User:John Wilkins/sandbox

Geoffrey David Cumming (‘Geoff’) Born	9 Feb 1946 Melbourne Australia Spouse	Lindsay Cumming (née Brown) Nationality	Australian Alma mater	 Monash University Magdalen College, Oxford Scientific career	Experimental Psychology Statistics Influences	Cohen, Rosenthal, Rothman, Schmidt, Thomason Website	https://thenewstatistics.com/itns/ https://www.latrobe.edu.au/psychology/staff/profile?uname=GDCumming

Geoff Cumming (Pronunciation of “Cumming” needed; born 9 Feb 1946) is a Rhodes Scholar, experimental psychologist, and statistical reformer. He is best known for The New Statistics, aimed at replacing p-values and Null Hypothesis Significance Testing with point estimation and confidence intervals in those biological, medical, and social sciences that rely on NHST. The New Statistics is part of his larger program of close integration of Open Science with improved statistical analyses, particularly meta-analyses.

Contents 1. Life 2. Career a. Experimental Psychology b. Statistical Reform 3. Publications 4. See also (WIKIPEDIA Articles) 5. References 6. External Links

Life Raised in Melbourne Australia, Cumming’s BSc(Hons) was from Monash University, with an Honours thesis on psychology and statistics. In 1968 he was elected to a Rhodes Scholarship to Magdalen College, Oxford. Anne Treisman supervised his DPhil thesis on visual masking. In 1974, Cumming joined La Trobe University. He retired with a Personal Chair in 2008 to work full time on statistical reform, especially by writing a technical overview of statistical reform and an introductory textbook. He married Lindsay Brown in 1969; they have three children and seven grandchildren. Scientific Career Cumming is an experimental psychologist best known for his work on statistical reform, particularly in the social sciences. His research focused originally on visual perception, computer assisted instruction , and statistical education. Until about 1990, Cumming accepted the legitimacy and accuracy of standard statistical NHST techniques, and his early statistical education research focused on improving student understanding of NHST concepts such as statistical power and accept/reject the null hypothesis. His first major project, StatPlay, was statistical software to help students visualize these NHST concepts and concepts such as Confidence intervals, meta-analysis, and Law of Large Numbers. IS THIS RIGHT? As he developed in, he became increasingly sympathetic to claims by Paul Meehl, Jacob Cohen , and Frank Schmidt  and others that NHST badly damaging psychology because of intrinsic conceptual difficulties. Later he expanded his concerns to the social sciences more generally and to those parts of the natural science that relied on NHST, such as ecology and parts of medical research. This produced a slow but radical shift away from statistical education (how to help students and often faculty members understand NHST) to replacing NHST with Confidence Intervals. Since then, the reforms have included meta-analysis and Open Science.

Initially, like other reformers, Cumming focused in on cases where the statistical issues were straightforward, such as simple t-tests. It became increasingly clear that, for many widely used NHST techniques, such as multi-variant analysis, statisticians had not developed the analogous Confidence Interval techniques. Often, even when they had been developed, they were in obscure technical statistics journals and the psych stats software did not calculate them. Although by the early 2000s, the APA Manual called for an increased emphasis on Confidence Intervals, almost all of its examples NHST. Even psychologists who accepted the need for statistical reform could not adopt the reformer’s recommendations.

This led to technical challenges. Cumming’s approach was to systematical work through technically challenges. As part of this, Cumming’s synthesized statistical literature on solutions to technical problems before developing psychologically accessible visualizations and software to accelerate disseminate these solutions, such as the Cumming Plot and StatPlay. These developments were further supported by Cumming research into the cognition of statistical inference and the sociology of the uses of statistics in scientific practices. In addition to his role in statistical reform, Cumming also contributed to the field of experimental psychology through his research on learning and visual perception.

That last paragraph and a couple subsequent ones can cover these topic Statistical reform: 	Evaluation of attempts at statistical reform 	The cognition of statistics 	The sociology of the use of statistics 	The visual representation of statistics ideas and inference 	Improving statistical understanding  	P-replication contributions 	The New Statistics 	Teaching statistics 	The Cumming Plot 	P-replication contributions (invented p-intervals with or without known population mean 80%)  a couple hundred citations. 	Non-central t distribution contributions (in fact not a big difference, the asymetry is smallest) 	Inference by Eye for independent (only?) samples 	Psychology’s statistical practices (With Fiona Fidler) 	Psych Science quasi-endorsement

Books:
Cumming, G., & Calin-Jageman, R. (2017). Introduction to The New Statistics: Estimation, Open Science, and Beyond. New York: Routledge.

Cumming, G. (2012). Understanding the new statistics: Effect sizes, confidence intervals, and meta-analysis. New York: Routledge. Software packages:

Cumming, G. (2001-2016). ESCI, Exploratory Software for Confidence Intervals. Computer software, available for free to everyone from: www.thenewstatistics.com

Teaching Resources:
Cumming, G. (2014). The New Statistics: Why and How. Psychological Science, 25, 7-29) Commissioned by Association for Psychological Science (APS)

Confidence Intervals and The New Statistics - http://www.apa.org/education/ce/1360300.aspx

American Psychological Association Article-Based Exam: Confidence Intervals and The New Statistics

Cumming, G., & Finch, S. (2001). A primer on the understanding, use and calculation of confidence intervals that are based on central and noncentral distributions. Educational and Psychological Measurement, 61, 532-574.

Cumming, G., & Finch, S. (2005). Inference by eye: Confidence intervals, and how to read pictures of data. American Psychologist, 60, 170-180. This article is a free download from the APA Style website: http://www.apastyle.org/manual/related/cumming-and-finch.pdf

Selected Academic Articles
Cumming, G. D. (1978). Eye movements and visual perception. In E. C. Carterette & M. P. Friedman (Eds.), Handbook of Perception (Vol. 9, pp. 221-255). New York: Academic Press.

Cumming, G., & Willox, N. (1982). Supplementary computer-assisted instruction for beginning readers: Australian trials. Australian Educational Researcher, 9, 31-45.

Cumming, G. (1984). Self-selection of student groups in a group Keller course. Teaching of Psychology, 11, 181-182.

Cumming, G., & Self, J. (1990). Intelligent educational systems: Identifying and decoupling the conversational levels. Instructional Science, 19, 1-17.

Cumming, G., Thomason, N., & Zangari, M. (1995). Designing software for cognitive change: StatPlay and understanding statistics. In D. Tinsley & T. van Weert (Eds.) Proceedings of the 1995 World Conference on Computers in Education (pp. 753-765). London: Chapman and Hall.

Cumming, G., & Finch, S. (2001). A primer on the understanding, use and calculation of confidence intervals based on central and noncentral distributions. Educational and Psychological Measurement, 61, 530-572. *

Finch, S., Thomason, N., & Cumming, G. (2002). Past and future American Psychological Association guidelines for statistical practice. Theory & Psychology, 12, 825-853.

Cumming, G., Williams, J., & Fidler, F. (2004). Replication, and researchers’ understanding of confidence intervals and standard error bars. Understanding Statistics, 3, 299-311.

Fidler, F., Cumming, G., Thomason, N. & Burgman, M. (2004). Statistical reform in medicine, psychology and ecology. Journal of Socio-Economics, 33, 615-630.

Fidler, F., Thomason, N., Cumming, G., Finch, S., & Leeman, J. (2004). Editors can lead researchers to confidence intervals, but can’t make them think: Statistical reform lessons from medicine. Psychological Science, 15, 119-126.

Cumming, G., & Finch, S. (2005). Inference by eye: Confidence intervals, and how to read pictures of data. American Psychologist, 60, 170–180. [download article] *

Cumming, G., & Maillardet, R. (2006). Confidence intervals and replication: Where will the next mean fall? Psychological Methods, 11, 217-227. *

Cumming, G., Fidler, F., et al. (2007). Statistical reform in psychology: Is anything changing? Psychological Science, 18, 230-232.

Cumming, G. (2008). Replication and p intervals: p values predict the future only vaguely, but confidence intervals do much better. Perspectives on Psychological Science, 3, 286-300.

Cumming, G., & Fidler, F. (2009). Confidence intervals: Better answers to better questions. Zeitschrift für Psychologie / Journal of Psychology, 217, 15-26.

Cumming, G. (2010). p values versus confidence intervals as warrants for conclusions that results will replicate. In B. Thompson & R. Subotnik (Eds.) Methodologies for Conducting Research on Giftedness (pp. 53-69). Washington, DC: APA Books.

Selected Popular Works -	The New Statistics Blog -	Dance of the p values [2009] Introduction to Statistics Lecture     -	Significance Roulette Lecture  Part1 and Part 2   -	APS Workshop: The New Statistics: Estimation and Research Integrity -	Part 1: Confidence Intervals, NHST, and p Values Part 2: Research Integrity and the New Statistics Part 3: Effect Sizes and Confidence Intervals Part 4: The New Statistics in Action Part 5: Planning, Power and Precision Part 6: Meta-analysis and Meta-analytic thinking -	Articles in the Conversation  See Also -	Estimation Statistics -	Open Science -	“P-values”  https://en.wikipedia.org/wiki/P-value -	“Misunderstandings of p-values” “https://en.wikipedia.org/wiki/Misunderstandings_of_p-values -	Cumming Plot (forthcumming)