This repository contains materials for my workshop on Bayesian Statistics in Numerical Cognition, held at MCLS 2019 in Ottawa, ON. In addition, it hosts the source files for our tutorial paper "Bayesian Inference in Numerical Cognition: A Tutorial using JASP", jointly written with Alexander Ly and E.J. Wagenmakers.
- Workshop slides (pdf)
- JASP software
- Online Bayes factor calculator for ANOVA summaries
- Tutorial paper (pdf)
- Technical supplement (pdf)
Each of the datasets below is formatted in CSV (comma separated value) format, which is the format required for JASP. You will want to download the files somewhere on your computer. Windows users should "right-click" on the CSV link. Mac users should "ctrl+click" on the CSV link. Alternatively, you can download a single ZIP archive of the files here.
Note: all datasets below are simulated and should be used for instruction only. Each dataset was generated in R and to the best extent possible reflects the structure of the data reported in the accompanying paper. The R script for generating these datasets is here.
-
t-test example (csv)
- these data are based on Verguts and De Moor (2005), who tested decomposed versus holistic processing in two-digit number comparison. The variables in this dataset include:
condition
: participants were randomly assigned to choose either the "smaller" or "larger" of each number pair.distSame
: a measure of distance effect for same-decade pairs (e.g., 52 vs. 58). Specifically, it is defined as the mean RT difference (ms) between "close" pairs (i.e., numerical distance 1 to 4) and "far" pairs (i.e., numerical distance 5-8).distDifferent
: a measure of distance effect for different-decade pairs (e.g., 68 vs. 73)
- these data are based on Verguts and De Moor (2005), who tested decomposed versus holistic processing in two-digit number comparison. The variables in this dataset include:
-
regression example (csv)
- these data are based on Holloway and Ansari (2009), who studied the relationship between symbolic and nonsymbolic distance effects and mathematical fluency. The variables in this dataset include:
symbolicDE
: a standardized measure of distance effect for symbolic number comparison, defined as (close RT - far RT)/(closeRT)nonsymbolicDE
: a standardized measure of distance effect for nonsymbolic comparison, defined as abovefluency
: a measure of math fluency from Woodcock-Johnson III (timed math facts for 3 minutes)
- these data are based on Holloway and Ansari (2009), who studied the relationship between symbolic and nonsymbolic distance effects and mathematical fluency. The variables in this dataset include:
-
factorial ANOVA example (csv)
- these data are based on a replication of Campbell and Fugelsang (2001), who found an interaction between problem size and format in mental addition verification. The variables all represent RTs (in msec) for the following experimental conditions:
small.digit
: small problems, digit formatsmall.word
: small problems, word formatlarge.digit
: large problems, digit formatlarge.word
: large problems, word format
- these data are based on a replication of Campbell and Fugelsang (2001), who found an interaction between problem size and format in mental addition verification. The variables all represent RTs (in msec) for the following experimental conditions: