June 1, 2026

How much trust should we place in opinion polls, especially when election outcomes or public sentiment contradict them at times? While polls are often criticised as biased and politically motivated, in ‘Polls are a “public good”, they deserve to be better understood’ (LSE Impact, May 2026), Assistant Professor Ozan Kuru (NUS Communications and New Media) argues that even in an age of misinformation, polls remain a crucial democratic mechanism linking public opinion with researchers and the wider public.

Drawing on his article, ‘Conditioning Public Opinion Perceptions by “Survey Methods 101”: Informing, Engaging, and Motivating Individuals for Critical Processing of Public Opinion Polls’ (Public Opinion Quarterly, 2026), Asst Prof Kuru highlights how surveys function not merely as statistical exercises, but as symbolic manifestations of representative democracy. From election polling and policy sentiment to public health surveys, they provide structured ways of understanding collective preferences and social realities.

However, polls today face three major credibility challenges. The first is methodological concerns; while some surveys employ rigorous probability-based sampling, others rely on weaker methods such as misleading online polls, clickbait surveys, or poorly designed questionnaires. Emerging technologies such as artificially generated datasets created by algorithms further raise questions of reliability. Secondly, declining public trust, particularly after unexpected election outcomes, causes many to perceive polls as inaccurate or manipulative. Finally, partisan bias shapes how people interpret poll results, with individuals more likely to trust polls that support their preferred political positions while dismissing unfavourable findings as untrue.

Asst Prof Kuru argues that improving polling literacy is therefore essential. In his longitudinal experiment, participants underwent varying forms of survey methods training, ranging from informational modules to interactive quizzes and motivational messaging warning against low-quality polls. Those exposed to these interventions became significantly better at distinguishing between robust and misleading polls, unlike control participants who received no training. His findings suggest that relatively simple educational interventions can go a long way in strengthening critical engagement with public opinion data. Since existing educational resources on survey methods tend to be inaccessible or overly technical, Asst Prof Kuru also advocates for more public-facing approaches, such as gamified learning tools and stronger collaboration between university researchers and media organisations to strengthen public comprehension of polls.

In Singapore where surveys increasingly shape debates on household topics such as cost-of-living concerns, the real challenge may not be whether polls are accurate, but whether the public knows how to read them critically. As misinformation and online echo chambers grow, stronger polling literacy can help Singaporeans better distinguish credible findings from misleading claims, thus maintaining public trust and informed democratic discourse.

Read the ‘Polls are a “public good”, they deserve to be better understood’ here.

Read ‘Conditioning Public Opinion Perceptions by “Survey Methods 101”: Informing, Engaging, and Motivating Individuals for Critical Processing of Public Opinion Polls’ here.