GS Paper II
News Excerpt:
On June 4, many people slammed the exit poll predictions after counting began for the 2024 India Lok Sabha elections.
More about the news:
- Exit polls in the recent elections grossly overestimated the number of seats the National Democratic Alliance (NDA) would win, predicting 350-400 seats, while they won only 293 seats.
- This is not the first time exit polls have gotten it wrong. In 2014 and 2016, they underestimated NDA's numbers.
- In 2014, exit polls underestimated the NDA's strength and overestimated the strength of the United Progressive Alliance (UPA), even though they predicted the NDA's victory.
- In 2019, most exit polls again got the numbers wrong, except for two of them.
- Experts have raised questions about the inaccuracy of exit polls, suggesting that either the voters lied to the pollsters or the polls employed flawed methodologies.
What are exit polls?
- Exit polls are surveys conducted immediately after voters leave the polling stations. Pollsters use probability and statistics to forecast election results based on the exit poll data.
- Exit polls can provide insights into voter behavior and early projections of election outcomes.
- Exit polls can analyze how different socioeconomic groups voted by providing demographic breakdowns.
- In India, such polls are prohibited during the voting process by the Election Commission. However, exit polls can be made public 30 minutes after the entire voting process ends.
- This is to make sure that the polls don't influence people while they are still voting, but also still allows for immediate analysis and discussion once the process is complete.
Factors Contributing to the Rise of Exit Polls in India:
- The unpredictability of Indian elections increased in the 1980s, along with the rise of regional politics.
- Despite Congress's historic win in 1984, their mandate was reversed by 1989. This unpredictability made exit polls more popular, as the Indian voter became a mystery to the political class.
- The proliferation of electronic media in the 1990s further popularized election surveys and exit polls in India.
- Exit polls are primarily for public consumption and media interest, with minimal direct impact on the election process itself.
How is it calculated?
Exit polls are calculated using a combination of sampling methods and statistical analysis by various agencies. These include:
- Sampling: Pollsters pick a bunch of voters from different areas to represent the whole population. They want to make sure they talk to people across demography.
- Questioning: Selected voters are asked who they voted for and other basic questions, like their age and where they live.
- Data Collection: The answers are collected and recorded.
- Weightage: The data is adjusted to make sure it is a fair representation. For example, if the pollsters didn't talk to enough young people, they might give their answers more weight to balance things out.
- Analysis: Experts look at all this data and analyse it to make predictions about the election results based on what the voters said.
- Margin of Error: The exit polls also figure out how certain or uncertain their predictions are.
Finally, these predictions are shared with the public.
Do polling errors occur in other nations too?
- The US has seen two most notable failures in the 1948 and 2016 presidential elections.
- The 1948 US presidential election poll failure was attributed to non-random sampling issues.
- In the 2016 US election, a very high non-response rate from voters and resulting bias despite weighting adjustments led to inaccurate polls, especially in rust-belt states.
- UK pollsters struggled to predict the Brexit referendum outcome, with only around a third of polls predicting the "leave" vote correctly.
- Exit polls in the UK have generally been more accurate, potentially due to smaller constituency sizes compared to India, allowing for reasonably accurate predictions with smaller sample sizes.
- The larger geographic scale and constituency sizes in India pose a greater challenge for achieving accurate exit poll predictions with feasible sample sizes.
