We have prepared this article to help Medical PG aspirants
understand ‘Normalization’ and what can be done to counter stress arising from
it. Many people offer explanation about this but it is way too technical and
complicated for majority of students, so we have tried to make it as easy as
possible to understand.
Just last year paper-pencil test was the method for
conducting exams like AIPMEE and AIIMS, when the same or a similar test papers
(with reordering of the questions) were given to all test takers and it was
relatively easier to compare student’s performance as they only had to create a
merit list of scores of students. With the introduction of an online exam by
MCI which is conducted by NBE (as of now), the test and the questions vary
across different test days and slots. Hence normalization is needed – scaling
up/down of scores using psychometrics, statistics and test takers’ responses to
a standard set of questions. This helps evaluate students on a common standard
and is necessary because the difficulty level of the test varies across test
days.
Why is normalization needed?
Let us consider a simple hypothetical scenario: Assume one
student is attempting NEET this year along with a friend. Both of us are
equally good in medicine. But while my strength lies in Anatomy, his strength
is Biochemistry. Since we gave the exams on different days and slots our test
papers were different. We both got considerable questions from our respective
strong areas and performed equally well. But does that mean that we are equally
good? No. Say on a scale of 1-5, I got questions of difficulty level 4 from
Anatomy. Being my strong area I did well. On the other hand he got questions of
difficulty level 5 from Biochemistry. Now the only way to benchmark his and my
performance is by looking at the relative scores of people who gave the test
with us. In my slot out of 80/100 students were able to work out the questions
from Anatomy. On the other hand, in my friend’s slot only 50/100 students were
able to work out questions from Biochemistry. That means when we take a large
enough sample size (students), my friend’s performance is much better than
average as compared to mine (even when we are scoring equally well).
Personal choices that bring in subjectivity – a student
might not like anatomy at all but might be strong in Physiology or
Pharmacology. This is countered by taking a large sample size.
Also there is something called ‘an equating block’. Now
there are questions in papers which are same in many papers which are called
‘an equating block’. Now suppose there are 24 questions out of 240 in a paper
which are same in papers of many days and slots. If I answer all 24 of them
correctly, when most of the students have answered only say 20 on average then
psychometric analysis will consider my high score as superior compared to other
students. So for example I have answered 200/240 questions correctly my raw
score will be same as all the students who scored 200 (as there is +1 for right
answer and +0 for wrong) with same number of wrong answers. But my scaled score
will be higher.
Also let us consider ‘no negative marking’ scenario. If
anyone attempted 240 and he/she got 200 right and there is a student who
attempted 230 but he/she also got 200 right then his percentile score will be
put above mine as NBE has said in their TIE – BREAKER CRITERIA that ‘In the
event of two or more candidates obtaining same percentile, the merit position
shall be determined by the number of wrong responses of such candidates.
Candidate with less number of wrong responses shall be placed at higher merit.’
Also as another example if a candidate scores 200/240 in a
paper in which all students performed poorly in equating block of questions
(hence a difficult paper) will be placed higher than candidate scoring 200/240
in a paper in which all candidates have performed well in equating block of questions
(hence an easy paper).
So conclusion is that a student who answers more number of
questions correctly, scores higher in equating block of questions and also has
less number of wrong answer will be put above all candidates.
Now lot of candidates where complaining after the results
that their score was much lower than what it should have been due to
normalization. It has also led to widespread debate about whether the technique
is even reliable.
How to deal with normalization then? Since normalization or
some other form of standardization is here to stay as even AIIMS is going to
conduct online exams, it is important that we make peace with the idea and try
to deal with it. My advice is to ignore the entire concept of normalization. And
we have good reason for the same: one cannot really predict how others in your
same slot are going to perform in the same test (over 3000 students had taken
NEET in a single slot). You cannot even predict whether the question you have
attempted/left are going to be branded as easy or difficult based on statistics
– in short, there is no way to know whether a question is that ‘equating block’
question that one needs to attempt and get right at all costs. Therefore it
makes sense to attempt as many questions as possible just the same way as one
would have done in a normal exam. Yes there might be some loophole with the
laws of statistics, but then which law in this cosmos is without its own set of
cracks and glitches.
Do not link your mock test scores to NEET scores and blame
normalization – dip in scores can be due to exam stress too. There is no point
in shooting arrows in the dark and creating unnecessary anxiety unless you want
to pin the blame of your performance on normalization. Give it your best shot
and forget the rest.
The other point to note is that the sample size taken for
standardization in this case is very large (over 0.9 Lac students giving NEET)
– thus low chances of statistical selection error.
Also these king of normalization is done is most
international level competitive exams like GMAT and MBA exams like CAT. MBA
aspirants have been complaining about it since years but it hasn’t helped
scraping the exam.
So just work hard and forget about the rest.
