ITING its new Code of Conduct for media, the Election Commission of Pakistan earlier this week asked Pemra to prohibit TV channels from airing the results of opinion surveys, arguing that `these can influence the free choice of voters`. This appears to be an overly conservative view of what is otherwise a seemingly harmless enumeration exercise that is conducted widely around the world in the run-up to elections.

Surveys are used to gauge public sentiment and predict the outcomes of different electoral contests. They not only provide election observers with a means to better evaluate results based on pre-existing data, but they also act as a safeguard against possible manipulation of the election process by providing a public reference point against which the freeness and fairness of the overall exercise can be judged after the polls. It defies understanding why the ECP would seek to prohibit them altogether instead of introducing measures to ensure that they are not misused.

Clause 12 of the ECP`s Code of Conduct states that traditional media, as well as social media influencers, should refrain from conducting entrance or exit polls `or any kind of survey` at `polling stations or constituency [sic]`. While the restriction on election day polling, especially at busy polling stations, is still understandable, it seems excessive to expand the restriction to entire constituencies and well before the election as well.

Indeed, those inclined to take a more conspiratorial view have been saying that the ECP`s decision to ban surveys altogether is, in fact, aimed at suppressing mentions of the PTI`s popularity, which has proven rather resilient. Instead of banning, the ECP may consider issuing guidelines through which it may demand a greater degree of accuracy, technical rigour and more transparency regarding how each survey is designed and conducted. Banning them outright will deny observers important reference material and also give rise to further controversy about an exercise whose fairness is already under debate.