The struggle to make transparency mainstream: initial evidence for a slow uptake of open science practices in PhD theses.

Open science (OS) practices-such as data sharing, study preregistration and transparent methods-aim to increase transparency of research. While OS practices are gaining popularity-particularly through bottom-up initiatives-their adoption rate among early career researchers remains unclear. To investigate this, we analysed dissertations from two German-speaking psychology departments with varying degree of OS implementation from 2018 to 2022. We manually coded n = 379 studies from k = 91 theses and surveyed former PhD students about perceived norms, attitudes and perceived behavioural control regarding OS practices. Our findings revealed a modest increase of OS over time but no significant difference between departments with more or less-established OS practices. Additionally, attitudes and perceived control appeared to affect OS use more than perceived norms of PhD students' surroundings. As more than a decade has passed since the replication crisis emerged, this highlights a need to intensify measures at universities to implement OS.