International Journal of Human and Health Sciences Vol. 04 No.

02 April’20

Review article:
Multiple Choice Questions in Medical Education: How to Construct High Quality Questions
Abdus Salam1. Rabeya Yousuf2, Sheikh Muhammad Abu Bakar3
Abstract:
Multiple choice questions (MCQ) are the most widely used objective test items. Students
often learn what we assess, and not what we teach, although teaching and assessment are
the two sides of the same coin. So, assessment in medical education is very important to
ensure that qualified competent doctors are being produced.A good test is the test that
assesses higher level of thinking skills. Many inhouse MCQs are found faulty which
assess lower level of thinking skills. The main problems in constructing good MCQs are
that (i) very few faculty members have formal training in questions construction, (ii) most
of the questions are prepared in the last minutes where little time exist for vetting to review
the quality of questions and (iii) lack of promise on the standard of the question format
and underestimation of the use of blueprint in medical schools. Constructing good MCQs,
emphasis should be given that, the stem is meaningful and present a definite problem, it
contains only relevant material and avoid negativity. It should be ensuring that, all options
present as plausible, clear and concise, mutually exclusive, logical in order, free from clues
and avoid ‘all of the above’ and ‘none of the above’. The MCQs can tests well any higher
level of the cognitive domain, if it is constructed well. Efforts must be made to prepare
and use of test blueprint as a guide to construct good MCQs. This paper describes and
offers medical teachers a window to a comprehensive understanding of different types and
aspects of MCQs and how to construct test blueprint and good MCQs that tests higher
order thinking skills in the future medical graduates, thereby ensures competent doctors
are being produced.
Keywords: MCQs, construction, high quality, assessment, higher order thinking skills,
test-blueprint.
International Journal of Human and Health Sciences Vol. 04 No. 02 April’20 Page : 79-88
DOI: http://dx.doi.org/10.31344/ijhhs.v4i2.180

Introduction: common type of written test item used in medical

Assessment in medical education is very education.
important to ensure qualified competent doctors Multiple-choice question: Multiple choice
to be produced. However, no single method is questions are most familiar as the widely
appropriate for assessment of skills, knowledge applicable and valuable type of objective test
and attitudes needed in medicine, so a combination items2,3, introduced into medical examinations
of assessment techniques is always required1. since 19501. Such a test usually consists of a
Among the various test methods, multiple choice number of items that pose a question to which
questions (MCQs) are widely used to assess students must select an answer from among a
learning and knowledge. These are the most number of choices4. An appropriately constructed

1. Abdus Salam, Medical Education Unit, Faculty of Medicine, Widad University College, Kuantan,
Malaysia.
2. Rabeya Yousuf, Blood Bank Unit, Department of Medical Lab Services, Faculty of Medicine,
Hospital University Kebangsaan Malaysia, Cheras, Malaysia.
3. Sheikh Muhammad Abu Bakar, Department of Anatomy, Faculty of Medicine International Medical
College & Hospital,Gazipur, Bangladesh.

Correspondence to: Abdus Salam MBBS (Ctg), MPH (Dhaka), DipMedEd (Dundee), MMedEd
(Dundee). Head of Medical Education Unit and Manager -Academic Scheduling & Curriculum,
Faculty of Medicine, Widad University College, Kuantan, Malaysia, ORCID iD: 0000-0002-2629-4903
Phone: 06 0199061724(H/p), E mail: [email protected]
79
 International Journal of Human and Health Sciences Vol. 04 No. 02 April’20

MCQs result in objective testing that can measure organize their thoughts or express their creative
knowledge, comprehension, application, and ideas6. It allows students to perform excellently
analysis. The entire test question is called the well with repeated questions and there are
item, which consists of a stem and several options. chances of errors in collation if answers (to 100s
The stem is the question, statement, or lead-in of questions) are manually marked9. Construction
to the possible answers. The possible answers of MCQ is most difficult and time consuming even
are called options, alternatives, or choices. The for the experienced and trained personnel. There
correct option is called the keyed response. The is evidence that the basic item-writing principles
incorrect options are called distractors or foils5. A are not followed mostly while constructing
good distracter needs to be inferior to the correct MCQs. Flawed MCQs interfere with accurate
answer but should be able to be plausible to a non- and meaningful interpretation of test scores and
competent candidate. negatively affect student pass rates. Therefore,
There are many advantages as well as limitations to develop reliable and valid tests, items must be
of using the MCQs. It has an excellent reliability constructed that are free of such flaws.
which is the reproducibility or consistency of a Types of MCQ
test score and can be efficiently employed to a MCQ can be presented in various format,
large number of students for stratifying individual commonly used format are ‘one correct answer’,
performance5. Their scoring is easier to understand ‘single or one best answer (SBA/OBA)’, ‘true or
than short-answer test scoring because there is no false’ and ‘multiple true or false’, ‘matching’ and
need to resolve partial and misspelled answers. the ‘extended matching questions or items (EMQ/
Furthermore, MCQ test assessment is more EMI)’ and ‘case or problem’ format or type4. The
objective than the assessment including essay most popularly used one is the SBA/OBA type,
questions. The essay test scores can be affected also called A-type MCQ which asks the students
by the examiner’s inconsistencies and are not to choose the best answer from 4-5 options10,11and
immune to the influence of bluffing and writing the second more popular is the EMQ/EMI, also
ability factors, which can lower their reliability. known as R­-type MCQ10.
The usage of MCQ tests is very important for
i) One correct answer format consists of two
the examiners because they allow easy and quick
parts: (i) the stem, which identifies the question
evaluation. These tests are particularly essential
or problem and (ii) the responses alternatives.
for the examiners who cover multiple courses
Students are asked to select the one alternative
with large number of enrolments6. MCQs can
that best completes the statement or answer
provide a large number of examination items that
the question. For example: Which one of the
encompass many content areas, so a broad domain
following antibodies is involved in Type 1
can be covered. They are time efficient as can
hypersensitivity? (A) IgA, (B) IgD, *(C)
be administered in a relatively short period and
IgE, (D) IgG, (E) IgM.
graded by a computer and have a high reliability7.
It can also assess taxonomically higher-order ii) Single or one best-option MCQs are the most
cognitive processing such as interpretation, versatile and widely used MCQs. This format
application and evaluation rather than the test of refers to a list of options that can all be correct
recall of isolated facts. Thus, they could test a in the sense that each has an advantage, but
number of skills in addition to the recall of factual one of them is the best. The example is shown
knowledge, and are reliable, reproducible and in Table 5.
cost-effective1. However, it is difficult to construct iii) True-false questions contain statements that
the question with little or no experience or training the student marks as being either true or false.
and examiners need to spend considerable time The purpose of this question is to test whether
and effort for this purpose1. The MCQ allow for students are able to evaluate the correctness
guessing without knowing the actual fact. Also, it of an assumption12. All parts of the statement
does not provide a measure of writing ability and have to be absolutely true or false in order
the scores can be influenced by reading ability8. to qualify as true or false. It is difficult to
It does not assess communication skill which is construct flawlessly and in general, true-false
a very important part of medical education. It questions check knowledge of facts for which
also does not assess writing skills. Therefore, this they are best avoided4,12. An example of true-
is not an effective way to test students’ ability to false question is: Anabolic Steroids provide
80
 International Journal of Human and Health Sciences Vol. 04 No. 02 April’20

an unfair advantage in sports where strength scenario or vignette also called the stem.
plays a large part. *(A) True (B) False13. There needs more than one, at least two
iv) Multiple true / false questions consist of a clinical problems / scenario or vignette for
stem which can be a statement, question, each theme. The candidate is asked to match
case history or clinical data followed by five one or more options to each item stem1.
true or false options; candidate has to answer Extended matching items are most commonly
separately to each of the five options as true used in medicine as they can be used to test
or false9,14. Multiple true false questions reveal diagnostic ability and clinical judgment1. The
the limits of the SBA format in that the student example is shown in Table 8.
must think and evaluate each possible answer, vii) Case or problem format can use a real or
rather than selecting the one best answer15,16. realistic resource such as documents or images
It is easy to change wording delicately, so to use to answer the question. This type of
that a true statement becomes false or vice questions can effectively test the application
versa16. It can cover breadth of knowledge of knowledge.
(widely sample) though may not assess depth Guidelines for Writing MCQs: It is important
of knowledge. In a paper comprising 60 to use good grammar, punctuation, and spelling
questions, 300 different items of knowledge consistently and minimize the time required
can be tested. For each true answer student to read each item. The ideal question will be
will get 1 mark, for each false answer get answered by 60-65% of the tested population. In
-0.5 mark, there will be no mark for options a well-constructed MCQ, unintended cues should
not attended3. Time required to answer each be avoided such as making the correct answer
question is about 70 seconds. There needs longer in length than the distractors. Instruction for
carefulness while formulating, if not it may answering the MCQ could be common for a set of
test only recall. Terminologies such as never, questions. For example, for the single best response
none, always, all, impossible tend to be false type, the instruction could be: “Select the most
while sometimes, often, usually, generally appropriate answer and darken the corresponding
tend to be true, as such both should be avoided. circle in the answer sheet provided”14. A good
The example is given in Table 6. distracter needs to be inferior to the correct
v) Matching type: In this format, students need answer but should be plausible or as nearer to the
to match the two lists of statements or words, correct answer as possible and yet maintaining the
with specific instructions. The two lists should correctness of the right answer11. The number of
contain different number of items to avoid options in MCQ varies from three to five. In case
cueing9. The advanced version is extended of three options, the chances of guessing are as
matching questions or items (EMQ/EMI) high as 33%, in case of five options one or two
type. distractors tend to be non-functional. It is suitable
vi) Extended matching questions/items (EMQ/ to use four or five alternatives depending upon the
EMI) consists of a number of options and a availability of good distractors11.
series of questions where the answers are Writing objectives : In order to prepare an MCQ
needed to be selected or matched from the test, it is important to consider the sampling
options. A typical set of EMQ/EMI begins of objectives and the content areas to be tested.
with an option list of four to 26 options; eight It is necessary to choose an important learning
to ten options are usually used, followed by outcome before constructing an MCQ. Therefore,
two or more patient-based items requiring the the item in an MCQ test should be addressed to
examinee to indicate a clinical decision for test a significant learning outcome11. The MCQs
each item1. But they are not easy to prepare. drawn from a representative sample of content
Generally EMQ/EMI comprises of four areas of predetermined learning outcomes will
parts (a) a theme, which include a symptom, allow for a high degree of test validity5. Learning
investigation, diagnosis or treatment, e.g. back objectives are statements of intent or desire which
pain, dyspnoea, diabetes, corticosteroids, (b) we expect to achieve by our learners at the end
an “extended” list (A-H) of possible answer of any educational program. Learning objectives
options (c) the question, also called the lead- should be written as much as possible to SMART,
in statement, and (d) a clinical problem / an acronym for specific, measurable, attainable,
81
 International Journal of Human and Health Sciences Vol. 04 No. 02 April’20

realistic and time-bound17,18. the objective it is designed to assess.

Benjamin Bloom : There are many threats to the validity of any test
Benjamin Bloom, an educational psychologist which can be categorized into (i) Content under­
divided what and how we learn into three separate representation(CU), meaning to under sampling
domains of learning: Cognitive domain–related of the content/domain and (ii) Construct
to thinking/knowledge (K), Affective domain– Irrelevance variance (CIV) which refers to the use
related to feeling/attitudes (A), psychomotor of inappropriate adjustments (flawed items, ill-
domain–related to doing/skills or practice (P). In structured items) which interfere with the ability
1956, Bloom published a taxonomy of cognitive to meaningfully interpret scores or ratings10.
learning, which was described as a hierarchy of (i) The content under­representation can be avoided
knowledge, (ii) comprehension, (iii) application, by the use of test blueprint10.A test blueprint
(iv) analysis, (v) synthesis, and (vi) evaluation19.In is also calledtest specification table which is
2001, 45 years later, the last two taxonomies i.e., a document that reflects the assessment. It
synthesis and evaluation revised as (v) evaluation contains (i) content /objectives to be tested (ii)
and (vi) creation20. Buck Walters’ Taxonomy the questions that designs to test the content/
describes learning at three level as follows: Level objective (iii) learning domain and levels therein
I:Recall of information, Level II: Comprehension testing. Blueprint is a three-dimensional chart
and application and Level III: Problem solving which shows the placement of each question in
ability -involving analysis, evaluation and respect of the objective and the content areathat
creation21. The Table-1 below shows the revised it tests22,23.The purpose of making blueprint
Blooms’ cognitive taxonomy that suitable to test is to provides a solid foundation for all test
different levels of learning by MCQs20. development activity thatoffersan evidence for
Table-1: Suitability of testing revised Blooms’ the content related validity; it is a strategic plan
cognitive taxonomic level learning by MCQs. in order to ensure that the assessment is more
meaningful10,23,24. The Table-2 shows an example
Revised Blooms’ Cognitive Taxonomy
of test blueprint for better understanding.
Creating Level-VI
Higher level, not suitable How towrite the Stem : The basic rule of writing
for MCQs
Higher
stem is that students should be able to understand
Evaluating Level-V order the question without reading it several times and
learning without having to read all the options14. The stem is
Analyzing Level-1V Higher order learning that
can be tested with MCQs the question or statement, or lead-in to the possible
Applying Level-1II
answers which is usually composed first and is best
Understanding Level-II Lower written as a complete sentence or question with all
Lower order learning is
order
what most MCQs target the relevant information and avoid repetition in the
Remembering Level-1 learning

MCQs written to test knowledge (lower-level options. Any maps, diagrams, graphs, or images
learning) would not be appropriate to test can also be incorporated in the stem which should
competence for objectives that reflect analysis be accompanied by a complete statement. Stems
(higher-level learning). For example, an MCQ should be written in a simple language so that the
that asks the learner to recognize benign dermal stem is not a test of the examinees’ reading ability.
calcifications on a mammogram does not test It should not be tricky or misleading, such that
the learners’ problem-solving ability. A question they might deceive the examinee into answering
that provides specific patient information and incorrectly5.Lead-in part of the stem should
imaging data (i.e., a patient vignette) and that be construct in such a way that it builds on the
asks the learner to choose the most appropriate information in the stem and pose a clear question
management is an example of an item that tests or a statement25,26. The stem should be stated so
problem solving ability. Objectives should define that only one of the options can be authenticated
important knowledge or skills and should be and that option should be indisputably correct.
supported by the instruction provided through When more than one option has some element
the educational program. Observable, measurable of truth or accuracy but the keyed response is the
objectives allow for accurate assessment of best, the stem should ask the student to select the
whether the learner has achieved the objectives. best answer rather than the correct answer5. As a
MCQ should test at the same level of learning as general guide, students can complete between one

82
 International Journal of Human and Health Sciences Vol. 04 No. 02 April’20

Tables-2: Example of a test blueprint or test specification table

Test Blueprint or Test Specification Table
Level of Blooms’ Cognitive Domain Total question
Questions Format Questions Format Questions Format
Objectives
measuring recall / measuring measuring
comprehension application/ analysis evaluation / creation % test
devoted to
Contents S M M S S M M S S M M S n topic
/Topics to be B T E A B T E A B T E A
tested A F Q Q A F Q Q A F Q Q
Topic-1
1 1 0 2 2 0 2 2 1 0 0 0 11 18

Topic-2
1 2 1 1 2 2 2 1 1 0 0 2 15 25

Topic-3
2 0 1 2 1 1 1 3 1 1 0 0 13 22

Topic-4
2 1 1 2 2 3 1 3 2 2 1 1 21 35

Total
6 4 3 7 7 6 6 9 5 3 1 3 60 100
question
% test devoted
10 6 5 12 12 10 10 15 8 5 2 5 100
toeach level

SBA=Single best answer; MTF=Multiple true false; MEQ=Modified essay question; SAQ=Short answer question
and two MCQs per minute . 5
(iv) Duration of symptoms: (e.g., that has
Central idea should include in stem instead of the continued for 3 days).
choices. Any excessive, unnecessary and irrelevant (v) Patient health history, such as: family history,
statements in the stem should be avoided. Language chronic illness, previous attacks, current
should be clear and straightforward; complex or treatments.
imprecise wording may become a test of reading (vi) Physical findings: examination findings,
comprehension rather than an assessment of the as results of diagnostic studies, as initial
subject matter. Imprecise terms such as seldom, treatment, subsequent findings, etc.
rarely, occasionally, sometimes, few, and many are The first three items are vital in the vignette and
not uniformly understood and should be avoided5. the other depends on the scenario and the question
Word the stem positively and asked a direct to be asked.
question (?) or present a definite statement using a How towrite the Options : The options are
colon (:) or a semicolon (;). Avoid negative words the possible answers; they are also called as
such as not or except; as negative questions tend to alternatives or choices or foils. Three to four option
be less effective and more difficult for the examinee items are effective while more than five options
to understand. If negative stem in some instances is burdensome and often leads to faulty options
cannot be avoided, ensure that the word appears while increasing the reading demands of the
capitalized such as NOT or EXCEPT, e.g. Which student. All the options need to keep homogeneous
one of the following is NOT mechanism of VSD? in content and of similar consistent length5.
To develop a good a clinical problem /scenario Options should be in logical or numerical order;
or vignette, it should include all or part of the avoid repetition in the options; length of options
following parts10. about to be equal; only one of options be the right
(i) General information about a patient, such as: answer and all are plausible. Options should be
age, gender, socio economic status, education independent, not be overlapping10. Avoid clues to
the right option such as always, never, completely,
level. etc. (e.g., A 40 y­ ear­old male, low socio-
absolutely, may, could, can and resembling words
economic class).
in the stem, as test wise examinees will know
(ii) Site of care, for example: hospital, health that almost anything is possible. Avoid obvious
center, department, clinic (e.g., comes to the correct as well as irrational options. Avoid all of
outpatient clinic). the above and none of the above. Using ‘all of the
(iii) Presenting complaint: (e.g., complaining of above’as correct option, students are exposed to
chest pain). a lot of correct information, but answers may be
83
 International Journal of Human and Health Sciences Vol. 04 No. 02 April’20

more obvious /noticeable, hence learning comes Table-4: Example of SBA type of MCQ assessing
from recall processing; the examinee only need higher order thinking level-V “Evaluation”.
to have partial information in order to answer the
Table-4.
question. If they know that only two of the options
are correct in a four or five choices question, they A 78-year-old man presents to the Emergency Department with
sudden onset of left sided chest pain radiating to his back. He
can determine that all of the above is the correct has history of stable angina and peripheral vascular disease.
answer option. Equally, students only need to His blood pressure is 80/50 mm of Hg with a heart rate of 120/
minute28,29.
eliminate one answer option as implausible in order Which one of the following tests would most likely confirm the
to eliminate ‘all of the above’ as an answer option. diagnosis?
Using ‘none-of-the-above’ as correct, students are A. Chest radiography
exposed a lot of incorrect information and may not B. Computerized tomography of the chest
C. Echocardiography
need to retrieve correct information. Both question D. Electrocardiography
types can be harmful to accurate assessment and
Comments: MCQs can assess well, any higher level of the
possible benefits to learning are small27. Avoid cognitive domain, if it is constructed well. It depends mainly on
grammatical inconsistencies that cue the test the level of competency in the construction of the MCQ items.
taker. Options should start with capitalizing the Table-5: The following tables (5a, 5b, 5c) shows SBA
first alphabet responding to a direct question and
type of MCQs using different style of lead-in such as
should start with capitalizing or uncapitalizing of
first alphabet depending on whether the statement using direct question (?), using statement ends with
ends with colon or semi colon. First, key answer colon at the end (:) or using statement ends with semi
of the question needs to be identified followed colon (;)at the end.
by distractors assuring that they are plausible.
Table-5a: Example of SBA type of MCQ asking for appropriate
Distractors should directly relate to the lead- option using lead-in as direct question (?)
in question or statement. Applyingappropriate
A 20-year-old lady presents to a clinic with a history of lower
distractor is a challenging task for the examiners abdominal pain, abnormal vaginal discharge, dyspareunia
and needs lots of efforts in construction. associated with fever, nausea and vomiting for 5 days.
What is the most likely diagnosis?
Some examples of MCQs with different types A. Acute appendicitis
B. Acute pyelonephritis
or formats, context in the stem, style, options, C. Ovarian torsion
technical flaws and improved versions are D. Pelvic inflammatory disease
E. Ruptured ovarian cyst
shown here:
Comments: The options or choices starts with capitalizing the
Table-3: Example of SBA type of MCQ with first alphabet of the options answering to a direct question.
context-free and context-rich stem.
Table-5b: Example of SBA type of MCQ asking for appropriate
Table-3. option using lead-in as a statement ends with colon (:) at the
end.
Context-rich stem /improved
Context-free stem
version A 20-year-old lady presents to a clinic with a history of lower
abdominal pain, abnormal vaginal discharge, dyspareunia
A 1­year­old baby who is a known associated with fever, nausea and vomiting for 5 days.
case of VSD, was admitted to the
ER by his mother, complaining of The most likely diagnosis is:
cough, shortness of breath which A. acute appendicitis
disturbs his feeding for the last B. acute pyelonephritis
What is the most important 3 days. On examination the baby C. ovarian torsion
mechanism of Heart Failure was found ill, his temperature D. pelvic inflammatory disease
in achild of one-year old10? was 38.7C, his pulse was 110/ E. ruptured ovarian cyst
min, which is regular, dyspneic
A. Pressure overload with respiratory rate of 70/min Comments: The options or choices starts with small letter
B. Ventricular dilatation and intercostals recessions, and alphabet relating to answer to the statement ends with colon (:)
has a liver was 4 cm below the
C. Ventricular dysfunction
costal margin10. Table-5c: Example of SBA type of MCQ asking for appropriate option
D. Ventricular weakness using lead-in as a statement ends with semi colon (;) at the end.
E. Volume overload What is the most likely
mechanism of his symptoms? A 20-year-old lady presents to a clinic with a history of lower
A. Pressure overload abdominal pain, abnormal vaginal discharge, dyspareunia
B. Ventricular dilatation associated with fever, nausea and vomiting for 5 days.
C. Ventricular dysfunction
D. Ventricular weakness The most likely diagnosis would be;
E. Volume overload A. Acute appendicitis
B. Acute pyelonephritis
Comment: Context-free version asked totally recall of information, C. Ovarian torsion
while the context-rich one is stimulating to think and it resembles D. Pelvic inflammatory disease
the real-life situation which the students will face in the future. E. Ruptured ovarian cyst

84
 International Journal of Human and Health Sciences Vol. 04 No. 02 April’20

Comments: The options or choices starts with capitalizing the Table-7: Example of MCQs (Table-7a) SBA type
first alphabet relating to answer to the statement ends with with repetition of word/phrase in the options with
semi colon (;) which similar to responses answering to direct
question. improved version, and (Table-7b) SBA type with
overlapped options including improved version.
Over all comments: How many options will be accepted whether
4 or 5 need to be determined by the faculty in consensus.
Also, answering to the lead-in statement ends with colon (:) and Table-7
semicolon (;), in which aspect option will start with capitalizing
first alphabet and in which aspect will start without capitalizing
Repetition avoided /
the first alphabet should be standardized by the faculty in Repetition of word /phrase in
improved version
consensus. options

The following action that

Table-6: Example of MCQ of multiple true-false Which of the following action would decrease the radiation
(MTF) type/format would decrease the radiation dose dose from the chest CT least,
from the chest CT least?5,28 by decreasing;
Table-6. Multiple true-false type of MCQ. Here the examinee
requires to answer all options that are appropriate as true or false. A. Decreasing kVp from A. kVp from 140 to
140 to 120 120
B. Decreasing mA from B. mA from 250 to
A 3-year-old girl is brought to the ED by her mother who presented
250 to 125 125
with shock. C. Decreasing the pitch C. the pitch from 2
from 2 to 1 to 1
What are common causes of shock in this age group? D. Decreasing the scan D. the scan time 1
A. Arrhythmias time 1 to 0.5 to 0.5

B. Cardiac tamponade
C. Gastroenteritis
D. Intussusception
Comments: In improved version, repetitions of word in all options
E. Peritonitis are avoided and used the stem as statement instead of direct
question.

Comments:The above question is asked using a lead-in as direct

question (?)

Table-7b:
The question may be asked using a lead-in as statement ends with
colon (:), where the options will start with small letter. For example;
Options are not mutually Improved version, while
The common causes of shock in this age group are: exclusive, while D is correct answer A is correct, not
A. arrhythmia answer, overlapped with A, B overlapped with other options
B. cardiac tamponade and C30.
C. gastroenteritis
D. intussusception
How many chromosomes are How many chromosomes are
E. peritonitis
found in a typical human cell? found in a typical human cell?
A. 18 A. 46
The question may also be asked using a lead-in as statement ends
B. 22 B. 48
with semi colon (;), where the options will start with capitalizing
C. 32 C. 50
first alphabet, similar to responses answering to direct question. For D. 46 D. 52
example; E. 54 E. 54

The common causes of shock in this age group;

A. Arrhythmia
B. Cardiac tamponade Comments: A cell that contain 22 chromosomes also contain
C. Gastroenteritis 18; a cell that contain 32, also contain 22 and 18 etc. and so
D. Intussusception on. Overlapped options may be considered as “trick” items by
E. Peritonitis examinee and excessive use of which can erode trust and respect
for the testing process.
The acceptability of the style of lead-in in the stem need to be
decided by the faculty in consensus.

85
 International Journal of Human and Health Sciences Vol. 04 No. 02 April’20

Table-8: Example of extending matching Mr. X is a patient admitted to the A & E in shock. After
questions or items (EMQ/EMI)26. 10 minutes, the doctor observes an ECG abnormality and
subsequently a ventricular fibrillation (VF). Mr. X is unconscious
Table-8. and notbreathing. Which one of the following is mostreliable sign
Theme: of cardiac arrest10?
Back pain
A. Absence of breathing.
Option list: B. Absence of pulse.*
A. Ankylosing spondylitis. C. Unconsciousness.
B. Aortic dissection. D. All of the above
C. Intervertebral disc infection.
D. Lumbar spondylosis.
E. Metastatic malignancy. Comments: Unnecessary information; unclear or ambiguous
F. Pars interarticularis defect. information e.g. ECG abnormality and the use of ‘all of the above’
G. Prolapsed intervertebral disc. in the optionsis a flaw.
H. Vertebral fracture.

Lead-in statement:
For each patient with back pain, select the most likely diagnosis. Table-9d:

Stems: Which one of the following factors is NOT associated with

Question 1 Type-2 diabetes10?
A 23-year-old man has a 6-month history of lower back
pain. His pain is predominantly at the thoracolumbar A. IGT and IFG
junction and in the right buttock. The pain is worse in the B. Obesity
morning and he has difficulty in getting out of bed. There C. Physical inactivity
is some improvement during the day. Examination shows D. Viral infections. *
restriction of lumbar spinal movements, particularly
lateral flexion. (Answer: A)
Comments: Use of negative (NOT) in the stem is not
recommended. Stem also not context-rich.
Question 2
A 32-year-old lady presents with acute onset of low
back pain. The pain is constant and is not significantly
affected by posture. All spinal movements are painful Table-9e:
and difficult. Three weeks earlier, she had a urinary
tract infection, which had been treated with amoxicillin. You, as a resident doctor assessing a patient with brainstem injury
(Answer: F) in a tertiary care hospital. After performing the Glasgow Coma
Scale, which one of the following would be the next step in your
assessment plan10?
Table-9: The following table shows different
A. Conduct lumbar puncture.
examples of MCQ with technical item flaws. B. Perform arterial blood gas measurement
C. Check cranial nerve functioning and respiratory rate.*
Table-9a: D. Perform a pulmonary wedge pressure measurement.

Which one of the following strategies is the best for prevention Comments: Logical cue is that the word “cranial” used in option C
of type-2 Diabetes Mellitus10? is logically homogeneous with the word “brainstem” in stem. The
options are not in chronological order.
A. High-risk strategy.
B. Population, high risk and screening strategy.*
C. Population-based strategy
D. Screening strategy. Assessment of students is a matter of continuing
concerns for medical educators31. The major
Comments: Merging options and longest option is the correct problems in constructing MCQs are that (i) very
option. Context-free stem.
few faculty members have formal training in
questions construction, (ii) most of the questions
Table-9b: are prepared as context-free in the last minutes,
where little time exist for vetting or review the
What is the target population for primordial preventionactivities10? quality of questions before examination and
(iii) may be no agreement on the standard of
A. High risk groups
B. Patients with complications. the question format and underestimation of the
C. Patients. use of blueprint of assessment by teachers in the
D. Total population.*
medical schools. The core part of the student life
Comments: Repetition of word in the stem and key response
is assessment32 and students always learn what we
(correct option); while it is acceptable to repeat the word in the assessnot what we teach10, although teaching and
stem and distracter10. assessment are the two sides of the same coin. The
The stem is also a context-free stem which assess recall of facts.
assessment strategies direct and influence the way
Table-9c:
students learn31. A good assessment is the test that
assesses higher level of thinking skills. Students
86
 International Journal of Human and Health Sciences Vol. 04 No. 02 April’20

need some scenario or information in assessment so, efforts must be made to construct high quality
tools to solve a problem which assess higher level MCQs.
skills and which should not be just recall of the This paper offers medical and health professional
fact10. Constructing a good scenario, emphasis teachers a window to a comprehensive
need to be given so that, the stem should present understanding of different types and aspects of
a definite problem to be meaningful and contain
MCQs and how to construct good MCQs that tests
only relevant material, it should be a question or
higher order thinking skills in the future medical
statement, and avoid negatively stated question
and health professional graduates, thereby ensures
as examinee may find incorrect answer without
knowing the correct33. All options should be competent doctors and other health professionals
plausible, clear and concise, mutually exclusive, are being produced. Constructinghigh quality
homogeneous, should not include ‘all of the above’ MCQs, institutions must emphasize on faculty
and ‘none of the above’, should be in logical order development programs concentrating on question
to avoid bias for certain position, should free from construction and implementation of test blueprint
clues30. Although there are criticisms that, MCQs or test specification table. A test specification table
tests only recall of information, it is evidenced can guide the test development process especially
that MCQs can tests well any higher level of the item writing efforts and test assembly that provides
cognitive domain, if it is constructed well. The strong content validity and legal defensibility of
criticism is basically due to poor construction; the examination process10.

References: 543-551.
1. Al-Rukban MO. Guidelines for the construction of 6. Jovanovska J. Designing effective multiple-
multiple choice questions tests. J Family Community choice questions for assessing learning outcomes.
Med. 2006; 3(3): 125-133. Infotheca 2018; 18(1): 25-42. DOI 10.18485/
2. Salam A, Rahim AFA, Aziz RA, Fakir NMRM, infotheca.2018.18.1.2.
Jaafar R. Assessment of the students: tools used 7. Amin Z, Seng CY, Eng KH. Practical Guide to
in UniversitiSains Malaysia. Bangladesh Medical Medical Student Assessment. World Scientific
Journal 2005;34(1):11-13. Publishing Co. Pte. Ltd. Singapore, 2006.
3. Menon PSN, Multiple choice question construction. 8. Zimmaro DM. Writing Good Multiple-Choice
In: Sood R, Paul VK, Sahni P, Mittal S, Kharbanda Exams. Faculty Innovation Centre. University of
OP, Adkoli BV, Verma K, Nayar U (Eds.). Assessment Texas – Austin. 2016. Retrieved on 10 November
in medical education. Trends and tools. K L Wig 2019 from: https://facultyinnovate.utexas.edu/sites/
Centre for medical education and technology. All default/files/writing-good-multiple-choice-exams-
India institute of medical sciences. New Delhi-India. fic-120116.pdf.
SEARO WHO project WR/IND HRH 001/LCS, 1995. 9. Asani M. Assessment methods in undergraduate
4. Sood R, Singh T. Assessment in medical education: medical schools. Niger J Basic Clin Sci 2012;
Evolving perspectives and contemporary trends. The 9(2):53-60.
National Medical Journal of India 2012; 25(6): 357-364. 10. Abdalla ME, Gaffar AM, Sulaiman RA. Blueprints
5. Collins, J. Writing multiple-choice questions for in Health Profession Education Series. Constructing
continuing medical education activities and self- A-Type Multiple Choice Questions (MCQs): Step By
assessment modules. Radio Graphics 2006; 26(2): Step Manual, 2011.
87
 International Journal of Human and Health Sciences Vol. 04 No. 02 April’20

11. Srinivasa DK, Adkoll BV. Multiple Choice Questions: Hungund. Blueprinting in Assessment:How much
How to Construct and How to Evaluate? Indian J is imprinted in our practice? Journal of Educational
Pediatric 1989; 56: 69-74. Research & Medical Teacher 2014; 2(1):4-6.
12. Schuwirth LW, van der Vleuten CP. Different written 23. Adkoli B. Attributes of a Good Question Paper. In. R.
assessment methods: What can be said about their Sood (Ed), Assessment in Medical Education: Trends
strengths and weaknesses? Med Educ 2004;38:974- and Tools New Delhi. KL Wig Center for Medical
979. Education and Technology. AIIMS, 1995.
13. Sian Millard, BenChavez. Writing Multiple Choice 24. Downing SM, Haladyna TM. Validity and its threats.
and True/False Exam Questions. A good practice In: Downing SM, Rachel Yudkowsky (Editors).
guide. University of Hawai’i at Hilo. January 2012. Assessment in Health Professions Education.
Retrievedon 10 November 2019 from Routledge, New York, 2009.
h t t p : / / u h h c o p f a c u l t y r e s o u r c e . w e e b l y. c o m / 25. Featherstone C&Hurst Y. (2014), adapted from Case,
uploads/2/1/9/8/2198211/multiple_choice_and_true_ S. M and Swanson, D. B. (2001). Constructing Written
false_exam_question_design_booklet.pdf. Test Questions for the Basic Clinical Sciences,
14. Chaudhary N, Bhatia BD, Mahato SK, Agrawal KK. NBME: Philadelphia. SBA writing guidelines. RCSI
Multiple Choice Questions- Part II (Classification, Joint committee on intercollegiate examination.
Item Preparation, Analysis And Banking). Journal of Policy No. G33, v.1.0; 2016.
Universal College of Medical Sciences 2014; 2(3): 26. Case SM, Swanson DB. Extended-matching items: A
54-59. practical alternative to free-response questions. Teach
15. Brian A. Couch, Joanna K. Hubbard, Chad E Brassil. Learn Med 1993; 5:107-115.
Multiple–True–False Questions Reveal the Limits of 27. Butler A. For better multiple-choice tests, avoid
the Multiple–Choice Format for Detecting Students tricky questions, study finds. EurekAlertAAAS 2018.
with Incomplete Understandings. BioScience 2018; Retrieved on 18/2/2018 from: https://www.eurekalert.
68(6): 455-462. org/pub_releases/2018-10/wuis-fbm100218.php.
16. Bender DA. MCQ, EMSQ or multiple true/false 28. Javaeed A. Assessment of Higher Ordered Thinking
questions? Bioscience Education 2003; 2(1):1-3. in Medical Education: Multiple Choice Questions
DOI: 10.3108/beej.2003.02000009. and Modified Essay Questions. University of Ottawa.
17. Salam A. Input, process, output: system approach MedEd Publish -An innovative approach to publishing
in education to assure the quality and excellence in medical and health professions education, 2018.
in performance. Bangladesh Journal of Medical DOI: https://doi.org/10.15694/mep.2018.0000128.1
Sciences 2015; 14(1): 1-2. 29. Touchie C. Medical Council of Canada Guidelines
18. Salam A. Issues of objective, content, method and for the Development of Multiple-Choice Questions,
assessment in the development of relevant curriculum 2010.
in medical school. Malaysian Medical Association 30. Brame C. Writing good multiple-choice test
(MMA) News; April 2010; 40(4): 22-24. questions, 2013. Centre for teaching. Vanderbilt
19. Bloom, BS, Engelhart, MD, & Committee of College University, 2013.
and University Examiners. Taxonomy of educational 31. Retrieved on 27 November 2019.from https://cft.
objectives: The classification of educational goals. vanderbilt.edu/guides-sub-pages/writing-good-
London: Longmans, 1956. multiple-choice-test-questions/.
20. Anderson LW & Krathwohl DR.A taxonomy for 32. Salam A, Yousuf R, Mahmood CB. Assessment in
learning, teaching, and assessing, Abridged Edition. medical education. JCMCTA 2004; 15(1-2): 15-18.
Boston, MA: Allyn and Bacon, 2001. 33. Haque M, Yousuf R, Abu Bakar SM, Salam A.
21. Jamaludin R, Jaafar R, Kaur S. Training module Assessment in medical education: Bangladesh
series: student-centered learning (SCL), Approaches perspectives. Bangladesh Journal of Medical
for innovative teaching. Module 3: Learning Sciences 2014; 12(4): 357-363.
Taxonomies. Centre for Development of Academic 34. Bothell TW. Assessment of student learning. Brigham
Excellence (CDAE).UniversitiSains Malaysia, 2012. Young University Faculty Center. Annual University
22. Sunita Y Patil, Nayana K Hashilkar, Bhagyashri R Conference, 2001.

88

View publication stats