Skip to content
/ ARR Public
forked from QuantLet/ARR

Academic Rankings Research

Notifications You must be signed in to change notification settings

ningboyuan/ARR

 
 

Repository files navigation

Visit QuantNet

Visit QuantNet ARR - Academic Rankings Research Visit QuantNet 2.0

Authors: Alona Zharova, Andrija Mihoci and Wolfgang Karl Härdle

Title: Academic Ranking Scales in Economics: Prediction and Imputation

Published in: SFB 649 Discussion Paper 2016-020

JEL classification: C14, C53, C81, M10

Available at: http://sfb649.wiwi.hu-berlin.de/papers/pdf/SFB649DP2016-020.pdf

Abstract

Publications are a vital element of any scientist’s career. It is not only the number of media outlets but aslo the quality of published research that enters decisions on jobs, salary, tenure, etc. Academic ranking scales in economics and other disciplines are, therefore, widely used in classification, judgment and scientific depth of individual research. These ranking systems are competing, allow for different disciplinary gravity and sometimes give orthogonal results. Here a statistical analysis of the interconnection between Handelsblatt (HB), Research Papers in Economics (RePEc, here RP) and Google Scholar (GS) systems is presented. Quantile regression allows us to successfully predict missing ranking data and to obtain a so-called HB Common Score and to carry out a cross-rankings analysis. Based on the merged ranking data from different data providers, we discuss the ranking systems dependence, analyze the age effect and study the relationship between the research expertise areas and the ranking performance.

Acknowledgements

Financial support from the German Research Foundation (DFG) via Collaborative Research Center 649 ”Economic Risk” and International Research Training Group 1792 ”High Dimensional Non Stationary Time Series”, Humboldt-Universität zu Berlin, is gratefully acknowledged. We are thankful for the research assistance provided by Marius Sterling.

About

Academic Rankings Research

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • R 56.8%
  • MATLAB 43.2%