Opinion polls and exit polls are surveys conducted to understand voter preferences and electoral behaviour.
They are important sources of:
Opinion Poll → Before voting
Exit Poll → After voting
An opinion poll is a survey conducted before or during the election period to assess:
Survey
↓
Voters/respondents
↓
Questions about:
↓
Statistical analysis
↓
Estimated electoral preference
An exit poll is conducted after voters have actually cast their votes.
Surveyors ask voters:
How did you vote?
or seek information intended to estimate:
How electors voted in the election.
Voter
↓
Casts vote
↓
Leaves polling station
↓
Survey
↓
Reports voting behaviour
↓
Estimated election outcome
The RPA, 1951 defines an exit poll in Section 126A in terms of an opinion survey concerning how electors have voted or how electors have identified with a political party or candidate.
| Feature | Opinion Poll | Exit Poll |
|---|---|---|
| Timing | Before voting | After voting |
| Main question | What are voters likely to do? | How did voters vote? |
| Basis | Voting intention | Reported voting behaviour |
| Purpose | Forecast preferences | Estimate actual result |
| Legal restriction | Mainly subject to election-period restrictions | Specifically regulated under Section 126A |
| Main risk | Influencing undecided voters | Influencing voters in later phases |
| Accuracy | Depends on sampling and timing | Depends on honest responses and sampling |
Opinion poll = intention
Exit poll = reported voting behaviour
Suppose an election is scheduled for next month.
A survey asks:
"If elections were held today, which party would you vote for?"
Results:
This is an:
Opinion poll
because voting has not yet taken place.
Suppose voting takes place today.
Surveyors ask voters as they leave:
"Which party/candidate did you vote for?"
After collecting responses:
This is an:
Exit poll
because respondents have already voted.
Opinion polls can help understand:
They provide information about the political preferences of citizens.
Exit polls seek to estimate:
They attempt to provide an early indication of:
The likely election result
before official counting.
Opinion polls can contribute to democracy by:
They can also influence:
Opinion polls can both inform and influence voters.
Exit polls can:
But premature publication can create:
India may conduct elections in:
Multiple phases
Suppose:
Phase 1
↓
Exit poll published
↓
Phase 2 voters
see the projected result
↓
They may change their voting behaviour.
Information from voters who have already voted could influence voters who have not yet voted.
Therefore, exit polls are subject to a specific statutory restriction.
The:
Representation of the People Act, 1951
contains:
Section 126A — Restriction on publication and dissemination of result of exit polls, etc.
It prohibits conducting an exit poll and publishing/publicising or otherwise disseminating its result during the period notified by the Election Commission.
126A → Exit Poll
For a general election, Section 126A provides that the restricted period may:
Begin from the beginning of the hours fixed for polling on the first day of poll
and continue until:
Half an hour after the closing of poll in all States and Union Territories.
This is different from the:
48-hour silence period under Section 126.
First phase of polling
↓
Section 126A restriction begins
↓
Exit-poll results cannot be conducted/published/publicised during notified period
↓
Later phases continue
↓
Last phase closes
↓
Half an hour after closing
↓
Restriction period ends as notified.
Contravention of Section 126A can result in:
Imprisonment up to two years
or:
Fine
or:
Both
This is expressly provided in Section 126A(3).
126A → Exit polls → Up to 2 years/fine/both
Section 126 deals with:
Prohibition of public meetings during the 48-hour period ending with the hour fixed for conclusion of poll.
It also restricts public display of election matter through specified media during that period.
Protect the final phase of voting from potentially influential election communication.
During the prescribed:
48 hours
before the conclusion of poll in a polling area, certain election-related activities are restricted.
This includes public display of:
Election matter
through television or similar apparatus.
"Election matter" means matter:
Intended or calculated to influence or affect the result of an election.
This is a major UPSC distinction.
| Section | Subject | Period/Scope |
|---|---|---|
| 126 | Silence period/election matter | 48 hours ending with conclusion of poll in the polling area |
| 126A | Exit-poll conduct/publication/dissemination | Period notified by ECI, linked to first-to-last phase |
| 126 | Protects final voting period | Election matter |
| 126A | Prevents premature exit-poll influence | Exit polls |
126 → 48 hours
126A → Exit polls
Opinion polls are not subject to the same blanket statutory regime as exit polls.
However, during the:
48-hour silence period
Section 126 restricts public display of election matter intended or calculated to influence or affect the result.
The ECI has stated that this can cover:
Results of opinion polls or other poll surveys in electronic media during the relevant 48-hour period.
Opinion polls are not the same as exit polls.
An opinion poll generally asks:
"What do you intend to vote for?"
It does not necessarily involve voters who have already voted.
Therefore:
The risk addressed by Section 126A is specifically associated with exit-poll information during multi-phase polling.
Opinion-poll content may nevertheless fall under Section 126's silence-period restrictions when it constitutes election matter intended or calculated to influence the result.
This is the key reason for regulation.
India has:
Phase 1
↓
Exit poll predicts:
Party A likely to win
↓
Phase 2
Voters see the prediction.
↓
Possible:
Exit-poll restrictions protect voters in later phases from early-phase exit-poll influence.
The:
Bandwagon effect
occurs when voters are influenced by the perception that a particular candidate or party is likely to win.
Poll predicts:
Party A → 60%
Some voters may think:
"Everyone is supporting Party A."
↓
They may support Party A.
Poll prediction can potentially become a self-reinforcing political signal.
The opposite can also occur.
A voter may see:
Party B is losing badly.
↓
They may sympathise with Party B.
↓
Support may increase.
This is sometimes called:
Underdog effect
The actual psychological effect is not uniform.
Polls can influence voters who do not simply vote for their first preference.
For example:
Party A
→ Preferred choice
Party B
→ Second choice but more likely to defeat Party C
A voter may use poll information to choose:
A strategically viable candidate/party
Polls can potentially influence:
Strategic voting
Poll results can influence parties' strategies.
Parties may:
Polls influence not only voters but also political campaigns.
Polls generate:
Media may create:
"Horse-race" politics
where the election becomes about:
Who is winning?
rather than:
What policies are being proposed?
Horse-race journalism focuses heavily on:
Policy issues may receive less attention.
Election discourse can shift from substantive issues to electoral arithmetic.
Poll accuracy depends on:
A poll is an estimate, not an actual election result.
A poll normally surveys:
A sample
rather than:
Every voter
Total voters:
10 crore
Survey sample:
20,000
The poll uses the sample to estimate broader voter preferences.
The sample must be:
Representative
Because a poll surveys a sample rather than the entire population:
Sampling error
can occur.
True support:
40%
Poll estimate:
43%
The difference may partly arise from sampling variation.
Poll results should not be interpreted as exact measurements.
Polls can also suffer from:
A large sample does not automatically guarantee an accurate poll.
Some voters may:
Survey responses may not perfectly reflect:
Actual voting behaviour.
A major challenge is:
Who will actually vote?
A poll may measure:
Preference among respondents
but the election result depends on:
Actual voters.
Turnout modelling is critical to election forecasting.
Prediction
Official outcome
The official result is determined through:
Actual votes cast + counting + declaration by competent election authorities.
Poll ≠ Result
Even exit polls are estimates.
They depend on:
Exit poll ≠ official election result.
Exit polls usually rely on voters who have:
Actually cast their votes
but the survey still uses a sample.
Actual vote count remains the final authority.
Shows voter preferences.
Highlights important issues.
Helps study electoral behaviour.
Parties understand voter concerns.
Encourages discussion.
Predictions may influence choices.
Voters may support the perceived winner.
Voters may abandon preferred candidates.
Polls may create an impression that the result is already known.
Campaigns may focus on polling numbers rather than policy.
Exit polls can provide:
These benefits must be balanced against:
Potential influence on voters in subsequent phases.
Especially in multi-phase elections.
Perceived winner may gain support.
Losing candidate may gain sympathy.
Some voters may become less motivated to vote.
Voters may alter preferences.
Suppose an exit poll says:
Candidate A is certain to win.
Some supporters of A may think:
"My vote is unnecessary."
↓
They may not vote.
Opponents may become more motivated.
Exit-poll information could potentially affect:
Turnout and mobilisation.
Poll results may be repeated by:
↓
Repeated information
↓
Perceived public consensus
↓
Further influence
Information cascade
Poll results can spread rapidly through:
A methodological limitation that would be obvious in a full survey report may disappear in a social-media post.
Example:
"Party A leading!"
without mentioning:
Complex survey evidence can become oversimplified political messaging.
Credible polls should ideally disclose:
Transparency allows voters to judge the reliability of the poll.
A poll may be commissioned by:
Sponsor influence
Disclosure of sponsorship helps readers assess possible conflicts of interest.
Professional polling organisations use statistical methods to estimate:
Different agencies can produce different results because of:
One poll should not automatically be treated as definitive.
A:
Poll of polls
combines results from multiple surveys.
May reduce reliance on a single survey.
It does not automatically eliminate:
Aggregation can improve robustness but cannot guarantee accuracy.
Opinion polls involve:
Political expression and dissemination of information.
Restrictions therefore raise:
Freedom of expression + free and fair elections
The distinction is:
Usually asks:
"What might you do?"
Asks:
"What did you do?"
The latter can reveal actual voting behaviour from earlier phases.
↓
Information from earlier voters
↓
Influences later voters
Exit polls create a special timing problem.
Election begins
↓
First phase voting starts
↓
Exit-poll restriction period begins as notified
↓
No prohibited conduct/publication/dissemination of exit-poll results
↓
Last phase voting ends
↓
Half an hour after close
↓
Restriction period ends.
The statutory framework and ECI notification determine the precise period for each election.
48 hours before conclusion of poll in a polling area
↓
Silence period
↓
No prohibited public election matter through specified means
↓
Poll concludes
↓
Silence period ends.
The ECI describes this as the 48-hour period before the conclusion of poll in the relevant polling area.
The law does not simply say:
"Exit polls can never be conducted."
Rather, Section 126A prohibits conducting an exit poll and publishing/publicising/disseminating its result:
during the notified restriction period.
Use:
"Restricted during the statutory/notified period."
rather than:
"Completely banned."
Similarly, do not write:
"All opinion polls are permanently banned during elections."
That is incorrect.
The important restriction arises from:
Section 126's silence-period rules
when the opinion-poll result constitutes election matter intended or calculated to influence the result.
The ECI has specifically explained that electronic display of opinion-poll or other poll-survey results falls within the relevant 48-hour restriction.
The ECI:
Article 324
Media organisations can:
Media freedom should coexist with electoral responsibility.
Voters should treat polls as:
Information, not certainty.
They should ask:
A poll can be useful when:
It becomes problematic when:
Headline:
"Party A WILL WIN."
Actual survey:
Party A: 42% ± 3%
These are not equivalent.
Correct interpretation:
The survey estimates support within a range subject to methodological uncertainty.
Forecasting is probabilistic, not deterministic.
Polls can be manipulated by:
Poll literacy is part of voter literacy.
Parties can use poll findings to:
Polls can influence the supply side of elections as well as the demand side.
Free and fair elections require:
No coercion.
No illegal advantage.
Voters can assess alternatives.
The information environment surrounding elections.
Polls can be wrong.
Different methods produce different results.
Voter preferences can change.
Exit polls can affect later voters.
Information spreads rapidly.
Polls can be weaponised as campaign tools.
Funding can create perceived or actual conflicts.
Publish:
Identify who commissioned the poll.
Separate:
Enforce Section 126 and 126A.
Teach voters how to interpret polls.
Track political amplification of poll results.
The fundamental policy dilemma is:
↓
Less risk of electoral manipulation
but
↓
Potentially greater restriction on information.
↓
Greater information freedom
but
↓
Potentially greater electoral influence.
Restrict narrowly during the sensitive voting period while preserving legitimate political information and analysis outside the prohibited period.
Survey
↓
Transparent methodology
↓
Disclose sponsor
↓
Disclose sample
↓
Explain uncertainty
↓
Responsible media presentation
↓
Voter interpretation
↓
Informed democratic choice
Voter casts vote
↓
Exit survey
↓
Data collection
↓
Restricted during notified period
↓
All phases complete
↓
Restriction ends as prescribed
↓
Exit-poll results may be disseminated subject to law
↓
Public analysis
↓
Official counting
↓
Official election result
Opinion Poll
↓
Voting intention
↓
Election
↓
Actual voting
↓
Exit Poll
↓
Estimated voting behaviour
↓
Counting
↓
Actual votes counted
↓
Official Result
Only the official counting process determines the election result.
Opinion poll and exit poll are the same.
❌ Incorrect.
Opinion poll → Voting intention
Exit poll → Reported voting behaviour after voting
Exit polls are permanently banned in India.
❌ Incorrect.
Section 126A imposes a restriction:
During the notified statutory period.
Section 126A deals with opinion polls.
❌ Incorrect.
Section 126A → Exit polls
Section 126 deals only with public meetings.
❌ Incomplete.
It deals with the 48-hour restriction and also prohibits public display of specified election matter through certain media.
Section 126 and Section 126A impose the same restriction.
❌ Incorrect.
126 → 48-hour silence period
126A → Exit-poll restriction
Opinion polls are completely prohibited throughout an election.
❌ Incorrect.
The important restriction concerns the applicable silence period under Section 126 and election-related requirements.
An exit poll gives the official election result.
❌ Incorrect.
It gives an:
Estimate
The official result comes after counting.
A larger sample always means a more accurate poll.
❌ Incorrect.
Sampling design, weighting, response quality and turnout assumptions also matter.
Exit polls are regulated only because they may be inaccurate.
❌ Incorrect.
A major concern is:
Potential influence on voters who have not yet voted, especially in multi-phase elections.
The ECI independently decides the election result based on exit polls.
❌ Incorrect.
Exit polls have no legal authority to determine election results.
Consider the following statements:
Answer: 1 and 2 only
Consider the following statements:
Answer: 1, 2 and 3
Consider the following statements:
Answer: 1 and 2 only
Consider the following statements:
Answer: 1 and 2 only
Consider the following statements:
Answer: 1, 2 and 3
Consider the following statements:
Answer: 1 and 2 only
Consider the following statements:
Answer: 1 and 2 only
| Provision/Concept | Association |
|---|---|
| Section 126 RPA | 48-hour silence period/election matter |
| Section 126A RPA | Exit polls |
| Article 324 | ECI |
| Opinion Poll | Voting intention |
| Exit Poll | Reported voting behaviour |
| Bandwagon Effect | Support for perceived winner |
| Underdog Effect | Sympathy for perceived loser |
| Sampling Error | Sample-based uncertainty |
| Poll of Polls | Aggregation of multiple surveys |
| Official Result | Actual counting and declaration |
| Question | Answer |
|---|---|
| Opinion poll | Measures voting intention/preferences |
| Exit poll | Surveys voters after voting |
| Opinion poll timing | Before/during election depending on survey |
| Exit poll timing | After voters cast ballots |
| Main exit-poll law | Section 126A RPA |
| Section 126 | 48-hour silence period |
| Section 126A | Exit-poll restriction |
| General-election restriction | First poll start → half hour after last poll closes |
| Penalty under 126A | Up to 2 years/fine/both |
| Election matter | Matter intended/calculated to influence result |
| Bandwagon | Perceived winner effect |
| Underdog | Perceived loser sympathy |
| Main methodological issue | Sampling/response/turnout |
| Official result | Actual counted votes |
| Constitutional institution | ECI |
| Constitutional basis | Article 324 |
| Core dilemma | Information vs electoral influence |
| Main reform | Transparency + narrow timing restrictions |
Opinion Poll
↓
"What are voters likely to do?"
↓
Voting
↓
Actual votes cast
↓
Exit Poll
↓
"How did voters say they voted?"
↓
Counting
↓
Actual votes counted
↓
Official Result
Section 126
↓
48-hour silence period
↓
Restricts specified election matter
Section 126A
↓
Exit-poll restriction during notified period
↓
Protects later-phase voters from premature exit-poll influence
Freedom of information
Freedom of expression
↓
Balanced against
↓
Free and fair elections
↓
Informed but unmanipulated voter
Opinion Poll
→ Measures voting intention
→ Usually before voting
→ Can inform voters
→ Can influence voters
→ Risks: Bandwagon + Underdog + Strategic voting
→ Accuracy depends on:
Sampling + Methodology + Timing + Turnout
→ During relevant 48-hour silence period, election matter is restricted under Section 126
Exit Poll
→ Surveys voters after voting
→ Estimates how voters voted
→ Especially sensitive in multi-phase elections
→ Section 126A
→ ECI notifies restriction period
→ General election:
First poll begins
↓
Restriction
↓
Last poll closes
↓
+ 30 minutes
↓
Restriction ends
→ Violation:
Up to 2 years +/or fine
→ Exit poll ≠ official result
What may voters do?
How did voters say they voted?
What did the counted votes actually show?
48-hour silence period
Exit-poll restriction
Perceived winner attracts support
Perceived loser attracts sympathy
Uncertainty from surveying a sample
Estimate
Official determination
Remember the topic through:
PREDICT → VOTE → SURVEY → COUNT → RESULT
Opinion Poll
↓
Estimates what voters may do.
↓
Citizens actually cast their votes.
↓
Exit Poll
↓
Estimates how voters say they voted.
↓
Election authorities count actual votes.
↓
Official result is declared.
The voter must receive information, but information released at the wrong time should not be allowed to distort the voting process.
Therefore:
Section 126 → Protects the final 48-hour voting period
Section 126A → Restricts exit-poll activity/results during the notified period covering the election phases
Opinion polls and exit polls are important tools of electoral analysis but must be distinguished carefully. An opinion poll generally measures voting intentions, whereas an exit poll surveys voters after they have voted to estimate actual voting behaviour. Their democratic value lies in informing public debate, but their potential to influence voter behaviour creates regulatory concerns. Section 126 of the Representation of the People Act, 1951 establishes a 48-hour silence period for specified election matter, while Section 126A specifically restricts conducting and publishing/publicising/disseminating exit-poll results during the period notified by the Election Commission. For a general election, this period can extend from the beginning of polling on the first day until half an hour after polling closes in all States and Union Territories. The key UPSC distinction is therefore: Opinion Poll = voting intention; Exit Poll = reported voting behaviour; Official Result = actual counted votes. The broader objective is to balance freedom of expression and voter information with the integrity of free and fair elections.
