Election campaigns often generate strong emotions, including when expectations do or do not align with electoral outcomes. This makes it all the more important to approach political analysis with sobriety, reflexivity, and respect for evidence.
In recent weeks, both before and after the results of Malta's general election, considerable attention was given to opinion polls commissioned and published by media organisations and pollsters. Here I am referring to surveys that employ methods widely used in social-scientific research and electoral polling around the world, including those published by Prof Vincent Marmara, Times of Malta, and MaltaToday.
Unsurprisingly, these polls generated extensive discussion in political circles, on social media, and in less visible social settings. This is healthy and entirely normal in a liberal democracy. Public debate about political trends, voter behaviour, and electoral prospects is part of democratic life.
Some commentary, however, focused less on engaging with the findings and more on attacking the messengers. Such reactions moved beyond legitimate methodological critique and veered into attempts to discredit researchers carrying out their professional work. On social media platforms such as Facebook, some commentators went as far as claiming that polls are manipulated and should be banned altogether.
Criticism of methods is both healthy and necessary in any scientific endeavour. Personal attacks and attempts to delegitimise research because its findings are politically inconvenient are something else entirely. Such reactions often reveal more about the partisan predispositions of the critics than about the quality of the research itself.
Another limitation of highly partisan interpretations is their tendency to privilege a single explanatory factor while ignoring the broader social and political context. Polls are only one tool among many for understanding electoral dynamics. Qualitative analysis is equally important.
Political outcomes are shaped by events, narratives, leadership perceptions, grassroots mobilisation, constituency work, house visits, organisational capacity, and countless interactions that occur away from the spotlight. Understanding politics requires methodological pluralism rather than the selective use of evidence that confirms prior assumptions.
Indeed, like all social-scientific methods, polling has both strengths and limitations. Concepts such as validity, reliability, sampling, margins of error, and response bias are familiar to anyone with a basic understanding of social research. This is precisely why polling is a social scientific exercise rather than a dogmatic one.
At the same time, social surveys should not be treated as infallible predictors of electoral outcomes. That is the business of prophecy, not social science. Researchers have long recognised that surveys can be performative: they may influence perceptions and expectations while also reflecting them. Yet such influence cannot be reduced to a single outcome. Respondents may answer strategically, withhold information, or change their minds. These complexities are part and parcel of social research and are regularly discussed within social science.
This is a far cry from claims that reduce polling to a simplistic narrative of manipulation whenever findings fail to align with political preferences. Such critiques often assume a one-directional process whereby researchers, media organisations, or political actors shape public opinion at will. In doing so, they overlook the complexity of social life and underestimate the agency of citizens, who interpret information, discuss politics, and make choices in diverse and at times unpredictable ways.
As sociologists know, social reality is complex and multi-causal. No single method can capture it in its entirety. This is why serious analysis requires a combination of quantitative and qualitative approaches, together with a willingness to revise interpretations in light of evidence.
In a democratic society, disagreement is both inevitable and healthy. What is less healthy is the temptation to dismiss evidence, attack researchers, or silence sober analysis whenever findings challenge one's expectations. Democracy is best served when debate is informed by critical inquiry rather than partisan tribalism.
Michael Briguglio is Associate Professor at the Department of Sociology, Faculty of Arts, University of Malta
www.michaelbriguglio.com