Is collaboration among scientists related to the citation impact of papers because their quality increases with collaboration? An analysis based on data from F1000Prime and normalized citation scores

In recent years, the relationship of collaboration among scientists and the citation impact of papers have been frequently investigated. Most of the studies show that the two variables are closely related: an increasing collaboration activity (measured in terms of number of authors, number of affiliations, and number of countries) is associated with an increased citation impact. However, it is not clear whether the increased citation impact is based on the higher quality of papers which profit from more than one scientist giving expert input or other (citation-specific) factors. Thus, the current study addresses this question by using two comprehensive datasets with publications (in the biomedical area) including quality assessments by experts (F1000Prime member scores) and citation data for the publications. The study is based on nearly 10,000 papers. Robust regression models are used to investigate the relationship between number of authors, number of affiliations, and number of countries, respectively, and citation impact - controlling for the papers' quality (measured by F1000Prime expert ratings). The results point out that the effect of collaboration activities on impact is largely independent of the papers' quality. The citation advantage is apparently not quality-related; citation specific factors (e.g. self-citations) seem to be important here.