Making the Most Of Statistical Analyses: Improving Interpretation and Presentation

作者: Michael Tomz , Jason Wittenberg , Gary King

DOI:

关键词: PresentationManagement scienceStatistical analysesDegree of certaintyStatistical assumptionComputer scienceStatistical simulationData scienceSoftwareReplicateInterpretation (philosophy)

摘要: Social Scientists rarely take full advantage of the information available in their statistical results. As a consequence, they miss opportunities to present quantities that are greatest substantive interest for research and express appropriate degree certainty about these quantities. In this article, we offer an approach, built on technique simulation, extract currently overlooked from any method interpret it reader-friendly manner. Using requires some expertise, which try provide herein, but its application should make results quantitative articles more informative transparent. To illustrate our recommendations, replicate several published works, showing each case how authors' own conclusions can be expressed sharply informatively, and, without changing data or assumptions, approach reveals important new questions at hand. We also very easy-to-use Clarify software implements suggestions.

参考文章(25)
Jorge I. Domínguez, James A. McCann, Democratizing Mexico: Public Opinion and Electoral Choices ,(1996)
Steven Stern, Simulation-based estimation Journal of Economic Literature. ,vol. 35, pp. 2006- 2039 ,(1997)
Robert E. Kass, Bradley P. Carlin, Andrew Gelman, Radford M. Neal, Markov Chain Monte Carlo in Practice: A Roundtable Discussion The American Statistician. ,vol. 52, pp. 93- 100 ,(1998) , 10.1080/00031305.1998.10480547
Eric W. Noreen, E. A. Hollas, Computer-intensive methods for testing hypotheses ,(1989)
Steven J. Rosenstone, John Mark Hansen, Mobilization, participation, and democracy in America ,(1993)