There is a lesson to be learned from the Tobago polls

Dear Editor,

In your news item ‘PNM wins all Tobago seats in historic result’ (SN, Jan 23), I conducted tracking opinion polls for NACTA from January 2011 till last Monday conducting no less than seven polls in as many trips to the island, and did not rule out such an outcome though I thought it would be highly improbable.  The result shocked many Tobagonians and as such an explanation is needed on why there was a blow-out.  There is also a lesson to be learned from different polls that gave conflicting findings not only to Trinis or Tobagonians but to Guyanese politicians as well.

The findings of the NACTA tracking poll findings were consistent for an incumbent PNM victory, but the number of seats was in doubt; the poll could not confirm a sweep although it had the PNM winning six seats by a huge margin and one with a smaller margin with five seats within its margin of error. At one time, the PNM was leading 11-1 in October, but the lead slipped thereafter with seats tightening up. In all the NACTA polls in December, the PNM was within a 3% margin in those seats where the opposition was leading, and it remained that way through January 21.

So a sweep was not impossible and in fact that was so stated in a NACTA release – all the seats were winnable by the PNM with the most the PNM could win being seven although it was leading in only three seats by a maximum of 3% while the other four were dead heats.

Numerous other polls were conducted. Another North American poll (not NACTA) had 11-1 and said a sweep was also possible. A poll from UWI Sociology Professors had it 10-1 with one seat a tie.  Another UWI poll from a different Professor (from Psychology), commissioned by the opposition showed a 7-5 victory in favour of the opposition. Another UWI government Professor (a constitutional expert) conducted two polls with the first showing the opposition six per cent behind the PNM but gave no seat numbers and the second poll showing the two parties neck and neck in support and seats. A poll conducted by a San Fernando based pollster showed 9-3 in favour of the PNM.  A poll conducted by a pollster affiliated with the PNM showed the PNM leading in all twelve seats, but gave the TOP support at less than 20% in each seat except in two seats: one had the TOP at 30% and trailing by 3% (a dead heat) another at 25% and trailing by 5% (another dead heat).

Another PNM commissioned poll had the PNM ahead in nine seats with the other three close.  And yet another pollster from central Trinidad commissioned by TOP, which conducted several polls, showed a close contest with its final poll showing TOP winning seven seats. And based on that projection, the opposition chilled the champagne for a victory celebration on Monday night. A TOP victory was difficult because the trends in all the independent polls were showing a clear PNM victory.  What no one predicted was the huge margin in votes in some of the seats that were supposed to be cliff-hangers.

What happened in Tobago is a lesson for pollsters as well as those parties that commissioned polls – ask the pollsters to be honest because it can’t be that independent polls were consistent with one another while the commissioned polls were showing victories for their parties.  This was somewhat similar to what happened in Guyana in November 2011 in which polls commissioned by the parties showed them winning. In Tobago, regarding polling, one can’t really blame the pipers because the masters don’t want to hear a tune out of sync.

Politicians have to be willing to listen to independent thinkers and those not affiliated with them. They would get a much more balanced and accurate view of their electoral fortunes. In Tobago, the independent pollsters had it right in terms of seats and the voting trends in the electoral districts. But they had a few elements wrong especially with regard to projected turnout. For example, NACTA expected a turnout over 60% but not quite 70%.  One poll had a turnout of 65%.  One poll had a turnout of 50%. The PNM affiliated pollsters had a moderate turnout similar to last election of 2009. The pollsters associated with TOP expected a low turnout and felt a higher turnout would benefit its candidates.  A higher turnout meant voters were coming out to vote against the incumbent in Trinidad and not the incumbent in Tobago as the voters were angrier with the Trinidad government than with the Tobago government.  They told our interviewers such although they were not pleased with the PNM.  They did not like the neglect they suffered from Port of Spain and the many mistakes made by the Prime Minister Kamla Persad-Bissessar led administration. They also felt they were being disrespected by POS.

The large presence of ministers from Trinidad in Tobago was not helpful.

In NACTA opinion surveys in May 2010, it appeared that the opposition TOP would win 11-1 in any local election, and general election results confirmed TOP would have won 12-0 and it remained that way for several months.  However, opinions began to change as the central government of which TOP was a member began to make blunders and did not take corrective measures. People saw TOP as being joined at the hip with the PP government and it was going to be punished.Had TOP stayed independent, it would have won the THA election or at least have done substantially better and the TOP affiliated pollsters failed to so inform their masters for whatever reasons.

Pollsters have a professional duty to tell those who commissioned them the exact findings of polls so campaign managers can plan accordingly to improve the chances of winning. Misleading party leaders or telling them what they want to hear will not help the parties.  And party leaders have to be wiling to accept results and not cuss out the pollsters.

Yours faithfully,
Vishnu Bisram



Join the Conversation

After you comment, click Post. If you're not already logged in you will be asked to log in or register.

The Comments section is intended to provide a forum for reasoned and reasonable debate on the newspaper's content and is an extension of the newspaper and what it has become well known for over its history: accuracy, balance and fairness. We reserve the right to edit/delete comments which contain attacks on other users, slander, coarse language and profanity, and gratuitous and incendiary references to race and ethnicity. We moderate ALL comments, so your comment will not be published until it has been reviewed by a moderator.