Quality assessment of scientific manuscripts in peer review and education

We report a vignette study and a survey to investigate which study characteristics influence quality ratings academics give of articles submitted for publication, and academics and students give of students’ theses. In the vignette study, 800 respondents evaluated the quality of an abstract of studies with small or large sample sizes, showing statistically significant or non-significant results, and containing statistical reporting errors or no errors. In the survey, the same participants rated the importance of 29 manuscript characteristics related to the study’s theory, design, conduct, data analyses, and presentation for assessing either the quality of a manuscript or its publishability (article) or grade (thesis). Results showed that quality ratings were affected by sample sizes but not by statistical significance or the presence of statistical reporting errors in the rated research
vignette. These results suggest that researchers’ assessments of manuscript quality are not responsible for publication bias. Furthermore, academics and students provided highly similar ratings of the importance of different aspects relevant to quality assessment of articles and theses. These results suggest that quality criteria for scientific manuscripts are already adopted by students and are similar for submitted manuscripts and theses.