Abstract
Background
While substantial attention has been paid to research misbehaviors, responsible research practices (RRPs) and their potential explanatory factors have not been studied extensively.
Methods
The National Survey on Research Integrity (NSRI) is an online survey targeting all disciplines and academic ranks in The Netherlands. Data was collected on 11 RRPs and 12 explanatory factor scales. Results were controlled for explanatory factor scales, academic rank, disciplinary field, gender, doing empirical research and if respondents belonged to a NSRI supporting institution or not.
Results
6,813 respondents completed the survey. The RRPs with the highest prevalence were avoiding plagiarism (99%), disclosing conflicts of interest (96.5%) and checking for errors before publication (94.3%). Preregistration of study protocols (42.8%), making accessible underlying data and syntaxes (47.2%), and keeping comprehensive research records (56.3%) had the lowest prevalence. Arts and humanities scholars, PhD candidates and junior researchers were associated with a lower RRP mean (-0.51 and -0.31 respectively) as was publication pressure (-0.05; 95% CI -0.08, -0.02). Mentoring (0.15; 95% CI 0.12, 0.17), scientific norm subscription (0.13; 95% CI 0.1, 0.15) and funding pressure (0.13; 95% CI0.10, 0.17) were significantly associated with a higher RRP mean.
Conclusions
We found publication pressure to affect RRPs negatively. Mentoring, scientific norm subscription and funding pressure may help foster RRPs. Arts and humanities scholars, PhD candidates and junior researchers need more efforts to raise awareness on RRPs. Further research on these groups is warranted in order to understand research integrity challenges that may be unique to them.
While substantial attention has been paid to research misbehaviors, responsible research practices (RRPs) and their potential explanatory factors have not been studied extensively.
Methods
The National Survey on Research Integrity (NSRI) is an online survey targeting all disciplines and academic ranks in The Netherlands. Data was collected on 11 RRPs and 12 explanatory factor scales. Results were controlled for explanatory factor scales, academic rank, disciplinary field, gender, doing empirical research and if respondents belonged to a NSRI supporting institution or not.
Results
6,813 respondents completed the survey. The RRPs with the highest prevalence were avoiding plagiarism (99%), disclosing conflicts of interest (96.5%) and checking for errors before publication (94.3%). Preregistration of study protocols (42.8%), making accessible underlying data and syntaxes (47.2%), and keeping comprehensive research records (56.3%) had the lowest prevalence. Arts and humanities scholars, PhD candidates and junior researchers were associated with a lower RRP mean (-0.51 and -0.31 respectively) as was publication pressure (-0.05; 95% CI -0.08, -0.02). Mentoring (0.15; 95% CI 0.12, 0.17), scientific norm subscription (0.13; 95% CI 0.1, 0.15) and funding pressure (0.13; 95% CI0.10, 0.17) were significantly associated with a higher RRP mean.
Conclusions
We found publication pressure to affect RRPs negatively. Mentoring, scientific norm subscription and funding pressure may help foster RRPs. Arts and humanities scholars, PhD candidates and junior researchers need more efforts to raise awareness on RRPs. Further research on these groups is warranted in order to understand research integrity challenges that may be unique to them.
| Original language | English |
|---|---|
| Publisher | MetaArXiv |
| Pages | 1-27 |
| DOIs | |
| Publication status | Published - 6 Jul 2021 |