Editorial: A love–hate relationship with polls

The results are in. The Antigua & Barbuda Labour Party (ABLP) and their supporters are not happy with the results of our most recent polling. The ABLP has dismissed the results out of hand. The conspiracy theorists in the ABLP camp claim that it is all part of giant plan to discredit the party, and by extension, the government, on its third anniversary.
Needless to say, the members of the United Progressive Party (UPP), on the other hand, are extremely happy. To their minds, the poll accurately represents how people think and they are buoyed that the results reflect their (i.e. the party’s) way of thinking.
Completely rejecting or completely accepting the results of any poll is a common error made by politicians and political parties. If the results comfort them, then they ask no questions. If it goes against the grain, then it is rejected as being faulty in design and/or execution. Seldom taken into consideration are the detailed poll numbers, the conditions for the poll, the technique utilised to collect the data, and the overall environment (among all the other factors).
Thinking in absolutes is one of the Achilles heels of politicians. The reason is simple. They surround themselves with people who say “yes” a lot, and shower them with accolades. This causes them to see the world through a distorted lens where opposing views are dismissed because they do not conform with the norm – an artificial norm that creates a protective cocoon around their minds, forcing them to reject any form of criticism.
Online polling is a particularly difficult thing from which to extract meaningful data. We are not bashing our polls but we are simply pointing out that the very nature of the polling technique will always be skewed to those that can get their people out to vote or are passionate about a particular topic. We have learned the hard way that online polling is subject to manipulation. That is why we have put in the safeguards to protect the basic integrity of our polling. For example, only one vote per IP address. In one of our early polls, we noticed a spike in poll numbers and our investigation revealed that the majority of votes came from a single IP address. Once we eliminated the multiple vote IP addresses, the results returned within the boundaries of normalcy. We may have eliminated legitimate results from a home or business that share an IP address but that is a price that must be paid.
With that being said, we will try to apply some common sense to interpreting the results. For the record, we are not statisticians and have no experience in soothsaying based on poll results. In fact, we continue to adhere to the wise words of E B White, who said, “The so-called science of poll-taking is not a science at all but mere necromancy. People are unpredictable by nature, and although you can take a nation’s pulse, you can’t be sure that the nation hasn’t just run up a flight of stairs, and although you can take a nation’s blood pressure, you can’t be sure that if you came back in twenty minutes you’d get the same reading. This is a damn fine thing.”
Having laid that groundwork, here is what we see coming from the poll. The first is that the UPP seemingly did a better job at activating their supporters and enticing them to participate. The positive results for caretakers, with little experience in the communities, coupled with the overall negative polling for the ABLP leads us to that conclusion. And that should be the takeaway. For the ABLP, it should be a bit concerning that the UPP supporters were more energised. Elections in our small constituencies are won and lost because the base is energised and people come out to vote.
Having said that, the rest of the results can be put in context. Notice we did not say thrown out. It is not unusual for incumbents to poll low but focus should be put into the different candidates rather than the UPP versus ABLP debate. For example, how did Molwyn Joseph and Samantha Marshall poll mostly positive while others did not fare so well? If the argument is that the overall poll respondents were biased, then that would conflict with the individual results.
A similar logic can be applied to the other side. Overall, it seems highly unlikely that caretakers would poll so well. They have limited experience in politics and in the constituencies so what would cause them to have such good results? We readily admit that sometimes fresh faces have a positive effect, since they carry little political baggage, but the chances of that single factor having a blanket positive effect is slim to none. The takeaway should then be that the ‘good feeling’ results may not convert to tangible results in a hard fought campaign. The positive of being a fresh-faced candidate may erode when the mud-slinging begins. The UPP would be wise to utilise caution and refrain from patting each other on the back.
So while this may be an unscientific poll which has all the weaknesses of being conducted online, it should not be dismissed or accepted in absolutes. All data has usefulness, the magic is in extracting that quality. We will continue to take the pulse of the nation as it runs up and down the flights of stairs that comprise life. The results will always reflect a moment in time but it is still a reflection that is worth a second glance.
We invite you to visit www.antiguaobserver.com and give us your feedback on our opinions.