Dear Candidates,

We have received emails regarding normalization of SNAP scores. 

We wish to inform you that we followed a rigorous process that was statistically tested, to ensure that across the question papers, unique types of questions were framed under a specific topic. Normal spread was ensured to avoid unfair advantage to candidates who opt for any specific test / multiple tests. This ensured normalisation of the spread of questions across sections. Furthermore, the question papers were normalised for level of difficulty across all three sets. An independent Committee evaluated the same. 

POST TEST:
The scores secured by candidates were analysed independently for each of the SNAP tests. The tests for normalcy along with skewed distribution were checked. It was found that the scores were normally distributed and were not skewed for any of the three tests. Hence, it was concluded that the scores of all the three tests were individually normal in their distribution.

All candidates were given an opportunity to appear for upto two SNAP tests. Approximately 50% candidates used this opportunity. While the candidates applied for SNAP, it was very clearly mentioned that the maximum of the two scores will be considered for those who have appeared twice. We were confident in mentioning this as we had taken all precautionary measures towards normalization of question selection and their difficulty. We drew the final scores for all candidates by applying the "best score rule" and found that there is no statistical difference in the performance of candidates who took the test multiple times, between the tests.

This analysis was again presented to an expert committee for their views. The committee reviewed and approved the process. The committee opined that there was no need for score normalization. Thus the University has decided that there is no need for normalization of marks (scores).

Dear Candidates,

We have received emails regarding normalization of SNAP scores. 

We wish to inform you that we followed a rigorous process that was statistically tested, to ensure that across the question papers, unique types of questions were framed under a specific topic. Normal spread was ensured to avoid unfair advantage to candidates who opt for any specific test / multiple tests. This ensured normalisation of the spread of questions across sections. Furthermore, the question papers were normalised for level of difficulty across all three sets. An independent Committee evaluated the same. 

POST TEST:
The scores secured by candidates were analysed independently for each of the SNAP tests. The tests for normalcy along with skewed distribution were checked. It was found that the scores were normally distributed and were not skewed for any of the three tests. Hence, it was concluded that the scores of all the three tests were individually normal in their distribution.

All candidates were given an opportunity to appear for upto two SNAP tests. Approximately 50% candidates used this opportunity. While the candidates applied for SNAP, it was very clearly mentioned that the maximum of the two scores will be considered for those who have appeared twice. We were confident in mentioning this as we had taken all precautionary measures towards normalization of question selection and their difficulty. We drew the final scores for all candidates by applying the "best score rule" and found that there is no statistical difference in the performance of candidates who took the test multiple times, between the tests.

This analysis was again presented to an expert committee for their views. The committee reviewed and approved the process. The committee opined that there was no need for score normalization. Thus the University has decided that there is no need for normalization of marks (scores).