Trust in scientists is linked to the feeling that they are on our side

Is knowing that someone has expertise enough to trust them? A survey by the Policy Institute at King's College London and the organization Sense about Science investigated how trust in scientists relates to perceived competence, transparency, and the feeling that research benefits people like the respondent. The report, funded by Wellcome and released September 9, examines the relationship between science and social belonging.

The team surveyed 4,262 adults aged 18 and older from May 8 to 11, 2026, through YouGov's online panel. Participants rated their trust in different groups of scientists on a scale from zero to ten. The report classifies scores of seven to ten as high trust, four to six as intermediate, and zero to three as low; don't-know responses are kept separate. The comparisons describe reported opinions, not tests of scientific knowledge.

Overall, 63% expressed high trust in scientists in general and 5% low trust. The share with high trust was 61% for university researchers, 42% for scientists working in government, and 34% for those in private companies. These categories were rated separately: the percentages do not divide scientists into mutually exclusive groups or measure the actual quality of each institution's work.

Perceived closeness clearly separated the respondent groups. Among the 2,817 participants classified as high-trust, 68% agreed that scientists were on the side of people like them. Agreement was 22% among the 1,107 with intermediate trust and 7% among the 206 with low trust. The report's percentages are weighted to represent the population and should not be recalculated as simple fractions of these counts. Transparency showed a similar contrast: 73% of the high-trust group perceived scientists as open and transparent, compared with 28% of the intermediate group and 10% of the low-trust group.

Political identification also accompanied differences in perception. Among those intending to vote for Reform UK, 32% saw scientists as being on their side, compared with 71% among those intending to vote Green—a gap of 39 percentage points. The survey included 778 and 626 participants in these two groups, respectively. Chris Simms's September 11 report in Nature highlights the interpretation of Bobby Duffy, one of the study's leaders: trust involves perceptions of whom scientists work for. Because responses were collected during the same period, the association does not establish whether political identity shaped trust, whether the reverse occurred, or whether other factors influenced both.

For scientific institutions, the result suggests that communicating credentials does not exhaust the task of demonstrating trustworthiness. Explaining interests, making procedures transparent, and clarifying who may benefit from research are approaches consistent with the observed perceptions. The survey did not test the effectiveness of those interventions. The next question is which practices build lasting relationships of trust without demanding political agreement or replacing critical examination of evidence with loyalty to a group.