A sensational sting, ‘Operation Prime Minister’, undertaken by News Express – not a very popular news channel, exposes malpractices of 11 opinion poll agencies. It showed how the pollsters were willing to manipulate data and provide “misleading results”. News Express editor-in-chief Vinod Kapri, claimed in a press conference, that they wished to “expose the mindset and intent” of the polling agencies. The channel, according to the editor was motivated by the Election Commission’s letter to political parties inviting their views on opinion polls and the mushrooming of such polls. The operation revealed that these polling agencies are willing to manipulate data to any extent at the behest of the client by way of deleting negative data or simply increasing the margin of error to show a spike in seats.
— Lokesh K Agrawal (@AapKaLokesh) February 27, 2014
The undercover journalists met 11 polling agencies as representatives of political parties. Most of these 11 agencies are not well established. AC Nielsen and CSDS-Lokniti refused to entertain the News Express reporters as their hand were full till election season.
In an interesting clip, Company C-Voter’s pollster Yashwant Deshmukh is seen telling the News Express’s undercover reporter that while 3% was the standard margin of error, “at best, we can put it to 5%”; “Rafu kar sakte hain, paiband nahii lagaa sakte hain.” We can darn, but not put a patch. C-Voter was also working for ‘India Today’ and the India Today Group has suspended all opinions carried out by C-Voter.
What are they?
It will take time to settle down the dust kicked up by the recent exposure. Meanwhile, let us understand here what the Opinion Poll business is all about? A scientific nonbiased public opinion poll is an inquiry designed to measure the public views regarding a topic of public interest where trained professional interviewers ask questions to people who are chosen at random from population being measured. A random sample is supposed to represent every segment of population being studied. Representative samples are chosen in order to make generalization about a particular population being studied. Opinion Polls tell us what proportion of a population has a specific viewpoint. This is a specialized work of social scientists and scholars. Statistical learning methods are now adopted in modelling and predicting voting intentions. Originally, the opinion polls were conducted to present the misinterpreted interests of large institutions to public. These polls would also gauge the ignored interests of the public.
There are polls which are taken before a candidate announces his bid for office. It could be soon after the announcement. It is technically called ‘Benchmark Polls.’ It gives a general idea where the candidate stands with the electorate before he starts his compaign. It also gives an idea of what messages or slogans could be appealing with the electorate.
‘Bushfire Polls’ are undertaken as the nest step which gives the idea how competitive the race is how much money the compaign needs to spend. It focuses on voters and efficacy of the messages delivered. The candidate or party gets a fair idea if any progress is made on ballet and how much demographic progress or regress has been made so far. Effects of ‘attack messages’ are analyzed here. The polls which are repeated regularly during the compaign are termed as ‘Tracking Polls.’ It could be a weekly exercise where the data from the past are discarded and updated.
Polls are based on samples of populations are subject to sampling error that reflects the effects of chance and uncertainty in the sampling process. The uncertainty is called Margin of Error.
A sample size of around 500–1,000 is a typical compromise for political polls. Another interesting calculation relies on ‘Poll Averages.’ This makes the assumption that the procedure is similar enough between many different polls and uses the sample size of each poll to create a polling average.
— News Express (@newsexpresslive) February 25, 2014
Coming back to latest sting, the representatives of QRS, another pollster company, allegedly told the undercover reporters that they first did a survey projecting the BJP as winning 200 seats in UP assembly elections and then created another fictitious company to project SP getting 200 seats.
By Naim Naqvi
Image Source: Opinion Polls@Facebook