Photo courtesy of Taiwan News
In the latest Sri Lanka Opinion Tracker Survey (SLOTS) polling in June 2023, support amongst likely voters for the NPP/JVP fell five points and for the SJB four points from May to 23% each. The UNP saw no gains, with a big increase of seven points to 28% in the share of voters saying they would vote for other parties.
In the four months since February the two leading parties, the NPP/JVP and SJB, have experienced substantial declines in support in the SLOTS polling with the share of voters saying they would vote for other minor and unnamed parties increasing by 17 points to 28%. The UNP and SLPP have seen modest gains but support for the UNP seems to be plateauing at one in ten likely voters, neck and neck with support for the SLPP.
“Other” parties in the SLOTS polling includes both smaller parties that ran in the 2020 general elections, such as the AITC and OPPP, as well responses where respondents indicated they would vote another unnamed party.
Dr Ravi Rannan-Eliya, director of the SLOTS polling, revealed that in most cases these respondents are opting for an unnamed other party, and that the SLOTS poll is working on identifying these further. He also commented that while the “other” category in the SLOTS polling in late 2021 was probably dominated by respondents who were too shy to voice opposition to the ruling party, the recent increase in this category seems to signal growing disillusionment with all the established parties in both government and in opposition.
SLOTS combines interviews from a national sample of adults (ages 18 and over) reached by random digit dialling of mobile numbers, and others coming from a national panel of respondents who were previously recruited through random selection. IHP estimates voting intent using an adaptation of Multilevel Regression and Post-Stratification (MRP), with multiple imputation to account for uncertainties in its modelling, exploiting data from all SLOTS interviews to estimate voting in a particular month.
The June 2023 MRP estimates are based on 506 interviews conducted in June 2023, and 11,926 interviews conducted from 1 October 2021 to 20 July 2023, with a margin of error assessed as 1–3% the various parties. As the June update uses a more recent data set than the previous update, there are small changes in estimates of voting shares for previous months. A total of 60 stochastic simulations were used in the modelling to estimate the uncertainty associated with the estimates arising from sampling noise in respondent interviews and uncertainties in the modelling process.
MRP is a method that is increasingly used by polling firms in other countries to leverage small samples, most notably by YouGov which used it to forecast results of the UK Brexit Referendum and recent UK general elections. All estimates are adjusted to ensure the sample matches the national population with respect to age, sex, ethnicity, socioeconomic status, education, geographical location, and voting in the 2019 Presidential and 2020 General Elections.