Sir, - You carried and responded to my letter of May 22nd on Tuesday, June 2nd. In thanking you for that courtesy, I have to point out that your response was factually inaccurate. You say that Irish Independent/IMS poll of May 18th "... suggested a Yes vote of 65 per cent," while the Harris poll for The Irish Times could be seen as "... predicting a 71 per cent Yes vote." Your own figure derives from re-calculating stated voting intentions, having excluded the "don't knows". In a similar manner, the Irish Independent/IMS Poll of May 18th showed 68 per cent of committed voters intending to vote "Yes" as of that Monday. I would be grateful if you would correct the inaccurate reference to a lower figure.
The question of "prediction" from polls conducted prior to an election or referendum is a broader one, where interpretation comes into play. Based on the IMS results, the Irish Independent as early as Wednesday May 20th headlined: "Final poll puts Trimble close to crucial 70 per cent". With the benefit of hindsight, you suggest the survey conducted by Harris for The Irish Times predicted a 71 per cent Yes vote. In fact, your headline to the detailed review of your own poll results, as late as Thursday May 21st (the eve of voting) only went so far as to suggest: "Vote in Northern Ireland of 65%-70 per cent in favour of agreement still possible". Hardly a resoundingly confident prognostication! - Yours, etc., Charles M. Coyle, Director, Irish Marketing Surveys,
Upper Pembroke Street, Dublin 2.