By RASHAD ROLLE

Tribune Senior Reporter

rrolle@tribunemedia.net

THE Free National Movement commissioned an internal poll that it says shows Prime Minister Dr Hubert Minnis leading Progressive Liberal Party leader Philip “Brave” Davis in approval ratings.

According to a memo obtained by The Tribune, Ragnar Research Partners conducted 400 interviews between July 1 and July 6 by live telephone operators and asked the respondents if they approve or disapprove of Dr Minnis and Mr Davis.

According to poll results, 54 percent of Bahamians approve of Dr Minnis while 37 percent disapprove of him; 24 percent of Bahamians approve of Mr Davis while 47 percent disapprove.

Twenty-nine percent of people strongly approve of Dr Minnis; 24 percent not so strongly approve of him; ten percent “don’t know”; 16 percent not so strongly disapprove of him and 21 percent strongly disapprove of him, according to the FNM’s internal poll results.

Twelve percent strongly approve of Mr Davis; 12 percent not so strongly approve of him; 29 percent don’t know; 19 percent not so strongly disapprove and 29 percent strongly disapprove of him.

Respondents were also asked who they trust more on the country’s response to COVID-19: Dr Minnis and the FNM, or Mr Davis and the PLP. Forty-three percent said “definitely Minnis”; 12 percent said “probably Minnis”; 27 percent “don’t know”; six percent said “probably Davis”; and 12 percent said “definitely Davis,” according to the poll.

The pollsters said: “Quotas on age, gender, ethnicity and region were used to ensure a representative distribution. The study’s margin of error is five percent.”

The poll comes as some in the Free National Movement suspect Dr Minnis will call an early election. Dr Minnis is said to be buoyed by the internal poll results and the FNM is expected to put the leadership battle between him and Mr Davis at the front and centre of their campaign. The party is said to be planning a short but vigorous campaign.

No public polls relating to the next election have been released to date.

Joey Gaskins, senior partner of Open Current, a research and government public relations firm, said yesterday that as with public polls, the methodology of internal party polls must be carefully scrutinised.

“Just because a poll is internal doesn’t mean it shouldn’t be trusted, although when internal polling is released it often comes with an agenda,” he said. “This is something to be especially wary of if only certain questions and not the results in their entirety have been released. Trusting a poll should be based on its methodology and how representations are made about the results of the poll based on that methodology.”

The FNM has relied on polls by Ragnar Research Partners head Chris Perkins before. A poll he conducted between November 29 and December 5, 2016 showed Bahamians disapproved of the direction the country was heading in at the time.

FiveThirtyEight, a well-known website that assigns ratings to pollsters based on the historical accuracy and methodology of an organisation’s polls, has assigned Ragnar Research Partners a B/C rating on the basis of two polls the company conducted in America.

Nonetheless, M’Wale Rahming, head of Open Current’s sister company Public Domain, a local market research and strategy firm, said he typically strives for polling samples larger than the 400 used for the FNM’s recent poll.

“Any poll is worth something as long as you state the methodology so that any reader of the poll can take the right things from it,” he said. “We shy away from using a sample of 400 if it’s anything where we need the margins to be more specific and you want to get as granular as possible.

“To draw conclusions at the national level of a poll of 400 randomly selected people might lead to conclusions that are not entirely representative. The margin of error is five percent, so that could result in a ten-point swing. If you have a ten-point swing, that could completely change what you are looking at. At Open Current, when doing a national poll we try to make sure we’re using as robust a methodology as possible. With a representative sample size of 400, for example, Grand Bahama is only 15 percent of the population, so you would essentially have about 60 people representing 50,000 people. That reduces the likelihood that you will find 60 people that adequately represent Grand Bahamians from East to West End. A chunk of that 60 could be concentrated in one area and that could skew your results and looking at the margin for error for that subset, you may not be as accurate on the ground as you’d want to be,” Mr Rahming said.

For his part, Mr Gaskins said: “Any poll we do for public release, we definitely aim to use a sample size of 1,000. For us at Open Current, a poll of this nature can be very informative but we would not in our internal operations consider it to be decisive. At Open Current we think that polling as it relates to government and policy in particular, it’s an important part of the democratic process.”

A spokesperson for the PLP said: “This is a garbage poll with a non-representative sample conducted by foreign interviewers -- it does not reflect national sentiment. Bahamians have had enough of Minnis’s failed policies and contempt for the people. The FNM is lying about the poll just like they lied about a fixed election date -  just like they’re lying about not intending to raise taxes (again) on the small man. Lies are all they have left.”