating

frequencies and percentages of the variables measured.

Results. Medical schools in SA make use of academic and non-academic criteria in their selection processes. The latter include indices of

socioeconomic disadvantage. Most undergraduate medical students in SA are black (38.7%), followed by white (33.0%), coloured (13.4%)

and Indian/Asian (13.6%). The majority of students are female (62.2%). The number of black students is still proportionately lower than in

the general population, while other groups are overrepresented.

Conclusion. Selection policies for undergraduate medical programmes aimed at redress should be continued and further refined, along

with the provision of support to ensure student success.

S Afr Med J 2016;106(1):76-81. DOI:10.7196/SAMJ.2016.v106i1.9913

RESEARCH

77 January 2016, Vol. 106, No. 1

questionnaires) and interviews. Each medical school employs a particular

combination of these admission criteria, but currently there is no

uniformity in how they are applied across the medical schools.

A positive correlation between past academic achievement

and future academic success has been demonstrated by previous

research.

[4]

However, in a group of applicants with exceptional aca-

demic achievement in certain compulsory subjects at NSC level, it is

not possible to predict among top achievers who is more or less likely

to succeed. The NBT provides an additional method of differentiating

cognitive ability by testing language, mathematical and reasoning

ability. Internationally, tests such as the General Cognitive Ability

test have been shown to have a significant relationship with future

academic and career success.

[1]

In SA, data are lacking regarding the

relationship between the NBT and success at medical school.

Non-academic tests are often used as additional measures to

differentiate between applicants with exemplary cognitive ability, to

ensure that candidates with desirable traits for future medical practice

are not excluded by focusing solely on academic performance.

However, the reliability and validity of non-academic performance as

selection criteria are debatable.

[5]

Scientific evidence of the predictive

value of various selection criteria and processes is needed to guide

future practice.

In SA, in an attempt to address historical inequalities with regard

to access, medical schools attempt to select candidates reflecting the

demographic profile of the country, and therefore aim to increase

the number of black African and coloured students selected for

medical training.

[6]

However, the demographic data regarding racial

composition of student cohorts at individual medical schools may

still not reflect that of the country at large. There are different

demographic distributions in the geographical areas and provinces

of SA, and the number of applicants who comply with minimum

selection criteria in each of these areas may be insufficient to meet the

required targets for transformation. All medical schools, however, are

committed to transformation in their selection practices. The number

of applicants who decline an offer of a place at a particular medical

school because they have been accepted at another institution, poses

difficulties for medical schools to increase the number of students

in specific target groups. In addition to optimal selection processes,

the throughput of selected candidates from first year to graduation

needs to be monitored in order to reduce challenges experienced by

students to complete their studies in the minimum time.

With 2014 marking the 20-year anniversary of democracy in SA,

selection procedures continue moving towards methods of providing

improved access to students from disadvantaged backgrounds.

However, defining disadvantage by race has become complicated by

the fact that there is an emerging black middle class. Factors such as

schooling and educational background (for example, quintile 1 and 2

schools

[7]

), rurality and socioeconomic circumstances have therefore

been included as indicators of disadvantage. Further complexities

exist, such as the emerging middle class and the fact that children

of domestic workers, for example, often attend good schools in the

area where their parents are employed. Students’ financial need,

[8]

the variable quality of primary and secondary schooling

[9]

and

the importance of taking into account cultural differences in the

selection process, are additional issues pointing to the need for a

comprehensive re-evaluation of selection criteria.

Objectives

To provide information regarding the current selection criteria and

medical student demographics at the eight SA medical schools, and

to determine to what extent collective practices are achieving the

goals of student diversity and inclusivity.

Methods

A retrospective, quantitative, descriptive study design was followed.

All eight medical schools in SA, as listed in Table 1, participated in

the study in 2014. Representatives from the South African Committee

of Medical Deans at each institution provided information regarding

selection criteria, selection procedures, and student demographics

regarding race and gender.

Information on selection policies, criteria and procedures for 2014

included both academic and non-academic performance indicators

used to select candidates for medical studies at SA medical schools.

Demographic variables included gender and race (black, coloured,

Indian/Asian, white and ‘other’, where ‘other’ mainly referred to

students from Southern African Development Community (SADC)

countries) of students who were offered places at the respective

medical schools, students who accepted these places, and the

registered students’ demographic characteristics in 2014. Descriptive

analysis of data was done by calculating frequencies and percentages

of the variables measured.

Results

As shown in Table 1, all eight medical schools in SA participated in

the study. All the medical schools offer a 6-year curriculum for the

undergraduate medical degree, with the exception of the UFS, where

a 5-year curriculum is followed. Extended programmes are available

at UCT, UFS, SMU, SUN and UP, and a graduate entry programme

at WITS. UCT has not yet introduced a graduate entry programme,

although one or two students with degrees are offered places in

second year. All the universities offer structured student support

programmes.

In 2014, a total of approximately 1 900 places were available at

these institutions for first-year medical students (UCT 220, UFS 140,

UKZN 250, SMU 250, UP 300, SUN 290, WITS 250, WSU 120). These

places included positions offered to students who have completed the

NSC (matriculants/school leavers), as well as senior students, foreign

students, dean’s/rector’s places, and students transferring from other

medical schools, health sciences professions or degree programmes.

Each university, however, has its own rules regarding the admission

of medical students.

Table 2 provides an overview of the NSC and NBT academic

requirements and the weighting applied at each of the institutions

for admission to their undergraduate medical programmes in 2014.

The demographic composition of selected applicants with regard

to race and gender varies at different medical schools. All medical

schools aim to include students from all races and both genders

in their admissions, and make use of different criteria to achieve

Table 1. SA medical schools and abbreviations (in alphabetical

order)

University Abbreviation

Sefako Makgatho Health Sciences University* SMU

Stellenbosch University SUN

University of Cape Town UCT

University of the Free State UFS

University of KwaZulu-Natal UKZN

University of Pretoria UP

University of the Witwatersrand WITS

Walter Sisulu University WSU

*Formerly University of Limpopo Medunsa Campus.

RESEARCH

78 January 2016, Vol. 106, No. 1

inclusivity in terms of black and coloured students, as well as those

from disadvantaged backgrounds. At UCT, the Senate approves

targets by population group for school leavers annually. No targets

are set for gender, and students from rural areas may enter through

special consideration with slightly lower cut-off points. At the UFS,

which follows a parallel-medium language policy, the selection

committee aims to select English and Afrikaans students in a ratio

of 60:40, while the female/male ratio is 60:40 in order to ensure

that minorities are represented. The ratio with regard to gender is

maintained so that, for example, female students cannot be more than

60% of the selected applicants. A definite attempt is made to include

candidates from disadvantaged backgrounds who demonstrate

academic potential, while two points are allocated for rural origin.

At UKZN, 28% of students are selected from quintile 1 and 2

schools

[7]

without racial quotas, while 52% are selected on merit,

with 69% of these places reserved for black students, 19% for Indian,

9% for coloured, 2% for white and 1% for other (Chinese) students.

The remaining 20% of the total number of students admitted are

selected from students with prior tertiary education. At SMU,

20% of the 250 spaces available for first-year medical students are

reserved for students from the foundation programme (i.e. the

extended curriculum programme). The remaining students are

selected from matriculants and senior students transferring from

other programmes. Of these, 78% of places are offered to black

students, 11% to white, 8% to coloured and 3% to Indian students.

No allowance is made for gender or disadvantage.

At UP, targets are set for historically disadvantaged students,

namely black and coloured, while at SUN male and female students

are selected in the same ratio as the pool of qualifying applicants,

and selection based on race is in accordance with enrolment

planning of the university and its commitment to inclusivity and

diversity. At WITS, selection is made in two racially defined groups,

namely coloured/black and white/Indian, to achieve a representative

demographic distribution.

Academic requirements include NSC results and compulsory

subjects at specified levels of achievement (level 5 (>60%)). These

subjects include Mathematics, Physical Science, Life Science and

Language (English). NBTs are required by UCT (70:30), UFS (60:40),

UP (60:40), SUN (45:30) and WITS (50:50), with varying weight

attributed to these results (NSC:NBT ratios indicated in parentheses).

UKZN, SMU and WSU do not require students to write NBTs in their

selection process.

Table 2. Academic (NSC and NBT) requirements for selection for undergraduate medical training at eight SA medical schools

University Academic (weight)

APS (minimum required

based on NSC)

NCS compulsory subjects + minimum

achievement level NBT requirements NSC:NBT

SMU 100% 30 Level 5 (60 - 69%)

Maths, Physical Science, Life Science

Not required NA

SUN 45% of selection

factor (grade 11/

matric average)

Not applicable Level 6 (70 - 79%) or higher

Level 4 (50 - 59%) or higher

Maths, Physical Science, Life Science

Intermediate or

proficient

Minimum 38% per

component

Weight: 30% of

selection factor

45:30

UCT NSC score out

of 600 and NBT

score out of 300 +

(for Medicine) an

optional personal

report out of 100

450 (APS = NSC) Level 5 (60 - 69%)

Compulsory: English, Maths, Physical

Science, plus next 3 best subjects

excluding LO

Intermediate or

proficient.

70:30

UFS 100 points (84.76%) 36 Level 5 (60 - 69%)

Language, Maths, Physical Science,

Life Science

Must pass each

component and

average 50% overall

60:40

UKZN Aggregate of 4

compulsory subjects

+ 2 best subjects,

excluding Maths 3

and LO

Not applicable Level 5 (60 - 69%)

English, Maths, Physical Science,

Life Science

Average 65%

Not required NA

UP 100% 35 (excluding LO) Level 5 (60 - 69%)

Language, Maths, Physical Science or

Life Science

AL 20%, QL 20%,

Maths 60%

60:40

WITS Composite index:

NSC 40%, NBT 40%,

BQ20%

Not applicable Level 5 (60 - 69%)

English, Maths, Physical Science or

Life Science, plus next 2 best subjects

incl. LO

Required 50:50

WSU Academic 50%;

interview 50%

20 (4 subjects) Level 5 (60 - 69%)

English, Maths, Physical Science,

Life Science

Not required NA

APS = admission point score; LO = life orientation; AL = academic literacy; QL = quantitative literacy; Maths = mathematics; BQ = biographical questionnaire.

RESEARCH

79 January 2016, Vol. 106, No. 1

The non-academic requirements for selec tion

for medical training carry between 10% and

25% of the total weight during the selection

process, and comprise the following:

Extracurricular activities. These include

leadership, community service, cultural

and sporting achievements, of which UCT,

UFS, UP, SUN and WITS include different

components in the selection process.

Region of origin. Although UCT does

not allocate extra points for region of

origin, rural students may be admitted with

lower cut-off points than the general pool

of applicants. UFS offers an additional 2

points and SUN an additional 4 points for

applicants of rural origin, while WSU tends

to admit more of its students from rural

Eastern Cape areas.

Advantage based on parents being

alumni or staff members of the institution

is given in the form of 2 additional points at

UFS and SUN.

Indices of disadvantage are specified

at UCT to include redress for groups 1

and 2 (black and coloured students) who

enter with lower points, redress for group 3

(Indian students) who enter with the same

cut-off point as ‘Open’ students (who must

meet requirements without dropping cut-off

points), and redress for group 4 (Chinese)

who may enter with points slightly lower

than the Open group. At UP disadvantage

is seen as historically based on race, while

at SUN no disadvantage index is used as

it is implicit in the selection process. At

UKZN and SUN (for selection of students

to the extended programme), schooling

background (e.g. quintile 1 and 2 schools

[7]

)

is taken into consideration. At SMU

applicants from quin tile 1 and 2 schools

are considered for the extended curriculum

programme if they have not been selected

directly into the medical programme on

competitive basis.

Biographical questionnaires, personal

reports or interviews. At UCT a personal

report will be required from 2016 if

students achieve at a high level with their

NSC and NBT results. WITS has made

use of a biographical questionnaire prior

to admission for 2015. UFS conducted

interviews in the past, while WSU currently

uses interviews for selected students based

on academic merit. The interview aims to

assess students’ personal attributes.

Fig. 1 shows the student demographic

profile with regard to race and gender of

all undergraduate students in the medical

training programmes offered at SA medical

schools during 2014. Among 9293 students,

male students constitute 37.8% of the total

group, with 62.2% female students. The

majority of students are black (38.7%),

followed by white (33.0%), coloured (13.4%),

Indian/Asian (13.6%) and other (1.3%).

Fig. 2 compares the racial distribution

of undergraduate medical students in SA

in 2014 with demographic findings of the

National Census of 2011.

[10]

Although the majority of medical students

at SA universities are black (approximately

39%), this group is still under-represented in

relation to the national population, of which

80% of individuals are black.

[10]

Medical

students from the other racial groups are

over-represented in comparison with the

national demographic profile. Slightly more

than 33% of medical students in 2014 were

white, while whites represent less than 10%

of the total population. In 2014, nearly 15%

of medical students were coloured, while

the coloured population comprises less

than 10% of the total population in the

2011 census. More than 10% of medical

students were Indian/Asian, while Indians/

Asians represent less than 5% of the total

population.

[10]

At all eight medical schools, large

differences are prevalent in the number of

offers made compared with offers accepted.

Students may receive more than one

admission offer as they may have applied

to more than one university. Therefore, the

large cumulative number of offers made

by the eight medical schools may be due

to duplication. Large numbers of students

4 000

3 500

2 500

3 000

2 000

1 500

500

1 000

Black Coloured Indian/Asian Other White

1 310

2 289

3 599

426

818

1 244

577

687

1 264

64 59

123

1 133

1 930

3 063

Male Female Total

Students, n

Fig. 1. Race and gender of undergraduate students at SA medical schools in 2014.

Black Coloured Asian White

100

79.6

39.2

8.9

13.6

2.5

13.8

8.9

33.4

Census 2011 (N=51.5 m*) UG medical students (N= 9 170

†

)

Fig. 2. Comparison of the racial distribution of the general SA population and undergraduate medical

students attending SA medical schools in 2014. (UG = undergraduate; *51.5 million;

†

UG students

enrolled at eight South African medical schools in 2014.)

RESEARCH

80 January 2016, Vol. 106, No. 1

who are selected and do not take up the

positions offered may therefore skew the

representation of the different population

groups eventually admitted at these

institutions.

For 2014 admissions, only 67% of students

accepted the offers made by medical schools.

Black students made up 49.4% of the total

offers made to students, and 46.9% of those

who accepted these offers. Only 62.4% of

black students accepted the offers made.

White students comprised 27.3% of the offers

made and 27.9% of the offers accepted. In

this group, 61.8% accepted the offers made.

Coloured students comprised 11.6% of the

offers made, and made up 9.5% of those

who accepted offers at medical school. In

this group, 49.3% accepted the offers made.

Of the offers made, 11.4% were to Indian/

Asian students, and of the offers accepted,

11.5% had been made to Asians. In this

group, 61.2% accepted the offers made. Very

low numbers of students in the group ‘other’

(including students from SADC countries)

were offered places, and of these students

only 40.0% (6/15) accepted the offers. There

are no available data showing reasons for the

difference between offers made and eventual

admissions.

Fig. 3 represents the total number of offers

made and accepted in 2013 for admission in

2014 at all eight medical schools.

At all institutions, with the exception of

WSU, the number of students who take up

positions offered to them was lower than the

number of offers made. The largest number

of offers made was to black and coloured

students (53.3%), redressing previous

disadvantage, followed by white (31.3%)

and Indian/Asian (14.8%) students. When

looking at the number of students who

accepted these offers, the majority (56.3%)

were black and coloured students, followed

by white (29.9%) and Asian/Indian students

(13.4%).

Discussion

All eight medical schools in SA use

academic and non-academic requirements

in their selection processes, conforming to

international practice.

[1,5]

In addition, these

institutions aim to improve inclusivity

among the students selected in order to meet

the needs of the diverse socioeconomic and

cultural populations that qualifying doctors

will serve in future. For example, based

on existing evidence,

[11]

targeting students

from rural areas may enhance the number

of rural practitioners. This approach is in

line with the World Health Organization

(WHO) guidelines stating that admission

policies should be targeted to reflect student

diversity in terms of socioeconomic, ethnic

and geographic background.

[12]

We found that overall the demographic

profile of selected students is beginning

to reflect the diversity of the population

groups in SA. Although the proportion

of students in each population group

varies between medical schools (probably

reflecting the demographics of the province

and region where the institution is situated),

the percentage of students from historically

disadvantaged racial groups (black and

coloured students) has improved to a large

degree when compared with pre-1994,

[13]

although less so since 1999.

[6]

The largest

number of students enrolled for training as

doctors at SA medical schools are the black

and coloured groups (52.1%), followed by

white (33.0%) and Indian/Asian students

(13.6%). Black students remain under-

represented in medical schools compared

with the national demographics in 2011,

[10]

while the minority groups are over-

represented.

In an attempt to address past inequality, at

least 60% of offers made by medical schools

are to black and coloured students. Of the

students who accept offers, at least 60% are

black and coloured.

The percentage of female students is

higher than that of male students. Only the

UFS sets a gender ratio of 60:40 female/male

in order to ensure that the minority male

gender among the applicants is represented.

Various indices of disadvantage (such as

quintile 1 and 2 schools,

[7]

rural origin,

disadvantaged population groups) have

been suggested for inclusion, in order to

address disadvantage effectively and to move

beyond race as an indicator of disadvantage,

especially when taking the emerging black

middle class into account. However, more

precise indicators of disadvantage may have

to be developed for use in selection, or for

deciding on students who may not fulfil

the minimum academic requirements set in

selection policies. Furthermore, culturally

sensitive selection tools need to be explored

in order to address issues related to cultural,

socioeconomic and language diversity, which

may affect students’ possible selection. For

example, students from lower socioeconomic

backgrounds may not have access to the

wide array of cultural and sporting activities

or leadership development compared with

those from privileged backgrounds, placing

them at a further disadvantage when these

items are included in non-academic selection

criteria. The erratic quality of secondary

school education in SA

[9]

also leads to

disadvantage, particularly for students

whose schooling is affected by sociopolitical

upheaval, poor teaching proficiency or lack

of adequate learning resources. This is not

limited to schools in rural areas or even

historically black residential areas such as

urban informal settlements.

In order to address socioeconomic dis-

advantage effectively, students from poor

backgrounds must have greater access to

bursaries and financial support for the

duration of their studies. As stated by the

WHO

[12]

in 2013: ‘It is clear that admission

procedures by themselves will not overcome

inequalities in healthcare systems. Where

targeted admission policies are used, support

mechanisms must be in place to ensure

conditions in which students are able to

complete programmes. … Currently many

students who do not complete their courses

Black Coloured Indian/Asian White

Oers, n

1 200

1 000

800

400

600

200

304

1 062

761

154

217

15 6

804

486

378

Oers made in 2013 Oers accepted in 2013

Other/SADC

Fig. 3. Number of oers made and oers accepted in 2013 for admission for medical training in 2014 at

all eight South African medical schools.

RESEARCH

81 January 2016, Vol. 106, No. 1

do so for financial reasons, so disadvantaged students would need

financial support.’

[12]

Conclusion

The demographic profile of medical students selected at SA

universities is moving closer to the population distribution of SA.

However, the percentage of black students is still lower than that

in the general population, while white, Indian/Asian and coloured

students are overrepresented in relation to the general population.

Current policies target black and coloured students for selection

into undergraduate medical programmes across all medical schools

in SA. However, race-based selection policies may be inadequate

for addressing historical inequalities. Indices of disadvantage such

as origin in rural or underserved communities (possibly linked to

poorer educational opportunities) and socioeconomic status should

be considered to improve access. In compliance with international

trends, both academic and non-academic indicators are used in

the selection process. Stronger evidence is now needed to link

throughput, academic success and possibly even future career paths

to selection processes.

It is worth noting, however, that the quality and values of these

graduates – irrespective of where they come from – are influenced by

the training curriculum, quality of teaching and role models they are

exposed to once they are enrolled in medical school.

Acknowledgements. We thank Dr Daleen Struwig, Faculty of Health

Sciences, UFS, for technical and editorial preparation of the manuscript.

References

1. Bore M, Munro D, Powis D. A comprehensive model for the selection of medical students. Med Teach

2009;31(12):1066-1072. [http://dx.doi.org/10.3109/01421590903095510] [PMID 19995169]

2. Koenig TW, Parrish SK, Terregino CA, Williams JP, Dunleavy DM, Volsch JM. Core personal

competencies important to entering students’ success in medical school: What are they and how

could they be assessed early in the admission process? Acad Med 2013;88(5):603-613. [http://dx.doi.

org/10.1097/ACM.0b013e31828b3389]

3. Ross D, Loeer K, Schipper S, Vandermeer B, Allan GM. Do scores on three commonly used

measures of critical thinking correlate with academic success of health professions trainees? A

systematic review and meta-analysis. Acad Med 2013;88(5):724-734. [http://dx.doi.org/10.1097/

ACM.0b013e31828b0823] [PMID 23524925]

4. Kuncel NR, Ones DS, Hezlett SA. A comprehensive meta-analysis of the predictive validity of the

graduate record examinations: Implications for graduate student selection and performance. Psychol

Bull 2001;127(1):162-181. [http://dx.doi.org/10.1037/0033-2909.127.1.162]

5. Benbassat J, Baumal R. Uncertainties in the selection of applicants for medical school. Adv Health Sci

Educ 2007;12(4):509-521. [http://dx.doi.org/10.1007/s10459-007-9076-0]

6. Lehmann U, Andrews G, Sanders D. Change and Innovation at South African Medical Schools – An

Investigation of Student Demographics, Student Support and Curriculum Innovation. Durban: Health

Systems Trust, 2000. http://indicators.hst.org.za/uploads/les/medschools.pdf (accessed 22 June 2015).

7. Western Cape Government, Department of Education. Background to the national quintile system. Media

release 14 October 2014. http://wced.pgwc.gov.za/comms/press/2013/74_14oct.html (accessed 30 June 2015).

8. Naidoo SS, van Wyk J, Higgins-Opitz SB, Moodley K. An evaluation of stress in medical students at

a South African university. S Afr Fam Pract 2014;565(5):258-262. [http://dx.doi.org/10.1080/207861

90.2014.980157]

9. Spaull N. South Africa’s education crisis: e quality of education in South Africa 1994-2011. Report

Commissioned by Centre for Development and Enterprise (CDE), October 2013. http://www.

section27.org.za/wp-content/uploads/2013/10/Spaull-2013-CDE-report-South-Africas-Education-

Crisis.pdf (accessed 30 June 2015).

10. Statistics South Africa. Census 2011. Statistical release (revised) P0301.4. www.statssa.gov.za/

publications/P03014/P030142011.pdf (accessed 28 July 2015).

11. Grobler L, Marais BJ, Mabunda SA, Marindi PN, Reuter H, Volmink J. Interventions for increasing the

proportion of health professionals practising in rural and other underserved areas. Cochrane Database

Syst Rev 2009, Issue 1. Art. No.: CD005314. [http://dx.doi.org/10.1002/14651858.CD005314.pub2]

12. World Health Organization. Transforming and scaling up health professionals’ education and training.

Geneva: WHO, 2013. http://apps.who.int/iris/bitstream/10665/93635/1/9789241506502_eng.pdf

(accessed 30 June 2015).

13. Digby A. Black doctors and discrimination under South Africa’s apartheid regime. Med Hist

2013;57(2):269-290. [http://dx.doi.org/10.1017/mdh.2012.106]

Accepted 28 September 2015.