The fixation of the research community on the prestige of journals harms research quality as some researchers focus on where to publish instead of what. We examined researchers’ publication preferences using a discrete choice experiment in a cross-sectional survey of international health and medical researchers. We asked researchers to consider two hypothetical journals and decide which one they would prefer. The hypothetical journals varied in their impact factor, formatting requirements, speed of peer review, helpfulness of peer review, editor’s request to cut results and whether the paper would be useful for their next promotion. These attributes were designed using focus groups and interviews with researchers, with the aim of creating a tension between personal and societal benefit. Our survey found that the researchers’ strongest preference was for the highest impact factor and the second strongest for a moderate impact factor. The least important attribute was a preference to make changes in format and wording instead of cutting results. Some respondents were willing to cut results in exchange for a higher impact factor. Despite international efforts to reduce the importance of the impact factor, it remains a driver of researchers’ behaviour. The most prestigious journals may have the most partial evidence, as researchers are willing to trade their results for prestige.