PUBLIC OPINION pollsters at the University of Hong Kong have been updating their top 10 list of legislative councillors every two months.
Results have been fairly consistent over the past five years. Only 13 of the 60 incumbents have made it on to the list since the current Legco session started in October 2000. The exercise has become such a routine that perhaps it is time for a change.
Eight of the 10 slots are usually occupied by those returned by universal suffrage from the geographical constituencies. Unionists Lau Chin-shek and Lee Cheuk-yan and Legco President Rita Fan Hsu Lai-tai have again been named as the three most popular lawmakers in the latest survey conducted early last month. On a scale of 100, their respective scores ranged from 60.1 to 58.8.
Though electronic media continue to announce the results of the Chief Executive's popularity rating, some no longer bother to report those for councillors. But the print media remain obsessed about minor shifts in the ranking. In the latest case, some newspapers have chosen to magnify the fact that Mr Lau had again topped the list, while Emily Lau Wai-hing of the Frontier had slipped from sixth to ninth place.
Such titbits are indeed meaningless, if not downright misleading. More than 1,000 Cantonese-speaking citizens aged 18 or above were approached at random for each of the programme's telephone polls. The response rates were set at over 60 per cent. The pollster has pointed out that 'at 95 per cent confidence level, the sampling error of legislators' ratings was less than 1.4 marks'. That means the differences between the scores of Mr Lau, Mr Lee and Mrs Fan were statistically insignificant. Technically speaking, it was not possible to pinpoint which of them was indeed number one.
By the same token, Ms Lau could have ranked anywhere between fifth and 10th place, had the margin of error been taken into consideration. Despite the researchers' warnings, the media simply does not care to read the fine print. The opinion series was originally meant to be a political barometer. Unfortunately, it has now been largely relegated to indulging the media's appetite for stories of winners and losers.
The survey comprises two parts. Respondents are first asked, without prompting, to recall the names of up to 10 legislators. They are then requested to grade the performance of each of those legislators.