Open Access Open Access  Restricted Access Subscription Access

Developing Multiple Linear Regression Models for the Number of Citations: A Case Study of Florida International University Professors

Christina I. Guzman, B. M. Golam Kibria


Research Gate (RG) is a social networking website where experts from different fields can share their research papers, ask questions that spark a dialogue regarding research related topics, and recommend useful papers to other researchers. This paper aims to examine what factors, or regressors, influence the number of citations that a Florida International University (FIU) professor receives on Research Gate (RG). The nine regressors considered are the Total Research Interest, the number of recommendations, the number of reads, the number of research items, the number of projects, the number of questions, the number of answers, the number of followers, and the number of following associated with each professor. After fitting a number of multiple linear regression models and implementing a variety of techniques to evaluate each models’ adequacy, we concluded that the number of citations that a FIU professor has on RG is influenced by the total research interest, the number of recommendations, the number of reads, the number of research items, and the number of followers that a professor has.


Citations; Florida International University; Linear Regression Model; Model Adequacy Metrics; Research Gate; Stepwise Regression Analysis.

Full Text:



A.Abrishami; S. Aliakbary Predicting citation counts based on deep neural network learning techniques. Journal of Informetrics13(2) 485-499 (2019).

O. Alheyasat; Examination expertise sharing in academic social networks using graphs: The case of ResearchGate. Contemporary Engineering Sciences, 8(1–4) 137–151(2015).

About Research Gate.

X.Bai; Z. Fuli;L.; Predicting the citations of scholarly paper. Journal of Informetrics,13, 407–418(2019).

Florida International University Members.

C.P.Hoffmann; C. Lutz,; M. Meckel ; A relational altmetric?Network centrality on Research Gate as an indicator of scientificimpact. Journal of the Association for Information Science and Technology. doi: 10.1002/asi.23423(2015).

W. Jeng; S. DesAutelS; D. He; L. Li;Information exchange on an academic social networking site: a multidiscipline comparison on ResearchGate Q&A. Journal of the Association for InformationScience and Technology, 68(3) 638-652(2017).

A.R. Memon; Research Gate is no longer reliable: leniency towards ghost journals may decrease its impact on the scientific community. Journal of the Pakistan Medical Association, 66(12) 1643–1647(2016).

D.C. Montgomery; A. Peck; G.G.Vining; Introduction to Linear Regression Analysis - Fifth Edition, Wiley, NewYork (2012).

D.Nicholas;D. Clark; E. Herman; ResearchGate: Reputation uncovered. Learned Publishing,29(3), 173–182 (2016).

O.B. Onyancha; Social media and research: An assessment ofthe coverage of South African universities in ResearchGate, Web ofScience and the Webof metrics Ranking of World Universities. South African Journal of Libraries and Information Science, 81 (1) 8–20(2015).

S. Ovadia; Research Gate and Academic socialnetworks. Behavioral & Social Sciences Librarian, 33(3) 165–169(2014).

M. Thelwall; K. Kousha; Research Gate: Disseminating, communicating, and measuring scholarship? Journal of the Association for Information Science and Technology, 66(5) 876–889(2015).

M. Thelwall; M. K. Kousha; Research Gate Articles: Age, Discipline, Audience Size,and Impact Journal of the association for information science and technology,68(2) 468–479 2017.

R.VanNoorden; Online collaboration: Scientists and the social network. Nature, 512 (7513) 126- 129(2014).

Yu, M-C, Wu, Yen-Chun Jim;W. Alhalabi; H-Y. Kao; Wu, W-H.;ResearchGate: An effective altmetric indicator for ac tive researchers? Computers in Human Behavior, 551001-1006(2016).


  • There are currently no refbacks.



© 2015 IARS. All right reserved.