Back in the grimdark pre-Snapchat era of humanity (i.e. early 2011), I started teaching an introductory statistics class for psychology students offered at the University of Adelaide, using the R statistical package as the primary tool. I wrote my own lecture notes for the class, which have now expanded to the point of effectively being a book. The book is freely available, and as of version 0.6 it is released under a creative commons licence (CC BY-SA 4.0)

The book is associated with the lsr package on CRAN and GitHub. The package is probably okay for many introductory teaching purposes, but some care is required. The package does have some limitations (e.g., the etaSquared function does strange things for unbalanced ANOVA designs), and it has not been updated in a while.

There are now several variations of the original LSR book:

I have suggested that someone write a Learning Statistics with an Abacus adaptation but so far there has been little interest.

Due to the lack of demand for the abacus version, I have started work on a tidy version of the original LSR book, which will take a Tidyverse-oriented approach and rely on a new “tidylsr” package. It is early days in the revision, but if anyone has any suggestions for the new book, please feel free to open an issue on the github page

Table of Contents

I. Background

  • Chapter 1: Why do we learn statistics? Psychology and statistics. Statistics in everyday life. Some examples where intuition is misleading, and statistics is critical.
  • Chapter 2: A brief introduction to research design. Basics of psychological measurement. Reliability and validity of a measurement. Experimental and non-experimental design. Predictors versus outcomes.

II. An introduction to R

  • Chapter 3: Getting started with R. Getting R and Rstudio. Typing commands at the console. Simple calculations. Using functions. Introduction to variables. Numeric, character and logical data. Storing multiple values as a vector.
  • Chapter 4: Additional R concepts. Installing and loading packages. The workspace. Navigating the file system. More complicated data structures: factors, data frames, lists and formulas. A brief discussion of generic functions.

III. Working with data

  • Chapter 5: Descriptive statistics. Mean, median and mode. Range, interquartile range and standard deviations. Skew and kurtosis. Standard scores. Correlations. Tools for computing these things in R. Brief comments missing data.
  • Chapter 6: Drawing graphs. Discussion of R graphics. Histograms. Stem and leaf plots. Boxplots. Scatterplots. Bar graphs.
  • Chapter 7: Pragmatic matters. Tabulating data. Transforming a variable. Subsetting vectors and data frames. Sorting, transposing and merging data. Reshaping a data frame. Basics of text processing. Reading unusual data files. Basics of variable coercion. Even more data structures. Other miscellaneous topics, including floating point arithmetic.
  • Chapter 8: Basic programming. Scripts. Loops. Conditionals. Writing functions. Implicit loops.

IV. Statistical theory

  • Prelude. The riddle of induction, and why statisticians make assumptions.
  • Chapter 9: Introduction to probability. Probability versus statistics. Basics of probability theory. Common distributions: normal, binomial, t, chi-square, F. Bayesian versus frequentist probability.
  • Chapter 10: Estimating unknown quantities from a sample. Sampling from populations. Estimating population means and standard deviations. Sampling distributions. The central limit theorem. Confidence intervals.
  • Chapter 11: Hypothesis testing. Research hypotheses versus statistical hypotheses. Null versus alternative hypotheses. Type I and Type II errors. Sampling distributions for test statistics. Hypothesis testing as decision making. p-values. Reporting the results of a test. Effect size and power. Controversies and traps in hypothesis testing.

V. Statistical tools

  • Chapter 12: Categorical data analysis. Chi-square goodness of fit test. Chi-square test of independence. Yate’s continuity correction. Effect size with Cramer’s V. Assumptions of the tests. Other tests: Fisher exact test and McNemar’s test.
  • Chapter 13: Comparing two means. One sample z-test. One sample t-test. Student’s independent sample t-test. Welch’s independent samples t-test. Paired sample t-test. Effect size with Cohen’s d. Checking the normality assumption. Wilcoxon tests for non-normal data.
  • Chapter 14: Comparing several means (one-way ANOVA). Introduction to one-way ANOVA. Doing it in R. Effect size with eta-squared. Simple corrections for multiple comparisons (post hoc tests). Assumptions of one-way ANOVA. Checking homogeneity of variance using Levene tests. Avoiding the homogeneity of variance assumption. Checking and avoiding the normality assumption. Relationship between ANOVA and t-tests.
  • Chapter 15: Linear regression. Introduction to regression. Estimation by least squares. Multiple regression models. Measuring the fit of a regression model. Hypothesis tests for regression models. Standardised regression coefficient. Assumptions of regression models. Basic regression diagnostics. Model selection methods for regression.
  • Chapter 16: Factorial ANOVA. Factorial ANOVA without interactions. Factorial ANOVA with interactions. Effect sizes, estimated marginal means, confidence intervals for effects. Assumption checking. F-tests as model selection. Interpreting ANOVA as a linear model. Specifying contrasts. Post hoc testing via Tukey’s HSD. Factorial ANOVA with unbalanced data (Type I, III and III sums of squares)

VI. Other topics

  • Chapter 17: Bayesian statistics. Introduction to Bayesian inference. Bayesian analysis of contingency tables. Bayesian t-tests, ANOVAs and regressions.
  • Chapter 18: Epilogue. Comments on the content missing from this book. Advantages to using R.
  • References. An incomplete reference list.

—-             —-