Hong Kong election pollsters in U-turn on methodology after complaints by parties
From now on HKU academics will tell respondents who is second on the list of candidates in each geographical constituency
Angry parties, including the Democrats, People Power and the Liberal Party, had pointed to the small sample size of the poll – 100 to 300 people in each of the five geographical constituencies – and the pollsters’ failure to mention aspirants placed second on a slate when questioning respondents.
Several outgoing lawmakers are running second on tickets in an attempt to secure seats for their protégés. Under the proportional representation system these veterans have little chance of winning.
In a statement last Friday, the HKU Public Opinion Programme said it would not consider altering its methodology as it had found no significant statistical difference between mentioning one and two candidates on the slate in a test when polling 469 people from August 11 to 15.
But it backtracked and decided to mention two candidates on the slate to respondents from Monday night onwards.
“[HKU POP] made this decision after taking into account opinions from all walks of life and we believe [the change] is manageable,” Frank Lee, the programme’s research manager, told the Post.