Derek Achong

Se­nior re­porter

derek.achong@guardian.co.tt

Poll­ster and po­lit­i­cal sci­en­tist Pro­fes­sor Hamid Ghany has de­fend­ed the re­sults of a Guardian Me­dia-com­mis­sioned pub­lic opin­ion poll he con­duct­ed for the April 28 Gen­er­al Elec­tion.

In his poll, pub­lished in yes­ter­day’s Sun­day Guardian, Ghany ques­tioned 1,650 re­spon­dents in 11 mar­gin­al con­stituen­cies in Trinidad, and found that the Unit­ed Na­tion­al Con­gress (UNC) had a sta­tis­ti­cal edge over the in­cum­bent Peo­ple’s Na­tion­al Move­ment (PNM). The in­de­pen­dent sur­vey, which was al­so con­duct­ed in To­ba­go, found that the PNM is ahead of the To­ba­go Peo­ple’s Par­ty (TPP) in the two seats there.

Con­tact­ed yes­ter­day, sev­er­al po­lit­i­cal com­men­ta­tors, small­er po­lit­i­cal par­ties and can­di­dates ex­pressed scep­ti­cism over the re­sults and crit­i­cised the method­ol­o­gy used.

But Ghany was un­fazed as he de­fend­ed his work.

“You would have all kinds of views be­ing ex­pressed about it but the method­ol­o­gy is there and at the end of the day, that is the snap­shot from last week­end,” Ghany said.

Re­spond­ing to the crit­i­cism over his se­lec­tion of mar­gin­al con­stituen­cies us­ing a mar­gin­al­i­ty for­mu­la he in­vent­ed in 2002, Ghany said, “You have to cap mar­gin­al­i­ty on the out­er end at some point. There has to be some kind of method­ol­o­gy which would al­low you to cap the out­er part of mar­gin­al­i­ty. The in­ner part is al­ways ze­ro but how far out you go is the is­sue.”

He al­so dis­missed crit­i­cism over the sam­ple size utilised, in which 150 con­stituents, from three polling di­vi­sions in each mar­gin­al con­stituen­cy iden­ti­fied, were is­sued with ques­tion­naires.

“The sam­ple size was quite large be­cause oth­er polls that are usu­al­ly done use small­er sam­ple sizes than that. In fact, I have seen oth­er polls in which they use be­tween three and four hun­dred to give you an out­come,” Ghany said.

“So, this was a rel­a­tive­ly large sam­ple size for the is­land of Trinidad and it con­fined it­self to mar­gin­al con­stituen­cies where you are more like­ly to en­counter views that are mixed and are of a dif­fer­ent hue,” he added.

Deal­ing with the re­sults for To­ba­go, Ghany said he con­sid­ered the 2020 Gen­er­al Elec­tion re­sults and the re­sults of two To­ba­go House of As­sem­bly (THA) elec­tions in 2021.

“Us­ing all three, you are able to get a fair read of the shift­ing of po­lit­i­cal opin­ions, be­cause the THA elec­tion is much like a gen­er­al elec­tion in To­ba­go it is not like a lo­cal gov­ern­ment elec­tion in Trinidad,” Ghany said.

“The un­known fac­tor is the split with the Pro­gres­sive De­mo­c­ra­t­ic Pa­tri­ots (PDP). Some peo­ple may have de­cid­ed to go back to the PNM, while some peo­ple may have stuck with the PDP and some peo­ple may have moved on to the TPP. The chal­lenge was to mea­sure all of that on the ground.”

He al­so not­ed that the poll was not in­tend­ed to pre­dict the over­all win­ner of the elec­tion in terms of the even­tu­al di­vi­sion of con­stituen­cies be­tween the po­lit­i­cal par­ties.

“It is not pro­ject­ing who is go­ing to win which con­stituen­cy but it is telling you in a cu­mu­la­tive way what the po­si­tion is in those 11 mar­gin­als,” Ghany said.

PNM, UNC main­tain they have edge

Guardian Me­dia at­tempt­ed to con­tact Prime Min­is­ter Stu­art Young and Op­po­si­tion Leader Kam­la Per­sad-Bisses­sar to re­spond to the poll yes­ter­day but they did not re­spond to What­sApp mes­sages.

How­ev­er, ahead of the poll on Sat­ur­day, both Young and Per­sad-Bisses­sar cau­tioned their sup­port­ers about pub­lic opin­ion polls.

Speak­ing at a cam­paign meet­ing at the Moru­ga Gov­ern­ment Pri­ma­ry School, Young urged sup­port­ers to be vig­i­lant against what he de­scribed as UNC pro­pa­gan­da.

“I saw the lat­est polls. The PNM is lead­ing. The UNC is un­der se­ri­ous pres­sure in San Juan/Barataria, Ch­agua­nas East, To­co/Man­zanil­la, and Ma­yaro. And if we stay the course—Moru­ga/Table­land will be ours.”

Young warned that as elec­tion day draws near, sup­port­ers should brace for what he called an “on­slaught of lies and fake polls” from the UNC.

“My mes­sage is—we are less than ten days away from the elec­tion. Let me warn you, next week the UNC will step up its mis­in­for­ma­tion cam­paign and try to mis­lead. I told you, from to­mor­row they will start with fake polls.”

He re­mind­ed the crowd that, un­like oth­ers, he re­mained vi­brant, en­er­gised, and bat­tle-ready with his PNM team to lead T&T through tur­bu­lent times.

Blast­ing the Op­po­si­tion for what he termed “des­per­a­tion pol­i­tics,” Young said, “In San­gre Grande, their can­di­date can’t even speak prop­er­ly. And now, when they run out of ideas, they turn to race. They’re say­ing, ‘Run the Chi­nese out of town.’ But we love the Chi­nese in Trinidad and To­ba­go. That’s not who we are.”

He al­so said that the PNM’s in­ter­nal polls showed the UNC was los­ing San Fer­nan­do West.

“In San Fer­nan­do West, they’ve melt­ed down be­cause (PNM’s)Faris is putting a lick­ing on them. Tu­na­puna seat, (Es­mond) Forde is bring­ing them home.”

Mean­while, at a meet­ing in Las Lo­mas, Per­sad-Bisses­sar said she was not con­vinced that me­dia polls can ac­cu­rate­ly con­vey the out­come of the Gen­er­al Elec­tion.

She ac­cused Guardian Me­dia of be­ing po­lit­i­cal­ly in­flu­enced, de­scrib­ing the ex­er­cise as “sup­pres­sion polls” in­tend­ed to dis­suade her par­ty’s sup­port­ers and “keep them at home” for the elec­tion.

“We have been con­duct­ing our own in­de­pen­dent polling week­ly, since last year and that polling has been on­go­ing and up­dat­ing.”

In a post on his X (for­mer Twit­ter) so­cial me­dia ac­count, Pub­lic Util­i­ties Min­is­ter and PNM can­di­date for Diego Mar­tin North/East Colm Im­bert al­so de­scribed the poll as “ques­tion­able.”

“He (Ghany) has open­ly crit­i­cised the PNM for the last 30 years. His de­mo­graph­ics are to­tal­ly at vari­ance with of­fi­cial CSO da­ta, and his un­sci­en­tif­ic ‘find­ings’ should be dis­missed,” Im­bert said.

Youth De­vel­op­ment and Na­tion­al Ser­vice Min­is­ter and PNM La Hor­quet­ta/Tal­paro can­di­date Fos­ter Cum­mings al­so said, “As far as I am aware, the Pro­fes­sor is not a poll­ster so I will not com­ment on his find­ings ex­cept to say you call me back when a re­al poll is done.”

The PNM Women’s League al­so ques­tioned the cred­i­bil­i­ty of the poll based on what it said was the re­al­i­ty that Ghany is not known to be a qual­i­fied poll­ster but rather a con­sti­tu­tion­al ex­pert.

What oth­er par­ties said

Na­tion­al Trans­for­ma­tion Al­liance (NTA) po­lit­i­cal leader Gary Grif­fith ques­tioned the re­sults of the poll, which claimed six per cent of the re­spon­dents favoured his par­ty.

The poll al­so re­vealed that the PNM had 30 per cent sup­port, and the UNC-led coali­tion had 45 per cent sup­port in Trinidad. The Pa­tri­ot­ic Front (PF) re­ceived sev­en per cent sup­port, while oth­er po­lit­i­cal par­ties re­ceived 10 per cent sup­port.

“What is in to­tal con­trast in that poll is for the en­tire 37 years the UNC has been in ex­is­tence, they have nev­er won an elec­tion on their own when a third par­ty is in­volved,” Grif­fith said.

“For Dr Ghany to say that the UNC is ahead in mar­gin­al seats when there are 17 dif­fer­ent po­lit­i­cal par­ties, the more po­lit­i­cal par­ties out­side the PNM, it guar­an­tees that the UNC can­not win the mar­gin­al seats.”

“The ac­cu­ra­cy of these polls leave a lot to be de­sired,” he added.

Grif­fith al­so took is­sue with Ghany fo­cus­ing on mar­gin­al con­stituen­cies.

“He did a poll on­ly in the mar­gin­al seats, so it may be very well skewed... I would take this with a pinch of salt,” Grif­fith said.

PDP leader Wat­son Duke said his par­ty was not pay­ing close at­ten­tion to the re­sults of the poll, which claimed it was trail­ing be­hind the PNM and TPP in To­ba­go.

“We don’t take on poll­sters in the PDP. We use them as a guide, but that does not paint a true pic­ture,” Duke said.

“You see To­bag­o­ni­ans, the is­sue is very clear. It’s about jobs, it’s about op­por­tu­ni­ties, it’s about politi­cians not op­er­at­ing too hoity-toity when they get in­to of­fice,” he added.

Duke said the sam­ple size was too small and could not give a fair re­flec­tion of the mood of the elec­torate.

In an in­ter­view yes­ter­day, po­lit­i­cal an­a­lyst Dr Win­ford James said he still had to analyse the full re­port from the poll.

How­ev­er, he said, “These things take a lot of care and time to con­struct and if you are go­ing to crit­i­cise you bet­ter give more facts...These things are not sim­ple, and one can­not just crit­i­cise them. One must have a good ba­sis be­fore crit­i­cism is launched.”

How­ev­er, he dis­agreed with crit­i­cism over the sam­ple size be­ing in­ad­e­quate.

“In sam­ples of this kind that we are deal­ing with you don’t have to have big sam­ples. But, you have to have sam­ples that are well se­lect­ed giv­en the dif­fer­ent de­mo­graph­ics in the con­stituen­cies where your votes live,” James said.