Amplifying research influence through the social network, open access publishing, and international collaboration: A mediation analysis on nursing studies literature
Arthur Tang,Kin-Kit Li,SeungHee Han,Kin On Kwok,Neo Tung,Wilson Tam
DOI: https://doi.org/10.1111/jnu.12827
Abstract:Introduction: Research impact and influence are commonly measured quantitatively by citation count received by research articles. Many institutes also use citation count as one of the factors in faculty performance appraisal and candidate selection of academic positions. Various strategies were recommended to amplify and accelerate research influence, particularly citation counts, by bringing research articles to a wider reach for potential readers. However, no prior empirical study was conducted to examine and valid effects of those strategies on nursing studies. This study examines and verifies the direct effects and mediation effects of some strategies, namely, the use of Twitter, international collaboration, the use of ResearchGate, and open access publishing, for amplifying the citation of research and review articles in nursing studies. Design: Cross-sectional study design. Methods: Articles published in top nursing journals in 2016 were identified in PUBMED and the citation metrics for individual articles until 2021 were extracted from Scopus. The primary outcome was the citation count of the article, while the tweet count on Twitter of the article was considered a mediator. The predictors included paper type, the total number of authors, the proportion of authors with a ResearchGate account in the article, funding support, open-accessed article, and the number of different countries stated in the authors' affiliation. A mediation analysis was conducted to examine the predictors' direct and indirect effects (i.e., via tweet count) on the citation count of the article. Results: A total of 2210 articles were included in this study, of which 223 (10.1%) were review articles. The median (IQR) number of Scopus citations, tweets, countries, and percentage of authors with ResearchGate accounts were 12 (6-21), 2 (0-6), 1 (1-1), and 75% (50%-100%) respectively. In the mediation analysis, tweet count, article type, number of countries, percentage of authors with a ResearchGate account, and journal impact factors in 2014 were positively associated with the Scopus citation count. The effects of article type, open access, and journals' impact factors in 2014 on Scopus citation count were mediated by the tweet count. Conclusion: This study provides empirical support for some strategies researchers may employ to amplify the citation count of their research articles. The methodology of our study can be extended to compare research influence between entities (e.g., across countries or institutes). Clinical relevance: The citation refers to the research work cited by peers and is one of the indicators for research impact. Higher citations implied the research work is read and used by others, therefore, understanding the associated factors with higher citations is critical.