New Vision "Presidential" Poll Misleading.

By Arthur M.M. Katabalwa.

Do we have a law of limitations in Uganda? I have a confession to make. Over two decades ago, when my political mind started to grow, I managed to single handedly manipulate the outcome of an election. I was spending days watching American politics and any foreign television I could find. I started watching intently the polls as Bill Clinton was being elected President of America.

It was really fascinating to see how the polls were swinging the way people in the US were reacting to the campaign. After the first Gulf War, I thought that Bush senior winning the election was a given. But then I was lured into the election of Governor Bill Clinton from Arkansas. And it wasn't just his youthful looks then. It was the message. The oration but most importantly the polls were on his side.

So, armed with some little information that I had got from C-Span about polling, I decided to find out if I could influence an election that was dear to me that was taking place where I living. I decided to publish a poll. It was not scientific at all because, I made it all up. For the three weeks before election day, I posted a tracker poll on a public notice board. For starters, no one paid attention to it. By the fourth day people were waiting for me at this public notice board to read and see what the polls were saying. I never, ever polled anyone. In fact on many occasions I made the poll up minutes before. But I held the electorate and for two days, the two protagonists were neck and neck. When the poll started to "run away" with my chosen candidate the campaign became electric. Can I therefore say that I rigged a poll without even casting a ballot?

Fast forward to a New Vision poll that was published only a few days ago and I can tell you that I view these statistics with a pinch of salt. One has to ask many questions as to the authenticity and accuracy of these polls. But most importantly as in the New Vision Poll, we need to first examine the question itself. The respondents were asked to answer: "Which Presidential Candidate Will You Most likely Vote For Next Year?"

Quite easy isn't it? I bet many people already have a vague clue to that answer. We all know whom we will vote next year? Right? Wrong. The New Vision and its research team committed the first fundamental mistake in poll science. They came up with the wrong question on so many levels. First of all, the poll, according to the The New Vision was conducted by its researchers. Well, they had better sack that research team because they should have known that as of June 12 to June 20 2015 Uganda did not have any Presidential Candidates! Unless there is something that they know that the electorate isn't aware of. But the last time I checked, on that date we had a few aspirants, a number of whom were being arrested and tear gassed for having aspiring to the highest office in the land. And by the way that is a crime in Uganda. So that question was aiding and abetting illegal or premature electioneering. You had better thank the gods Kawesi has been moved somewhere in the police!

For anyone to get this polling stuff right, you need to get the question right. One which doesn't bias people on the onset. One that is understood. Recently we had a referendum in Greece about their financial bailout from the Euro Group and others. That question flummoxed even the most hardy economists. One needed banks of economic forecasts so as to answer that question. It is therefore regrettable that the New Vision and the research team it used failed at the first huddle, which then made the whole exercise invalid. They were asking the wrong question at the wrong time.

But, 6626 respondents were reached (This is already from a flawed premise, a flawed question) and asked a wrong question. One then continues to ask: What demographics did they target? Did they cover all the demographics? Which income bracket? What religion, educational level? As we have been told before, our society is predominantly peasant. So is it the case that the statistics presented were of mainly peasants? Which in my view will not be representative. But to their credit, the summary shows that President Yoweri  Museveni wins in the rural areas. The others win in the urban areas. If one has been observant of the political landscape, one might say that that is broadly true.

The statistics further provided a list of people whom The New Vision (I think) also branded "Presidential Candidates". (I am still struggling here to quantify and qualify the whole exercise because of that question). But it only gets better. Is Janet Kataha Museveni a Presidential Candidate for the 2016 elections? The poor lady only gets 0.1% of the total vote compared to her husbands' alleged 71%. (I would have liked to be a fly on the wall when they were having breakfast that day. The conversation!) Who in heavens sake is Ajok Lucy? At 0.1% she is as popular as the First Lady? Nantaba Idah Erios 0.1%, Tann Sanjay 0.1%.....Asamo Hellen Grace 0.1%! Are all these people Presidential Candidates? That is going to be a very long ballot paper. Then (and this is where some might get confused in the statistics), we have the Don't knows at 2.1%, None at 1.4% and the Refused To Answer at 1.3%.

The whole exercise was farcical in my view once they adopted that question. At 19, armed with a rudimentary understanding of the psychology behind polls and a rickety, faulty TV dish pointing at another equally rickety faulty TV dish 3 miles away so as to tap there signals, I was able to manipulate an election through unscientific polling and my candidate won. In the 21st Century we need to be doing a better job while carrying out polls and get the question right. Else the whole exercise is not worth the paper it was written on.

mwenky99@gmail.com