In celebration of the present giving season, and because I am procrastinating reading final proofs of our new book, I bring you the Twelve Days of Christmas in statistical style!
Maurits and I started discussing statistical methods in HCI with our paper for CHI 2012 "Rethinking Statistical Analysis Methods for CHI". Several years later, we are in the final stages of editing a book on the topic which draws together contributions from an amazing set of generous and gifted authors from within and beyond the HCI community. I'll post some concepts from the book in a series of short articles, one for each day of the season. (Yes, I know the twelve days start from Christmas and work to 6th Jan but who in their right mind would read a post about statistical methods on Christmas day?) The selection of topics is mostly related to whether I could torture the syllables into the song, so I wouldn't call it a balanced treatment of the book. (There is a bias against the most advanced topics because for some reason there are more syllables in the names of the trickier techniques. Show offs...)
On the first day of Christmas, my true love sent to me: fair stats communications.
In one of the last chapters in the book, Pierre Dragicevic argues that researchers should aim for fair statistical communication when presenting results: "While we cannot be fully objective when writing a study report, we can give our readers the freedom to decide whether or not they should trust our interpretations... this is the essence of fair statistical communication. " He describes the principles of fair statistical communications as:
Don't get hung up on whether you have followed the magical Null Hypothesis Significance Testing traditions which have evolved in your field. Let Pierre persuade you to consider using an estimation approach, following his comprehensive series of tips to help to make your statistical communication fair.
Modern Statistical Methods in Human Computer Interaction (edited by Judy Robertson and Maurits Kaptein) will be published by Springer in early 2016. Here is the table of contents:
Modern Statistical Methods for HCI
1. An introduction to Modern Statistical Methods for HCI.
J. Robertson & M.C. Kaptein
Section 1: Getting Started With Data Analysis.
2. Getting started with [R]: a brief introduction
3. Descriptive statistics, Graphs, and Visualization.
J. Young & J. Wessnitzer
4. Handling missing data
T. Baguley & M. Andrews
Section 2: Classical Null Hypothesis Significance Testing done properly
5. Effect sizes and Power in HCI
6. Using R for repeated and time-series observations
D. Fry & K. Wazny
7. Non-parametric Statistics in Human-Computer Interaction
J.O. Wobbrock and M. Kay
Section 3: Bayesian Inference
8. Bayesian Inference
9. Bayesian Testing of Constrained Hypothesis
Section 4: Advanced modeling in HCI
10. Latent Variable Models
A. Beaujean & G. Morgan
11. Using Generalized Linear (Mixed) Models in HCI
12. Mixture models: Latent profile and latent class analysis
Section 5: Improving statistical practice in HCI
13. Fair Statistical Communication in HCI
14. Improving statistical practice in HCI
J. Robertson & M.C. Kaptein
Online supplementary materials