A few days back the Congress party wrote to the Election Commission that lotuses in full bloom in various regions of Madhya Pradesh should be hidden, so that voters aren’t drawn towards the Bhartiya Janata Party (BJP). The state goes to the polls later this month. Lotus is the electoral symbol of the BJP.
If one were to extend the argument a little further, other parties could demand that Indians hide their hands when in public, before elections are due to happen. Hand is the electoral symbol of the Congress party.
Also, when elections are due, elephants should be hidden, given that they are the electoral symbol of the Bahujan Samaj Party and Asom Gana Parishad. People should not be allowed to move around in cycles because they are the electoral symbol of Samajwadi Party. Large parts of India where electricity is non existent should not be using the lantern because it is the electoral symbol of the Rashtriya Janata Dal. People should stop looking at their clocks because it is the electoral symbol of the Nationalst Congress Party. And this list can go on.
The Election Commission rejected the unreasonable demand of the Madhya Pradesh unit of the Congress party to hide lotuses. It termed the demand “absurd” and “an insult to the intelligence of the voter”.
After this, the Congress party has made another such absurd demand. It wants opinion polls to be banned. The position of the party is well summarised in a statement made by the party general secretary Digvijay Singh. As he put it “These have become a farce. They should be banned altogether. The kind of complaints, information that I have got, show that anybody can pay and get a survey as desired…Wondering how a few thousand people could predict election trends for a country of 1.2 billion people…It has become a racket. So many groups have sprung up.”
This statement needs to be examined threadbare.
A straightforward reason for the Congress party wanting opinion polls to be banned is the fact that all the opinion polls expect the party to perform badly in the forthcoming state assembly elections, as well as the Lok Sabha elections, due next year. Other parties like BSP and JD(U), which are similarly on a weak wicket, want the opinion polls to be banned as well.
A part of Singh’s statement was that “ The kind of complaints, information that I have got, show that anybody can pay and get a survey as desired.” This is a very serious charge. If Singh has any such information he should bring it to the attention of the nation, instead of just making vague statements. If there is any truth to the accusations, the Election Commission can investigate it and then make a reasonable decision.
Singh further wondered how “how a few thousand people could predict election trends for a country of 1.2 billion people.” Even the most sophisticated opinion polls do not have a sample size more than a few thousand people.
Take the case of opinion polls which happen in the United States before a Presidential election to predict to who is likely to be the next President. The sample size of most such surveys is less than 1500. This can be clearly seen from the following link.
Also, there is a great story from the 1936 US Presidential elections which is very relevant in the context what Singh said. The election had the incumbent President Franklin D Roosevelt, belonging to the Democratic Party, facing Alfred Landon of the Republican Party, who was also the governor of the state of Kansas.
Literary Digest one of the most well respected magazines of that era predicted that Landon would get 57% of the vote whereas Roosevelt would get 43%. Roosevelt got 62% of the vote whereas Landon got only 38%. The survey was way off the mark. Roosevelt got 19% more vote than the survey had predicted. This is believed to the largest ever sampling error in a major opinion poll.
The Literary Digest poll had a sample size of around 2.4 million people. At the same point of time George Gallup predicted a victory for Roosevelt using a much smaller sample size of 50,000 people.
There were problems with the way Literary Digest had chosen its sample. The sample was not a representation of the population. As a University of Pennsylvania Case Study on the issue points out “The first major problem with the poll was in the selection process for the names on the mailing list, which were taken from telephone directories, club membership lists, lists of magazine subscibers, etc. Such a list is guaranteed to be slanted toward middle- and upper-class voters, and by default to exclude lower-income voters. One must remember that in 1936, telephones were much more of a luxury than they are today. Furthermore, at a time when there were still 9 million people unemployed, the names of a significant segment of the population would not show up on lists of club memberships and magazine subscribers.”
Hence, the sample was skewed towards Republicans. In the United States, the upper class typically tends to support the Republican Party, whereas the poor largely go with the Democratic Party. Given this, it predicted a victory for the Republican candidate Landon.
Thus, it is important to ensure that the chosen sample is a good representation of the population as a whole. Small samples can be very effective as long as they are not riddled with sampling errors. “A badly chosen big sample is much worse than a well-chosen small sample,” concludes the University of Pennsylvania cited earlier.
Another article on the website of Constitution Rights Foundation makes a similar point “It is important to point out that large, national polling organizations have small national samples of under 2,000 that predict quite accurately for the entire electorate.” As statistical techniques have evolved samples much smaller than the sample of 50,000 people used by Gallup, to predict the 1936 elections, can give good results. The moral of the story is that bigger is not necessarily better. Something that Digivijay Singh needs to understand.
Opinion polls are also a very important part of the democratic process. People have a right to know what a state, a region or the nation as a whole is thinking on a particular issue, at a given point of time.
But Congress and democracy do not go together. In 1959, the then Congress prime minister Jawahar Lal Nehru dismissed the government of Kerala led by the communist chief minister EMS Namboodiripad, by invoking the controversial article 356. Namboodiripad’s government was the first democratically elected Communist government anywhere in the world.
This set the precedence for Nehru’s daughter Indira Gandhi and she turned dismissing democratically elected non Congress governments into an art form. Estimates suggest that she dismissed nearly 59 state governments during her several tenures as the Prime Minister of the nation.
She, egged by her younger son, Sanjay, declared a state of internal emergency on June 26, 1975. It stayed till March 21, 1977. During this period the leaders of opposition parties were put into jail and the fundamental rights of the citizens of this country remained suspended. The newspapers were heavily censored. Democracy came to a standstill during the period of nearly 19 months the country was in a state of emergency.
The senior most posts in the Congress party have been perpetually reserved for the Gandhi family and there have been no elections for the same. Given this, it is hardly surprising that the leaders of the Congress party are making demands for opinion polls to be suspended. They go against the party’s idea of democracy.
The article originally appeared on www.firstpost.com on November 5, 2013
(Vivek Kaul is a writer. He tweets @kaul_vivek)