A data analyst has suggested that the CBSE may not have kept its promise of discontinuing the practice of inflating Class XII results. Prashant Bhattacharji, 33, an electrical engineer from IIT Kharagpur and working in Hyderabad, has plotted the results of about 25,000 students from over 300 schools in Gujarat and Rajasthan to indicate distortion.
In 2013, Bhattacharji had carried out a similar, widely reported analysis suggesting misuse of “moderation” by the CBSE and CISCE to inflate their respective students’ marks. He had mined results of 7 lakh Class XII CBSE students and plotted a distribution of marks that showed an abnormally large number of students scoring 95 per cent.
On Monday, a day after CBSE declared its Class XII results, Bhattacharji repeated the exercise with a smaller sample — 25,000 students of 139 schools in Gujarat and 170 in Rajasthan. He says the graphs (see chart below) for Physics, Chemistry, Mathematics and Economics show an abnormally large number of students scoring 95 per cent.
In the CBSE results declared Sunday, the number of students scoring 95 per cent or more increased by 740 since last year, while those scoring 90 to 95 per cent decreased from 63,387 to 63,247. Despite the drop in the latter bracket, Bhattacharji claims the CBSE may have still have resorted to what he calls “a mix of distortion and inflation”.
“We expect something roughly resembling a bell-shaped curve, even a skewed one will do. What we see is a distribution badly distorted into a near-uniform distribution with jagged edges and massive spikes at specific points. There is a mix of distortion and inflation,” he wrote to The Indian Express. “Last year, CBSE, by its own admission, had awarded as many as 16 extra marks to students in Mathematics. The fact that the number of students scoring 90 to 95 per cent has dropped could also mean that the Board might have awarded five to seven marks extra this year instead of 16. This way, the number of 90 to 95 percenters will show a decrease, but that doesn’t mean marks were not inflated,” he added.
CBSE did not respond to questions emailed by The Indian Express Monday. Chairman R K Chaturvedi did not respond to calls. In his sample, Bhattacharji found 1,142 out of 11,141 students in Mathematics, 429 out of 10,343 in Economics, 941 out of 13,973 in Physics and 1,231 out of 13,966 in Chemistry as having secured 95 per cent.
The English results, he says, point to a more fair method of marking. “It has the characteristic bell-like structure, though somewhat skewed,” he wrote in his email.
Asked about restricting his sample to 25,000 students, Bhattacharji wrote, “A sample of 25,000 is a sufficiently large sample, though this does have regional bias in it and is restricted to one zone.” But why regional bias? “Data was analysed for over 300 schools, from Rajasthan and Gujarat, because I had information about the school codes and centres (required to view results) only for those — based on what was available on the web. Previously, one could view results for all students without having to know any specific codes for schools,” he said.
What if the sample of 25,000 just happens to be of students who have scored very well in Class XII exams this year? “The (graph) patterns are quite consistent with the flaws of past years which’d make it too big a coincidence for the flaws to be attributed to the characteristics of the specific cases which I downloaded,” he wrote.