• Compared with preprints that notpublished in journals, preprints subsequently published in journals have better altmetrics. • The better the altmetrics of preprints, the higher the impact factor when the preprints published in journals. • The better the altmetrics and citations of preprints, the better performance when the preprints published in journals. • There exists consistency in academic community evaluation (peer review and citation) and public attention (altmetrics). In this paper, we study the consistency of impact of preprints and their journal versions, with the impact being measured using altmetrics and bibliometric indicators. We found that, compared with preprints that have not been published in journals, preprints subsequently published in journals performed better in altmetrics statistically, which accords with the peer review system. There was a positive correlation between the altmetrics performance of preprints and the journal impact factor when the preprint had been published in a journal. We also found a positive correlation between the performance of preprints and their corresponding journal papers, which confirms the consistency in academic community evaluation (peer review and citation) and general public attention (measured by altmetrics indicators). The impact of research is driven by its value, which could be distinguished by both the general public and the academic community. Research evaluation refers to the evaluation of individual research instead of where it is published.