I heard of the perfect storm but never of a perfect poll until now. The recent CADRES poll derives its perfection not from its methodology or its design or even its timing but rather from its political intrigue.
Stuart receives more support than Sinckler among DLP voters; he has an approval of 71 per cent in his job as Deputy Prime Minister among DLP supporters; he is ahead of Sinckler as the preferred leader of the country.
Yet the uncertain voters are used to miraculously push Sinckler ahead of Stuart as the preferred alternate leader of the DLP.
The DLP support among the voters declines from 44 per cent to 33 per cent, a dramatic fall of 11 percentage points or a fall of 25 per cent. The BLP support is marginally up. Yet the unknown swing analysis shows the DLP support improving by 0.7 percentage points.
The poll reveals the following: (1) Stuart is more popular than Sinckler in the DLP; (2) Snickler is more popular than Stuart among all Barbadians because of how the uncertain voters were manipulated; (3) Mottley is more popular than Arthur in the BLP and (4) Arthur is more popular than Mottley among all Barbadians.
The intrigue comes from the seemingly unsettled issue of leadership in both political parties. According to the poll there is an internal – versus external appeal that has to be resolved. The pollster obviously prefers the candidate with the external appeal in the DLP, while he prefers the candidate with the internal appeal in the BLP.
Never in the history of CADRES’ polling was there an environment of unsettled leadership in both political parties at the same time. So immediately one can question the statement that “the methodology used by CADRES on this occasion is similar to that which was employed in previous NATION/CADRES surveys and therefore need not be detailed in this report”. I beg to differ.
The CADRES poll is not sufficiently stratified to differentiate leadership between the internal party level and among all Barbadians. The fact that a person who is polled is a supporter of Democratic Labour Party does not mean that the person is a member of the party and of course the same reasoning applies to the Barbados Labour Party.
A party member has an opportunity to determine the internal leadership of the party; therefore a polled BLP/DLP supporter has to be a member of the party to have a say in its internal leadership. The external leadership derives its power from within parliament by way of the people.
Of course, the two types of leadership are related and it is in this sense that the methodology would have had to be different from previous surveys to reflect this very reality.
There is no doubt that any potential change in leadership, especially political leadership, could become messy but this could not be deduced from the scientific data collected and certainly not from the unscientific distinction made between national (external) and among party supporters (internal) in the recent CADRES poll.
The collision between the scientific and the unscientific is truly captured in the “summary judgement”, when it is observed that “public opinion on leadership in the Caribbean is invariably always influenced heavily by leadership roles and status and not the other way around”.
“In this regard, Stuart has an
advantage over Sinckler since he has acted in a leadership role over the past months while Sinckler has never held a national leadership role.” Yet the poll is a “referendum” on Stuart as acting Prime Minister.
Ironically a poll does not have to represent reality to have good predictive power. In logic a false premise can lead to a true/desired outcome.
Polling should not be monopolised in a small, personal and polarised environment. It is wrong!
