Confidence Intervals and The New Statistics
This article describes an estimation approach to research, based on Confidence Intervals. Estimation is an excellent approach to designing research and analyzing data because it’s highly informative, and usually gives the best answers to our research questions.
Part of the Online Article CE Series (OACES).
- Explain what information an effect size and a confidence interval provide.
- Explain why estimation is likely to be more informative than an analysis based on null hypothesis significance testing.
Geoff Cumming, DPhil retired in 2008 after 33 years in the School of Psychological Science, La Trobe University, Melbourne, Australia. Before that I took a doctorate in experimental psychology at Oxford. I retired to write a book about The New Stats, to be accompanied by my ESCI software. My passion is trying to persuade psychologists to improve how they do statistics. My research includes statistical cognition—the study of how people understand, or misunderstand statistical ideas. I am married to Lindsay; we have three grown up children scattered around the world, and one grandson. I enjoy giving talks about statistics, mountain biking, woodwork, and word games.
- This exam is available online and can ONLY be purchased online. Phone orders will not be accepted.
- You can not purchase two or more of the same online exam.
- A valid email address or your UserID and a password are required during the ordering process. When the order is complete, you will receive an email notification with instructions on how to retrieve