Of research and reality
Recently, we read in ‘Herald’ about how Goa was the Indian state which had the least use of tobacco. We hope that the Goa-based National Organisation for Tobacco Eradication, better known as NOTE, has duly taken note. It can afford to give itself a quiet pat on the back and take things a little easier. However, all research is not always as reliable. This could be because it uses the wrong sampling methodology, as a result of which those it surveys are not representative of the population as a whole. Or, it could be because it structures the questionnaire wrongly, the respondents do not understand, and give inconsistent answers. But, most often, it is because the people hired to do the survey are not supervised properly, there is no system for verification of response sheets, and the majority of responses are fudged by the surveyors.
One example is based on data from the central government; the Union Ministry of Women and Child Development (WCD) to be precise. Released recently, it says that Goa is one of the two states in India with the least corporal punishment in schools. But Goa doesn’t stand first here; it’s second, with 34.25 per cent of children saying they faced corporal punishment, and 65.75 per cent saying they didn’t. The state that apparently ranks above Goa is, surprise, surprise, Rajasthan; with just 17.87 per cent saying ‘Yes’ and 82.13 saying ‘No’.
But this should not really come as a surprise. Data from the same survey led the WCD Ministry to create an uproar when it announced – on the basis of a survey it claimed was conducted among nearly 12,500 children, boys and girls, aged between five and 18 from 13 states. More than half of Indian children, it said, had faced sexual abuse! Do one in every two Indian children face sexual abuse? It sounds absurd. But that did not stop the Ministry from going ahead and putting it down in black and white…
The larger the scope of the research, the more stringent, rigorous and meticulous it needs to be in terms of methodology, and the more carefully it must supervise the surveyors and backcheck to see that responses are not fudged. Even the best in the world can slip up. The latest controversy doing the rounds is a study on deaths from Malaria in India, published in ‘Lancet’, which is one of the world’s most respected medical journals. It estimated that Malaria causes around 205,000 deaths per year in India; 13 times the figure estimated by the World Health Organisation (WHO), which has rejected these figures as “implausibly high”. The problem, it says, is the methodology – ‘verbal autopsy’ – which is based on a description of symptoms by a family member. It is a reliable method for diseases with distinct symptoms, but not those with mixed symptoms like Falciparum Malaria. Earlier, two of the study’s authors derived an estimate for HIV-related deaths in India using the same methodology, and got an estimate that was much lower than the official WHO figure.
Yes, even ‘Lancet’ can get it wrong, though it doesn’t happen very often. WHO’s indictiment of its Malaria paper ought to give a big boost to the Union Ministry of Health, which had lashed out at another paper recently published in ‘Lancet’ about the alleged ‘super-bug’ it christened ‘New Delhi’, which it said was resistant to most known antibiotics and was being spread through operation theatres in elite Indian hospitals that were catering to Westerners opting for cheaper surgery in India. While the authors defended their work as being based on ‘evidence’, the Indian government interpreted it as a conspiracy to scuttle the country’s fast-growing ‘medical tourism’ industry. Even scientific research, it seems, is not unimpeachable. It must be seen in context of reality.
