Dear Editor,
I have noted the article in June 8, 2016 edition of the Jamaica Observer titled ‘Don Anderson misses in St. Lucia’, making reference to a poll conducted by me there prior to the June 6 election, which was won convincingly by the Opposition United Workers Party (UWP).
For the record, and for the benefit of the general public, let me state the context and some of the data. I was commissioned by Dr Kenny Anthony’s ruling St Lucia Labour Party (SLP) to conduct a poll to test the level of public support for the party. Fieldwork for the poll was conducted April 7-14, 2016.
In that poll, taken significantly before even the election date was announced, the findings had the SLP ahead of the UWP in the majority of the 17 constituencies with at least two of them locked in a dead heat. The summary from that poll was that if the election was to be held then, the SLP would retain the government.
The election was not held then. Instead, on May 20, the Government called a snap election for June 6, allowing just 16 days to election day. No poll conducted eight weeks before an election can be used as a basis for predicting the outcome of an election.
My success and strong reputation as a pollster in Jamaica and several other Caribbean islands, (Grenada, St Kitts and Nevis and Antigua and Barbuda) has been built partially around the independent collection of data from several polls, generally a minimum of six for each election. This allows for a trend to be plotted and a clear indication established as to how the election could go. Certainly the last of these polls is normally conducted within a week of the election day. This is fairly standard for pollsters. So the poll conducted by me represents a reading of the popular support for the two parties at a point in time, April 2016, and was not intended to be used as a basis for predicting the outcome of the election. This would be impractical and unreliable.
The article goes on to say that my April poll was retested on May 28 and 29, a week before the election. Any retest was not done by me and, quite frankly, I do not understand what a retest in this context means.
So this article does not give a full representation of all the information or the context in which the poll done by me was carried out.
To complete the picture, my poll of April clearly showed that whilst the SLP was ahead of the UWP and likely to win the election if it were called then, there was a significant number of persons, over 52 per cent, that understandably had not yet made up their minds either whether they would vote or who they would vote for. Hence the possibility existed for these persons to significantly influence the eventual outcome of the election. This fact was communicated to the SLP leadership in my summary.
Further the campaign promise of the UWP to reduce/remove VAT was seen as tantamount to the $1.5-m tax plan of the Jamaica Labour Party here. This was always likely to be extremely attractive to a population experiencing hardships making ends meet. This was also something communicated to the SLP leadership in my summary.
These and other factors help to explain the massive swing to the UWP during the short campaign and certainly explain the difference between the level of popular support found in April and the results of the June 6 election. The essential point remains that the poll conducted by me in April, eight weeks ahead of the election, could not be and was not a prediction of the outcome of the election of June 6.
Don Anderson, CD
donanderson@flowja.com
I have noted the article in June 8, 2016 edition of the Jamaica Observer titled ‘Don Anderson misses in St. Lucia’, making reference to a poll conducted by me there prior to the June 6 election, which was won convincingly by the Opposition United Workers Party (UWP).
For the record, and for the benefit of the general public, let me state the context and some of the data. I was commissioned by Dr Kenny Anthony’s ruling St Lucia Labour Party (SLP) to conduct a poll to test the level of public support for the party. Fieldwork for the poll was conducted April 7-14, 2016.
In that poll, taken significantly before even the election date was announced, the findings had the SLP ahead of the UWP in the majority of the 17 constituencies with at least two of them locked in a dead heat. The summary from that poll was that if the election was to be held then, the SLP would retain the government.
The election was not held then. Instead, on May 20, the Government called a snap election for June 6, allowing just 16 days to election day. No poll conducted eight weeks before an election can be used as a basis for predicting the outcome of an election.
My success and strong reputation as a pollster in Jamaica and several other Caribbean islands, (Grenada, St Kitts and Nevis and Antigua and Barbuda) has been built partially around the independent collection of data from several polls, generally a minimum of six for each election. This allows for a trend to be plotted and a clear indication established as to how the election could go. Certainly the last of these polls is normally conducted within a week of the election day. This is fairly standard for pollsters. So the poll conducted by me represents a reading of the popular support for the two parties at a point in time, April 2016, and was not intended to be used as a basis for predicting the outcome of the election. This would be impractical and unreliable.
The article goes on to say that my April poll was retested on May 28 and 29, a week before the election. Any retest was not done by me and, quite frankly, I do not understand what a retest in this context means.
So this article does not give a full representation of all the information or the context in which the poll done by me was carried out.
To complete the picture, my poll of April clearly showed that whilst the SLP was ahead of the UWP and likely to win the election if it were called then, there was a significant number of persons, over 52 per cent, that understandably had not yet made up their minds either whether they would vote or who they would vote for. Hence the possibility existed for these persons to significantly influence the eventual outcome of the election. This fact was communicated to the SLP leadership in my summary.
Further the campaign promise of the UWP to reduce/remove VAT was seen as tantamount to the $1.5-m tax plan of the Jamaica Labour Party here. This was always likely to be extremely attractive to a population experiencing hardships making ends meet. This was also something communicated to the SLP leadership in my summary.
These and other factors help to explain the massive swing to the UWP during the short campaign and certainly explain the difference between the level of popular support found in April and the results of the June 6 election. The essential point remains that the poll conducted by me in April, eight weeks ahead of the election, could not be and was not a prediction of the outcome of the election of June 6.
Don Anderson, CD
donanderson@flowja.com