Qiang Ruan

Graduation Semester and Year

2020

Language

English

Document Type

Dissertation

Degree Name

Doctor of Philosophy in Management Science

Department

Information Systems and Operations Management

First Advisor

Mary Whiteside

Abstract

Do research trends in business and statistics predict or even reflect the emergence of analytics in business practice and programs in the 21st century? My dissertation explores the answer to this question from several perspectives. The first and second essays explore knowledge sharing among statistics and business academic journals. Both essays use citation and abstract data from 24 business journals and 12 statistics journals for the years 2000, 2005, 2010, and 2015 from the Web of Science. The first essay employs multidimensional scaling with factor analysis simple structure groups to analyze both citations and abstracts. The similarity of citations among statistics and probability and other business disciplines seems to have changed somewhat from 2000 to 2015, but the changes are not substantial. Journals from different disciplines are more likely to share research interests in 2015 than previously. Analysis of the citing and cited data based on disciplines and topic modeling also are used to describe how the different disciplines either influence or are influenced by other disciplines during the emergence of business analytics. Journals in accounting and ISOM are more likely to cite other disciplines than be cited by other disciplines. On the other hand, economics, finance, and probability are more likely to be cited by other disciplines. These three disciplines are more likely to "teach" than “learn”. The citing and cited patterns are not so evident in other disciplines. The second essay applies network analysis and log-multiplicative models to reexamine communications among statistics and business journals. UCINET and LEM are the tools used in this research. In this essay, we find that the rise of business analytics seems to have promoted and increased communication among different disciplines, but the changes are not pronounced. The influence of statistics journals as storers (citing) remains low and stable for the four years, the impact of statistics journals as sources (cited) starts high and increases in the four years we selected. We also discuss the influence of other disciplines. The third paper focuses on studying the landscape of data science research from a network perspective. Like previous essays, Web of Science provides the source of data in our research: we extract publication information related to data science from 1960 to February 2020. The paper presents descriptive statistics of the most cited journals, the most cited authors, and the most productive authors. We also use exponential Random Graph Models (ERGM) to analyze the citation network regarding paper quality and the number of authors of the paper. Statistics journals are influential in data science studies as they provide fundamental background for this new rising subject. Publications with fewer keywords, more pages, and more authors, and publications with funding support are more likely to be cited by articles about the same topic. This research finds that since the emergence of business analytics, knowledge sharing between statistics and academic business disciplines has increased but not substantially. Moreover, in the landscape of data science, the journals that publish the most are not statistics journals, but the most cited authors are statisticians.

Keywords

Business analytics

Disciplines

Business | Management Information Systems

Comments

Degree granted by The University of Texas at Arlington

Download

Share

COinS