Fudging the figures: the Split in the polls, 1955-1975

The Labor Party Split and the emergence of a substantial group of voters prepared to support a new party caught the managing director of Australia's Gallup Poll flat-footed: Roy Morgan's initial reaction was to treat the Split as if it did not exist. When he did decide to recognise it, on the eve of the 1955 federal election, he faced another problem: should those respondents who wanted to vote for the Australian Labor Party (Anti-Communist) be treated as part of the Labor vote (which was how Morgan handled those who intended to vote for the Communist Party), or should they be treated as anti-Labor voters who preferred to see Labor kept out of office - at least until Labor had freed itself of communist influence? This chapter looks at how the Gallup Poll came to register the Split. It examines the various attempts Morgan made to measure support for the new party. And it looks at how he reported his findings, at the various rules of interpretation he applied to them, as he struggled to predict how well the Democratic Labor Party (as it came to be known) would do, and at how he fudged the gap between what his polls suggested, what he published and what the election showed.