This material is based upon work supported by grants CAREER IOS-1150292 and BCS-1439267 from the National Science Foundation
Statistics and Experimental Design (NB257)
Introductory class for graduate students without much statistics training
Book
R.P. Runyon et al., (2000) Fundamentals of Behavioral Statistics, 9th Edition. McGraw Hill. ISBN: 9780072286410
Even if it is a bit dated, I found this to be an excellent book for an introductory statistics class. It focuses primarily on conceptual understanding of the basic statistical tests most neuroscientists will use in their graduate career and the writing emphasizes simplicity and clarity (over comprehensiveness). Affordable used versions are usually available online.
Supporting articles
Primarily “Points of View” and “Points of Significance” articles published in Nature Methods, but also includes material from other sources.
Correlation, regression, and prediction
Probability and sampling distributions
Error bars and other ways to visualize variability in your samples
Hypothesis testing and t-tests
Analysis of variance (ANOVA) and general linear model (GLM)
Non-parametric tests and resampling techniques
Scientific rigor and reproducibility :: What does the P value mean again?
Scientific rigor and reproducibility :: Power failures, p-hacking, and replicability issues
Scientific rigor and reproducibility :: Power analyses and determining sample sizes
We use G*Power to perform power analyses in class. The program can be downloaded for free and offers a detailed manual (see http://www.gpower.hhu.de/). Please note that there are of number of alternative programs to consider (but I have little experience with them). Below are articles discussing how to properly determine sample sizes.
Beyond "Intro to Stats"
Data visualization and presentation
Most useful articles (discussed in class)
POV Arrows POV Axes, ticks and grids POV Bar charts and boxplots POV Plotting symbols
POV Labels and callouts POV Power of the plane POV Data exploration POV Unentangling complex plots
POV Temporal data POV Binning high resolution data POV Heat maps POV Mapping quantitative data to color
POV Neural circuit diagrams POV Elements of visual style POV Intuitive design
POV Storytelling POV The overview figure POV Points of review (part 1) POV Points of review (part 2)
Supplementary (optional) readings
POV Avoiding color POV Color blindness POV Color coding POV Design of data figures
POV Gestalt principles (part 1) POV Gestalt principles (part 2) POV Integrating data POV Into the third dimension
POV Layout POV Multidimensional data POV Negative space POV Networks
POV Pencil and paper POV Representing the genome POV Salience to relevance POV Salience
POV Sets and intersections POV Simplify to clarify POV The design process POV Typography
POV Visualizing biological data POV Managing deep data in genome browsers POV Representing genomic structural variation
Data visualization and presentation
POS Classification and regression trees POS Classification evaluation POS Ensemble methods - bagging and random forests
POS Clustering POS Logistic regression POS Regression diagnostics
POS Regularization POS Split Plot Design POS Tabular (categorical) data
POS Model selection and overfitting POS Multiple linear regression POS Nested design
POS Machine learning - a primer POS Principal component analysis What is a hidden Markov model?
