Introduction to the Blog - INDIA543

Introduction

This is not another opinion poll but an aggregation and correction of many that are being published and made public every week and shown on news channels as well as printed in news papers and magazines. The results and analysis here rely on the survey data as well as analysis conducted by these pollsters. These results here are formed basis a detailed analysis of past opinion polls performance vis-a-vis the final election results. Each of these pollsters are then ranked (individually for each state) and then given weights, basis which the final results are derived. There is also an attempt to incorporate some corrections basis expected alliances or possible announcements by various political formations as it keeps coming out in the public domain. Our results may or may not prove to be correct in the end but as of now the attempt is to be the best at least statistically.

Why is our poll expected to be better than those published?

An opinion poll, is a survey of public opinion from a particular sample. Opinion polls are usually designed to represent the opinions of a population by conducting a series of questions and then extrapolating generalities in ratio or within confidence intervals. Its proven and is believed that if the sampling is truly random and scientific then the analysis would be able to predict the results accurately, with a margin of error of +/- 5%. What is then the reason that pollsters, in spite of running surveys for past 10-15 years, in the age of 24*7 television and social media, haven’t been able to guage the public mood correctly? Fortunately whats again well known and proven is that the reasons for this are also statistical and are very well known. Of these the main ones are:

1. Non Response Bias: Some people do not want to or cannot answer these complicated interviews while others are very savvy. This more often than not is a result of their socioeconomic backgrounds or their current profession. E.g. a software techie may be keen and enthusiastic to put his point of view across and a beggar may feel shy or under confident. But in universal franchise their voting weight ages are same. That may primarily be the reason why BSPs tally is always considered to be under projected.

2. Response Bias (A): Survey Results may be deliberately engineered by unscrupulous pollsters in order to generate a certain result or please their clients and also respondents may deliberately try to manipulate the outcome of a poll by e.g. advocating a more extreme position than they actually hold in order to boost their side of the argument or give rapid and ill-considered answers in order to hasten the end of their questioning.

3. Response Bias (B): Respondents may also feel under social pressure not to give an unpopular answer. For example, respondents might be unwilling to admit to unpopular attitudes caste and gender bias, and thus polls might not reflect the true incidence of these attitudes in the population. 

4. The Shy Tory Factor: The Conservatives in UK had suffered a sustained period of unpopularity as a result of economic difficulties and a series of minor scandals, leading to a spiral of silence in which some Conservative supporters were reluctant to disclose their sincere intentions to pollsters. This e.g. impacts the Communist opposition in Bengal for their perceived unpopularity in popular media.

So for the purpose of accurate forecasting, we would correct the current set of survey’s weighted average basis the proven errors and thence suggest our final projection. We would go about this state by state for all states and union territories and finally combine these to get a national picture too. In the process we would also try and understand some rapid upward or downward trends (like rise of AAP) and predict possible events (new forming 3^rd front 4^th front) and further improve our numbers. We would also take feedback of our results from the next opinion polls being published and continuously update ours till elections are held or election commissions bans us! 

However finally would like to say that I am doing it for the passion and fun of it. The opinions expressed here would be completely mine and would want those to be unbiased – only as much as possible!