Premium
Does research with statistics have more impact? The citation rank advantage of structural equation modeling
Author(s) -
Thelwall Mike,
Wilson Paul
Publication year - 2016
Publication title -
journal of the association for information science and technology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.903
H-Index - 145
eISSN - 2330-1643
pISSN - 2330-1635
DOI - 10.1002/asi.23474
Subject(s) - citation , structural equation modeling , computer science , rank (graph theory) , field (mathematics) , statistics , citation analysis , contrast (vision) , data science , simple (philosophy) , econometrics , mathematics , machine learning , artificial intelligence , epistemology , library science , philosophy , combinatorics , pure mathematics
Statistics are essential to many areas of research and individual statistical techniques may change the ways in which problems are addressed as well as the types of problems that can be tackled. Hence, specific techniques may tend to generate high‐impact findings within science. This article estimates the citation advantage of a technique by calculating the average citation rank of articles using it in the issue of the journal in which they were published. Applied to structural equation modeling ( SEM ) and four related techniques in 3 broad fields, the results show citation advantages that vary by technique and broad field. For example, SEM seems to be more influential in all broad fields than the 4 simpler methods, with one exception, and hence seems to be particularly worth adding to statistical curricula. In contrast, P earson correlation apparently has the highest average impact in medicine but the least in psychology. In conclusion, the results suggest that the importance of a statistical technique may vary by discipline and that even simple techniques can help to generate high‐impact research in some contexts.