As the election approaches, we are seeing the results of various types of surveys. In almost every country, different kinds of polls are conducted ahead of elections in an attempt to gain preliminary insights. Various organisations and political parties carry out their own surveys to better understand the state of the electoral landscape. Analysts often try to interpret the political situation by combining the results of these polls with their own insights.
Sometimes, the outcomes of these surveys align with reality; other times, they do not. There are instances where the actual election results turn out to be the opposite of what the polls predicted. Calculating voter behaviour is by no means an easy task, which is why poll results can often be inaccurate.
When surveys are conducted with a particular agenda, they are even more likely to diverge from the actual election results. Still, many organisations and political parties carry out such polls to serve their own interests.
In the language of research methodology, such surveys are referred to as biased or purposive surveys. These surveys are designed with a specific outcome in mind—usually what the organisers hope for or want to see. While surveys can reveal the real situation on the ground, they can also be manipulated to produce desired results. It all depends on how the data is collected and how the sampling framework is structured.
For example, suppose a survey is conducted in the city of Bonn, Germany, to study smoking habits. If the survey is carried out in and around central bus and train stations or bars, the results are likely to favour smoking.
On the other hand, if the survey is conducted among members of environmental organisations or student groups at the University of Bonn, the results will likely be negative toward smoking. A survey conducted across the entire population of Bonn might provide a more accurate picture—but that would require more time and resources.
Such long and comprehensive surveys are generally not undertaken for election or public opinion research due to practical limitations.
Recently, a number of politically motivated research findings have emerged in the country in the context of the election. It is important to note that these surveys are not easy to dismiss as incorrect or flawed. If one reviews the research design, sample size, and data collection methods, it would appear that the surveys were conducted following scientific standards and established methodologies. Yet, the results seem to favour a particular group.
For example, toward the end of Sheikh Hasina's last term, several surveys suggested she would return to power with overwhelming public support. However, the reality was that in the January 2024 election, not even 5 per cent of the population voted. Even supporters of the Awami League did not go to the polls. This clearly indicates that those surveys were politically driven.
A similar wave of biased surveys is now circulating. After Sheikh Hasina fled the country, a survey conducted by the BRAC Institute of Governance and Development (BIGD) in October of last year showed that if elections were held at that time, only 16 per cent of the population would support the BNP, while 38 per cent said they did not support any party. Yet public support for the BNP was at its peak during that period.
Sheikh Hasina had just fled the country. At that point, BNP members had not yet fully mobilized across different sectors. Now, a survey conducted in July shows a decline in public support for the BNP—only 12 per cent of respondents expressed support for the party, while 49 per cent said they were still undecided.
The October survey from last year didn’t receive much attention, likely because it came right after the Awami League was forced to step down. However, both surveys conducted by BIGD raise some concerns. It is true that people have become frustrated with certain criminal activities at the grassroots level involving BNP members—this cannot be denied. Still, it is hard to believe that BNP’s support has dropped to just 12 per cent.
In the 2008 election, BNP secured the lowest number of seats in its history, yet it still received 32.8 per cent of the vote. One would expect that number to increase now. Even if it hasn’t increased, it shouldn’t have dropped so drastically.
The claim that BNP’s popularity has collapsed due to extortion and land-grabbing is not entirely convincing. The political economy in our country—especially at the local level—is closely tied to such practices. Unfortunately, the reality is that our political culture has long depended on this kind of irregularity. A large class within society benefits from it, even though it is clearly unacceptable.
The question is: Why are such biased or agenda-driven surveys conducted? One reason is to influence public opinion. If voters can be convinced that public support for the BNP is declining, it may sway a portion of society. Another possible reason is to use the fear of declining support as leverage to extract concessions from the BNP.
We must bring change to this system. But even so, it should not result in a dramatic collapse in popularity. In our politics, there is often a strong sense of blind loyalty to political parties.
In another survey, it was shown that 31.56 per cent of people support Jamaat-e-Islami. Can this truly be the result of a credible survey?
This survey was published on 8 March 2025. It was conducted by Innovision Consulting, with support in promotion and policy coordination from the Bangladesh Research Analysis and Information Network (BRAIN) and an activist group called Voice for Reform. (Netra News, 9 March)
The Awami League has fled the country under the weight of accusations of fascism and genocide. The BNP has also lost popularity in various areas due to land-grabbing and extortion. But even then, the idea that 31 per cent of people now support Jamaat is simply not realistic.
Jamaat is a cadre-based political party. It operates under its own set of rules and structures, and it has effectively created a society within a society. The life of a Jamaat activist remains confined within the organisation’s system for a lifetime. A party like this doesn't suddenly gain or lose mass public support overnight.
In my opinion, even Jamaat itself does not believe the claim that it has 31.56 per cent public support. I believe that if Jamaat had even 15 to 20 per cent support, it would already hold a strong position in Parliament and would be willing to participate in the election without hesitation.
The survey results from Innovision Consulting do not align at all with those from BIGD regarding the public support for BNP and Jamaat. The difference between them is enormous.
In BIGD’s first survey, Jamaat’s support was recorded at 11 per cent; in the most recent one, it’s 10 per cent. That’s a difference of 20 to 21 percentage points compared to the 31.56 per cent reported by Innovision Consulting.
According to Innovision’s survey, BNP has 41.7 per cent public support, whereas in BIGD’s most recent survey, only 16 per cent expressed support for BNP.
Meanwhile, another survey by Shushashoner Jonno Nagorik (Shujan) claims that 71 per cent of people in the country support allocating seats in the upper house of Parliament based on proportional representation (PR). This result also seems unbelievable—because the country doesn’t even have an upper house yet. It might be established after the next election. Furthermore, the major party, BNP, is opposed to the PR system for an upper house. So where did this 71 per cent support come from? That’s the big question now.
Shujan head Badiul Alam Majumdar is a member of the Consensus Commission. This commission itself supports the proportional representation (PR) system. In fact, not only the PR system—survey respondents have more or less agreed with almost all of the proposals made by the Consensus Commission. These proposals are primarily backed by the newly formed party, the National Citizen Party (NCP), and Jamaat-e-Islami. Jamaat’s Nayeb-e-Ameer, Abdullah Mohammad Taher, has publicly spoken in support of the survey results.
It appears that the statements made by the NCP and Jamaat are completely aligned with the findings of these surveys. Whatever Jamaat and NCP are advocating for or intending to say seems to be reflected directly in the survey outcomes.
From a research methodology standpoint, these surveys likely cannot be dismissed as invalid—because the methodology used has not been discussed in detail here. Only selective results from each survey have been analysed. However, it seems that these surveys were either conducted with a specific agenda in mind, collecting data from a targeted group, or there were methodological flaws in the research that the researchers failed to identify.
These kinds of politically motivated surveys bring to mind the media polls conducted before the 2001 national election in Bangladesh. At that time, almost every newspaper predicted that the Awami League would return to power. Some surveys claimed the Awami League would win anywhere between 170 and 220 seats. Possibly, Prothom Alo was the only publication whose survey indicated that the BNP might come to power.
But the actual election results showed that the Awami League won only 57 seats. We also remember the US election between Hillary Clinton and Donald Trump. Before that election, many polls showed Hillary in the lead, yet she ended up losing.
The question is: Why are such biased or agenda-driven surveys conducted? One reason is to influence public opinion. If voters can be convinced that public support for the BNP is declining, it may sway a portion of society. Another possible reason is to use the fear of declining support as leverage to extract concessions from the BNP.
We don’t know for certain all the reasons behind these surveys, but we can sense the potential impact of such politically motivated research. These surveys are likely to create confusion among the public.
In the electoral field, these surveys may harm the BNP. There are already efforts underway to corner and isolate the party. These surveys could very well be part of that strategy. On social media, there is an aggressive narrative that BNP is monopolising the political landscape. Meanwhile, the biased surveys claim that this dominance is causing a drop in BNP's popularity.
Dr. Maruf Mallick is a political analyst
#The views expressed are the author’s own.