With all the sessions of NEET MDS being over, candidates are apprehensive as to how their performance will be affected since some sessions of NEET MDS were quite easy and some were on the difficult side.
Candidates are consistently sending E-mails quering whether they will be at a disadvantage owing to disparity in the difficulty level of the different sessions of NEET MDS.
With a view to put your confusion to rest, PrepLadder is the first platform to share the details as how the Normalization process used by the NBE works.
Normalization Procedure – Details
Psychometry procedure applied to standardize the score.
In order to arrive at the normalized score, 3 step-procedure is followed:
Step I: Calculation of Raw Marks
Raw marks are calculated based on the manner in which you have attempted the exam.
- For every Correct Answer, the candidate is awarded +1 mark
- For every Incorrect Answer/Not Attempted, the candidate is awarded 0 mark
The important aspect which matters is the number of points you obtain at this stage.
Remember, the raw marks do not constitute your final score.
Also Read: Colleges to Choose for MDS
Step II: Equating Raw Marks
In order to equate raw marks, a procedure known as ‘Item Response Theory’ is applied.
It refers to a psychometrically supported statistical model utilized in the process of equating marks.
It is based on the principle that ‘Candidates response to a particular MCQ determines the weight age of the MCQ’.
Item Response Theory takes into consideration the following 2 aspects:
- Performance of the candidate
- Difficulty level of the MCQs
1. Performance of the Candidate
We will explain this by taking an instance:
Suppose 100 candidates have attempted an MCQ.
Out of 100, almost 90-95 candidates answered it correctly.
This will infer that the question was so easy that majority of candidates answered it correctly.
The general principle behind this is that questions answered correctly by majority of candidates are marked as ‘Easy’ by the software and are assigned ‘Low weight age’ by the software.
For instance, an Easy question might be assigned weight age as low as 0.1 mark.
On the other hand, questions which are answered ‘Incorrectly’ by majority of candidates are regarded as ‘Difficult’ and are assigned ‘High Weight Age’ by the software.
For instance, a difficult MCQ is answered correctly by only 5-10% of candidates.
Such questions will be considered as ‘Difficult’ by the software and the software might assign 4-5 marks to such questions.
Also Read: Expert analysis of NEET MDS December-2016
2. Difficulty level of the MCQs
Difficulty level of the MCQs is determined using 2 factors:
- Difficulty Index
- Discrimination Index
Difficulty Index is computed using the formula: H+L/T * 100
Discrimination Index is computed using the formula: H-L/T * 2
H denotes: Number of students with Correct Answer in ‘High Achievers’
L denotes: Number of students with Correct Answer in ‘Low Achievers’
T denotes: Total number of candidates who attempted that question
After checking all the papers of a particular session, the papers are arranged in descending order of marks and are segregated into 3 groups:
These 3 Groups are:
1-Upper rd referred to as the ‘High Achievers’ Group
2-Lower rd referred to as the ‘Low Achievers’ Group
3-Middle rd – This group is not included in Analysis. This group comprises those candidates who answer some questions correctly and some wrong and use Blind Guesswork for attempting the exam. Therefore, these are excluded from Analysis
Also Read: Government Job Options after BDS
Step III: Equated Raw Score is Scaled/Normalized
The final step involves normalizing the equated raw score. For this Linear Transformation procedure is applied.
Linear Transformation includes Normal Distribution and Standard Normal Distribution.
Based on this, Z score is computed.
Z-score = Observed Value-Mean Value/Standard Deviation
Let us now compare how Z-score varies in an Easy and Difficult session:
For Easy Session
Suppose 100 candidates appeared. Mean score for an easy session will be high.
Suppose the Mean Score is 75 and Standard Deviation is 15
Then, suppose a candidate has obtained 90 marks as his raw score.
Then, Z-score = 90-75/15 = 1
We see that Z score turns out to be 1 for an easy session.
For Difficult Session
Suppose the candidate has obtained 75 marks as the Raw score and suppose that the standard deviation is 20.
Assuming a difficult session, let us say that the Mean score is 50
Then, Z-score = 75-50/20 = 1.25
As evident, Z score turns out to be 1.25 for a difficult session.
As it is clearly evident, Z score is higher for a Difficult session.
Finally, Normalized or Corrected Score is computed using the following formula:
Target Mean + (Z score * Target Standard Deviation)
Now, it is noteworthy that Target Mean & Target Standard Deviation is same for all students and for all sessions.
Also Read: Pioneer Dental School
Target Mean & Target Standard Deviation is computed using either of the 2 methods:
1. By computing the Mean score for each session separately and then computing the Mean & Standard Deviation from the Mean Values of all sessions
2. For every session, marks of Top 1% or Top 10% candidates are used to compute the Mean & Standard Deviation
Therefore, it is clearly evident that the sole factor which will determine your overall ranking and percentile will be the ‘Z Score’
Therefore, you must not be worried in case your session was Difficult.
Normalization procedure used by the NBE standardizes the score of all candidates to arrive at the final merit list.
Thus, Z-score will be low for easy sessions and high for difficult session, but overall the disparity in different sessions will not effect your overall rankings.
We are sure that this post will definitely provide you an in-depth understanding of the Normalization procedure used by the NBE.
Stay Tuned for More Updates!!