COLOMBO — Sri Lanka opposition leader Sajith Premadasa and President Ranil Wickremesinghe gained ground in a voting intent poll for June 2024, with the former surpassing leftist candidate Anura Kumara Dissanayake by 13 points, and the latter catching up at only 10 points behind.
The Institute for Health Policy (IHP) which carried out the survey found that support for Premadasa increased to 43% (+4) but decreased for Dissanayake to 30% (-6) compared to May. Support for President Wickremesinghe increased to 20% (+3) whilst a generic Sri Lanka Podujana Peramuna (SLPP) candidate ran at 7% (-1).
IHP noted, however, that the June estimates were associated with a substantial increase in model uncertainty, and therefore the two leading candidates remain within the margin of error.
These estimates use the IHP’s Sri Lanka Opinion Tracker Survey (SLOTS) Multilevel Regression and Poststratification (MRP) model (01/2024 revision). IHP said in a statement on Thursday (Aug 01) that the latest update is for all adults and uses data from 18,213 interviews conducted from October 2021 to 21 July 2024, including 446 interviews during June 2024.
“A hundred bootstraps were run to capture model uncertainty. Margins of error are assessed as 1–11% for June,” the IHP said, explaining that the margins of error for June estimates were substantially higher than the previous months, which appears to reflect increased volatility in voter preferences, so the two leading candidates remain neck and neck.
IHP it said was planning to expand its interview numbers in the coming weeks and hopes that the next MRP update will provide more clarity on current trends.
According to the IHP, its SLOTS MRP methodology first estimates the relationship between a wide variety of characteristics about respondents and their opinions – in this case, ‘If there was a Presidential Election today, who would you vote for?’– in a multilevel statistical model that also smooths month to month changes. It then uses a large data file calibrated to the national population to predict voting intent in each month since October 2021, based on what the multilevel model indicates about their probability of voting for various parties (‘post-stratification) at each point in time.
“The multilevel model was estimated 100 times to reflect underlying uncertainties in the model and to obtain margins of error. The voting intent results for previous months have changed in this update as more data were collected after the last release and these changes are within the margin of error,” the institute said.
-economynext.com
Comments are closed, but trackbacks and pingbacks are open.