Games psephologists play

Pic for representational purpose

Raju Korti
Now that a watershed election is over and just about everyone is going  to town with his "analysis" of the poll results, the question is should you believe every analysis reported by the Media?
Poll statistics are a dicey and misleading affair, or at best, a double-edged weapon. I rationalize this on my own definition of statistics. To me it is the specious arithmetic of producing unreliable facts from reliable figures. In layman's terms, statistics can establish non-truth with a logic that can be weird and bizarre.
As someone who is perpetually predisposed to diehard skepticism, I tend not to believe a lot of what I see until I can verify it for myself. Some people have agendas that they’d like to steer you and me towards, and I generally try to ensure that that doesn’t happen to me blindly. The other premise is a bit less conspiratorial in nature: Even though most polls are conducted properly, the results are often reported improperly. That usually has to do with not understanding the statistical nature of the poll on the part of the reporter. But, since you’ve learned how to calculate mean values and can answer the question what are the range and standard deviation, you now know everything you need to decipher poll results and to decide for yourself whether or not you believe them.
For most people glued to their seats trying to get a hold on self-proclaimed psephologists spouting a maze of numbers they would not otherwise hassle themselves with, it is a readymade post mortem minus the clinical precision. My reservation to poll statistics, leaving aside their academic interest, is it can unreasonably support or undercut any argument. It is one thing for the politicians to buttress their viewpoints with statistics and quite another for the Media and its psephologists to dissect those figures with a semblance of coherence and conventional wisdom. Statistics can be a substitute for mathematical inference but not judgment. I remember in one of the Test series tail-ender BS Chandrashekhar averaged better than Gavaskar since he remained not out on zero in all the innings he played.
My friend and fellow journalist Mayank Chhaya, however made a pertinent observation with a piece of statistic that paints the larger picture vividly. Said he a day before the polling: "We still do not know the overall voting percentage but let’s presume for the sake of argument that 60 percent of the 814 million eligible voters exercised their right. That makes it close to 490 million Indians voting in the election. This means that the destiny of 1.26 billion people was decided by 770 million out of them either not choosing to take part or not being eligible to take part because they are below the voting age. I know democracy is the best, albeit flawed, system that we have but the idea that nearly 65 percent of the country’s actual population has to live with the consequences of electoral preferences of the remaining 35% is rather disturbing."
According to the statistics issued by the Election Commission, an estimated -- and that's quite a staggering figure -- 50 lakh voters chose to press the NOTA (None Of The Above) button and play villain for many a contestant. Just how would it have affected the poll results? The answer is bread and butter for psephologists and politicians and play around with that figure.
Overall vote base is another statistic that is often thrown at people into believing the other side of the story. So you have a candidate who may have been returned with a thumping margin but the party may have suffered erosion in its vote base. You don't find anything absurd about it because according to statistical inference a man kept in scorching 500 degrees of heat for five hours and then kept in -500 degrees of cold for the same time has to be perfectly normal.
More often than not, stories based on poll statistics aren't even actually news. Not all statistics can  establish a sound correlation between facts and figures and all have a buffer within which their numbers should be viewed. For a Media whose credibility has chaffed considerably in the last decade or so, pliable statistics are far easier to seek than stubborn and inaccessible facts.
Pity the people! They have to put up with lies, damn lies and statistics.