Dr Alieu SK Manjang
Since the emergence of the recent survey measuring public opinion on preferred candidates for the 2026 elections, my initial intention was to ignore it and take the results at face value. However, upon reviewing the report, several critical issues stood out which warrant closer examination. In this first part, I will highlight concerns about the reliability and internal validity of the findings. It is also necessary to ask: who owns this research, and who funded it?
This question is valid because I came across a statement denying any association between the survey and Ceprass. This is further supported by the fact that the document does not bear the name of any institution. The only organisation mentioned in the introduction is Gambia Participates. As a civil society organisation, Gambia Participates does not self-fund its activities but relies on support from UN agencies and other partners. Thus, the question remains: who actually funded this survey? In academic research, this question is critical, as it exposes vested interests, underlying intentions, and the true aims of a project. More importantly, the report indicates that 48 enumerators were trained and deployed across all 53 constituencies, supervised by experts throughout the process, an exercise that clearly required significant resources. The public has the right to know the funding source; otherwise, speculation will persist, undermining the credibility of the results.
Another major shortcoming is the lack of clarity on the data collection process. The survey does not explain how long it took the 48 enumerators to collect data from all 53 constituencies, nor the specific period during which the fieldwork took place. Although this may appear trivial, such details are important for transparency, as some citizens might recall being interviewed or observing survey activities during that time.
Reliability is also questionable. Since this is a 2026 presidential election opinion poll, one must ask: to what extent would repeating the same survey with the same respondents at a different time yield similar results? This concern is heightened by the fact that more than a year remains before the elections, during which voter preferences could shift significantly due to various political and social factors. This raises the broader question: why conduct such a poll at this early stage?
The survey’s objectives are vague and ambiguous. According to the report, the first objective was to measure public “views” on President Adama Barrow and key opposition figures. However, “views” is a generic and imprecise term. It could refer to respondents’ views on personality, performance, relations with other political parties, or other factors influencing voter behavior. Yet, the survey narrows this “view” to whether Barrow and Darboe should seek a third term, and whether respondents would vote for them if they ran. Reducing such a broad concept to a narrow question undermines internal validity.
The second stated objective was to assess leadership succession and candidate viability within major political parties, particularly the United Democratic Party (UDP). However, the survey only addressed this issue in the case of the UDP. If the researchers believed this matter was relevant exclusively to the UDP, they should have stated so clearly. Otherwise, why was the same analysis not extended to the NPP, especially given debates about President Barrow’s third-term bid and speculation about potential successors? This selective approach suggests bias and subjective judgment.
A third objective was to examine the broader electoral outlook, including coalition prospects. While this is important, the survey provides no rationale for limiting coalition scenarios to Talib Bensouda, Mamma Kandeh, and Essa Faal. Why not consider alternatives such as Darboe, Kandeh, and Faal, or Yankuba Darboe with other figures? The criteria for narrowing coalition possibilities remain unexplained, further undermining the objectivity of the survey.
The biased selection of candidates directly threatens internal validity. For example, PDOIS and Halifa Sallah were excluded, despite being a longstanding opposition force, with Halifa outperforming Essa Faal in the 2021 elections. The question of “who stands the best chance of winning in 2026” would likely have yielded different results had Halifa been included. Furthermore, including both Talib and Darboe from the same party, while limiting other parties to a single candidate, distorts voter preference comparisons. Excluding Talib, for instance, could drastically shift support in Darboe’s favor relative to Barrow.
Similarly, the rationale behind it including some figures and excluding others remains unclear. On what basis was Talib prioritised over Yankuba Darboe in coalition scenarios? Why was Lamin Sanneh excluded from UDP succession discussions? Why did the survey specifically ask whether Talib Bensouda should run for president in 2026, irrespective of party affiliation, when he is clearly affiliated with a political party? Why was this question not posed for other figures like Royeh or Yankuba?
Such inconsistencies, lack of transparency, and selective framing raise serious credibility concerns. Unless these issues are addressed, the survey risks being dismissed as biased and methodologically unsound.
In part two, I will critically examine the survey sample method to highlights issues related to external validity of the results.
