Now that the curtain is closing on the world’s mammoth parliamentary elections, a significant issue that has taken centrestage in the last few days is the prediction of the exit polls. Most major exit polls conducted by local media post the last phase predict a clear majority for the BJP-led NDA and according to some, the party will surpass its 2014 record. For instance the India Today-My Axis and Chanakya exit polls have predicted a landslide win for BJP-led alliance and its allies with over 350 seats along with marked gains in places such as West Bengal, Bihar, Odisha among others.
What is marked about the exit polls is that while it scales up the political brouhaha amidst the parties, the results of such polls have historically been considerably mixed. In the international sphere, the Brexit phenomena is a more conscious example of the stark contrast between what the pollsters favoured and the actual results. To substantiate the aforementioned thesis, a closer examination of the past exit polls is necessary to understand the current prediction.
For a start, the pollsters from exit polls during the 2004 general elections gave the Atal Bihari Vajpayee-led NDA government 230-275 seats. But the result proved to be a misnomer, as the party won only 187 seats and was pushed out of power. Further, during the 2009 general elections, exit polls predicted a hung parliament with the UPA garnering around 199 seats. Instead, the UPA accumulated over 262 seats and formed a stable government of its own. However, on some occasions, like the 1999 general election and the recent 2014 elections, the exit poll predictions have been reasonably accurate.
A crucial feature that factors into making the state of exit polls a shaky affair is that there persistsconsiderable doubt regarding the technique and the sample sizes of many polling organizations. Although of late, there has been an increase in the sample size to make it representative of the population, the core issue lies in choosing a sample that gives a near accurate representation of the voter profile.
Another formidable challenge lies in estimating the vote share, given that India is a country of diversities; in terms of location, caste, religion, language, different levels of educational attainment and economic class all of which have a bearing on the voting behavior. So essentially, over- or under-representation of any particular societal group can affect the accuracy of the estimated vote share.
In addition, respondents often lie to pollsters, sometimes due to a phenomenon known as the “social desirability bias.” This entails that when particular preferences become socially undesirable, certain individuals have a tendency to suppress their choice in lieu of making the “politically correct choice.” They may even be scared to admit their true choice due to local political conditions, preferring to ride the bandwagon. (The US 2016 election is a case in point).
Thus, while the numbers from the various exit polls for 2019 elections are out, it is only later today that one will find out how accurate these predictions will be, once the results of the election will be announced.
The image used is for representative purposes only.
India Outbound
May 23, 2019
source https://indiaoutbound.org/2019-exit-polls-a-word-of-caution/
No comments:
Post a Comment