Here’s a brief bio:
      Geoff Cumming is Professor Emeritus at La Trobe University, Melbourne, and author of Understanding The New Statistics: Effect Sizes, Confidence Intervals, and Meta-Analysis (Routledge, 2012), and, with co-author Robert Calin-Jageman, of Introduction to The New Statistics: Estimation, Open Science, and Beyond (Routledge, 2017).  He has taught statistics for more than 40 years, and his statistics tutorial articles (tiny.cc/errorbars101 and tiny.cc/tnswhyhow) have been downloaded more than 370,000 times. The Association for Psychological Science published six videos of his highly successful workshop on the new statistics (tiny.cc/apsworkshop). His main research interests are the investigation of statistical understanding, and promotion of Open Science and improved statistical practices. A Rhodes Scholar, he received his Doctorate degree in experimental psychology from Oxford University. More information at tiny.cc/geofflatrobe and at www.thenewstatistics.com

Now for some background:
      I hope my books will change the world.
      Null hypothesis significance testing (NHST), the traditional way to analyze data in psychology and many other disciplines, is a terrible way to draw conclusions from data! The replicability crisis now adds further reasons—and urgency—for us to do better. I believe that currently the best way forward is to move to estimation, meaning effect sizes and confidence intervals. I refer to estimation, together with meta-analysis, as “the new statistics”—the techniques themselves are not new, but adopting them as our main way of drawing conclusions from data would, for most of us, be very new and also a major step forward.
       My first book takes an estimation approach, after explaining why that's much better than using statistical significance testing. It includes three chapters on meta-analysis.
      Cumming, G. (2012). Understanding The New Statistics: Effect Sizes, Confidence Intervals, and Meta-Analysis. New York: Routledge.

Happily, some things are changing.
       Psychological Science, probably the world’s top empirical journal in psychology, introduced drastically revised author submission guidelines from 1 January 2014, including the statement:
“Psychological Science recommends the use of the ‘new statistics’—effect sizes, confidence intervals, and meta-analysis—to avoid problems associated with null-hypothesis significance testing (NHST). Authors are encouraged to consult this Psychological Science tutorial by Geoff Cumming, which shows why estimation and meta-analysis are more informative than NHST and how they foster development of a cumulative, quantitative discipline.”
      The new guidelines are at http://tiny.cc/pssubguide There is an explanation at http://tiny.cc/apseichinterview
      Eric Eich, the journal's then editor-in-chief, commissioned me to write this tutorial article to support the changes: http://tiny.cc/tnswhyhow
       Cumming, G. (2014). The New Statistics: Why and How. Psychological Science, 25, 7-29. Since publication, this article has every month been in the top handful of the most-read articles in this journal.
       The second book is an introductory textbook, which assumes no previous statistics knowledge. It takes an estimation approach from the start, and also introduces and explains Open Science—the new techniques needed to increase the replicability of research. Later it explains statistical significance testing, and cautions about its problems. This is the first introductory statistics text that presents the new statistics and Open Science, and does so all through.
Cumming, G., & Calin-Jageman, R. (2017). Introduction to The New Statistics: Estimation, Open Science, and Beyond. New York: Routledge.
      It was released in September 2016.
      My software is ESCI ("ESS-key", Exploratory Software for Confidence Intervals), which runs under Microsoft Excel. I have developed ESCI over the last 20 years. It offers dynamic interactive simulations to help users build good intuitions about fundamental statistical concepts. It also provides graphing and some data analysis capabilities to support use of the new statistics. There is a version of ESCI for each of the two books. ESCI is a free download.
      For more than a decade before I retired main research interest was statistical cognition, which is the study of how people understand—or misunderstand—statistical concepts, and various different ways to present the results of statistical analyses. Statistical cognition research can provide the evidence needed for the evidence-based practice of statistics, meaning that our selection of a statistical technique should be supported by cognitive evidence that people understand it well.
      I am especially interested in replication, which is a key requirement of Open Science. One of many reasons that CIs are better than p values is that CIs generally give quite good information about what is likely to happen on replication of an experiment, whereas a p value gives almost no information about replication. The dance of the p values illustrates how p values vary enormously with replication, thus indicating how terribly uninformative they are. See tiny.cc/dancepvals
      I may be interested in visits to interesting labs in interesting places. I can offer research talks and various statistics workshops on the new statistics, Open Science, and the teaching of statistics.
Education
BSc (Honours), Monash University, Melbourne, 1968
DPhil, Oxford University, 1971
Areas of Research / Professional Expertise
Research methods and statistics
Cognitive psychology
Statistical cognition
Teaching of statistics
Personal Interests
Spending time with family--especially our seven grandchildren--travel, bike riding, building and house renovation, woodwork, and word games.