A blog on statistics, methods, philosophy of science, and open science. Understanding 20% of statistics will improve 80% of your inferences.

Thursday, October 6, 2016

Improving Your Statistical Inferences Coursera course



I’m really excited to be able to announce my “Improving Your Statistical Inferences” Coursera course. It’s a free massive open online course (MOOC) consisting of 22 videos, 10 assignments, 7 weekly exams, and a final exam. All course materials are freely available, and you can start whenever you want.

In this course, I try to teach all the stuff I wish I had learned when I was a student. It includes the basics (e.g., how to interpret p-values, what likelihoods and Bayesian statistics are, how to control error rates or calculate effect sizes) to what I think should also be the basics (e.g., equivalence testing, the positive predictive value, sequential analyses, p-curve analysis, open science). The hands on assignments will make sure you don’t just hear about these things, but know how to use them.

My hope is that busy scholars who want to learn about these things now have a convenient and efficient way to do so. I’ve taught many workshops, but there is only so much you can teach in one or two days. Moreover, most senior researchers don’t even have a single day to spare for education. When I teach PhD students about new methods, their supervisors often respond by saying ‘I've never heard of that, I don't think we need it’. It would be great if everyone has the time to watch some of my videos while doing the ironing, chopping vegetables, or doing the dishes (these are the times I myself watch Coursera videos), and see the need to change some research practices.

This content was tried out and developed over the last 4 years in lectures and workshops for hundreds of graduate students around the world – thank you all for your questions and feedback! Recording these videos was made possible by a grant from by the 4TU Centre for Engineering Education at the recording studio of the TU Eindhoven (if you need a great person to edit your videos, contact Tove Elfferich). The assignments were tested by Moritz Körber, Jill Jacobson, Hanne Melgård Watkins, and around 50 beta-testers who tried out the course in the last few weeks (special shout-out to Lilian Jans-Beken, the first person to complete the entire course!). I really enjoy seeing the positive feedback


Tim van der Zee helped with creating exam questions, and Hanne Duisterwinkel at the TU Eindhoven helped with all formalities. Thanks so much to all of you for your help.

This course is brand new – if you follow it, feel free to send feedback and suggestions for improvement.

I hope you enjoy the course.

11 comments:

  1. I am enjoying the course very much. I have some confusion about the classification of statistics in terms of Neyman-Pearson, Bayesian and likelihoods. Is this classification standard?

    I thought that Neyman-Pearson was the same approach that likelihoods. Given a null hypothesis test, I use a statistic that often corresponds to the statistic arising from a likelihood ration test (for example the t-test for the mean).

    ReplyDelete
    Replies
    1. The distinction is commonly used - see the references in the course to Zoltan Dienes' book for example. There is a difference between a likelihood ratio (see lecture 2.1) and maximum likelihood estimation. The 3 approaches are also discussed in Royall's book on Likelihoods.

      Delete
    2. Thanks! I missed the references in the course.

      Delete
  2. Great course. I'm thinking the systematic problems why many misunderstandings about p value have overwhelmed the researchers' minds for decades. One problem is the over-dominance of frequentist approach in the research and education. Look at the classical textbook, Roger Kirk's "Experimental Design", it collects the classical works of the frequentist approach and has inspired the generations of researchers. However, the cautions in this book have yet explicitly influence its readers make serious decision in their study. This is the time to adjust the way to teach the behavioral scientists use the statistical concepts and tools.

    Is there any idea to attract the non-English Coursera users join this course? I'm glad to help translate the materials to Chinese.

    ReplyDelete
    Replies
    1. That would be totally awesome! E-mail me, and we can work something out.

      Delete
  3. Daniel, what a great course. Thank you so much for taking the time to create it. I am a PhD student and have studied stats for years. I'm only up to Week 4 of the course and I have already recognised many fundamental misunderstandings I have gained through my university courses! Not only that, but this course is waaaaaay more interesting than the frequentist stats courses I've been forced to do. Again, thank you :) Keep up the amazing work.

    ReplyDelete
    Replies
    1. Thanks for the positive feedback! Glad you are enjoying it!

      Delete
  4. Hi, I am having trouble finding the forum/board for this course. I am just running into some errors running the script for the first assignment and feel it is something I could self-diagnose with other students instead of having to bother you with it. I have some experience with R, but trouble deciphering error messages. Thank you so much for this course.

    ReplyDelete
    Replies
    1. You enrolled in a course that starts in the future? Then the forum is not open yet.

      Delete
  5. Wow, thank you so much for your quick response. I was able to get the correct result by installing an updated version of R Studio. I actually was not able to fully install it, but it seems to run fine from the disk mounted image, and when I ran the script in this I was able to do the simulations. I still don't fully understand everything, but am getting a better feel/familiarity with the concepts. Thank you so much for offering this class.

    ReplyDelete