Does the use of unusual combinations of datasets contribute to greater scientific impact?

Significance Combined datasets in social science research, particularly those that use unusual or newly linked datasets, are significantly related to both scientific and public impact. This positive effect is observed across varying team sizes and experience levels, and its influence surpasses that of traditional paper novelty measures. Despite its relationship with impact, novel combinations of datasets are rarely found, even in this era of data-driven and multidisciplinary studies that encourage resource amalgamation from previously disconnected sources. Our findings underscore the potential of datasets combined unconventionally to elevate the impact of scientific discoveries. They also provide valuable insights for researchers, policymakers, and data curators.