The track record of exit pollsters in predicting Lok Sabha elections was remarkable until 1999 in India. My own projections for successive Lok Sabha elections in 1996, 1998 and 1999 for the Times of India were spot on. (Refer Table 1) Similarly, my projections for the assembly elections to Uttar Pradesh (2007), Gujarat (2007), Madhya Pradesh (2008) published in the MINT newspaper were also spot on.
It is therefore natural to ask why Indian pollsters in general (and that includes my own predictions) and media pundits in particular have failed to read the voter mood correctly in the 2009 and 2004 Lok Sabha elections when they were able to do so very accurately until 1999? (Refer Tables 2 and 3)
Is there something esoteric and mysterious in the fact that the only two parliamentary elections in India’s parliamentary history, where the pollsters in general have gone horribly wrong, were totally ‘electronic’ elections in which electronic voting machines (EVMs) were used all over the country? Are the electronic voting machines responsible for the jerky election outcomes in Indian elections? Do EVMs constitute the enigmatic ‘X factor’ in Indian elections?
TABLE 1: GVLN Rao’s EXIT POLL PREDICTIONS vs. ACTUAL RESULTS (1996, 1998, 1999)
- Poll prediction by G.V.L. Narasimha Rao for Times of India/ Doordarshan; Source: Indian Elections in Nineties by GVL Narasimha Rao and K. Balakrishnan (Published, 1999)
|TABLE 2:EXIT POLL PROJECTIONS: LOK SABHA ELECTION, 2009|
* Political Assessment
|TABLE 3:EXIT POLL PROJECTIONS: LOK SABHA ELECTION, 2004|
In any country, these doubts should warrant a serious examination and study of election results. But our EVMs do not offer any such facility as we have no verifiable, credible proof of voting in elections. The Election Commission of India says that the electronic record of the votes cast is stored in the memory. If somebody were to tamper the EVM, even the electronic data can be manipulated. It is due to this reason we have been demanding physical record of voting.
Do make an assessment for yourself if the results of elections warrant a closer scrutiny and examination. Lest I am not accused of party bias, let me make it clear that I am raising the issue in a non-partisan manner. I am questioning the efficacy of the technology per se and the possibility of its misuse by unscrupulous individuals, irrespective of any political affiliation.
To explore the reliability and accuracy of EVMs, the only way left was to validate the reported election results with the personal verbal accounts of the voters. For this purpose, we set out for an empirical study in about 50 polling stations where a single candidate had secured almost all the votes or a huge majority of them. This study was undertaken within two months after the Lok Sabha election results were declared in May 16., 2009.
Results of our investigation were quite revealing and shocking. I will post the details of this exercise in my future posts.
I can be contacted at nrao@indianEVM.com