The Learning by Example course will attempt to provide you everything you need to get your own analyses up and running in R. This five-day course will cover many of the models for the analyses that most common to the social sciences. Here are the topics we will cover throughout the week.
- Preliminaries and Foundations
- Data Visualization with
- Descriptive Analysis and Measures of Association
- Linear Models
- Non-linear relationships and effects plots
- GLMs (mostly focusing on logit)
- Non-linear relationships
- Effects Plots
- Ordered/Multinomial Logit
- Parallel Regressions Assumption
- Effects plots
- Conditional logit with choice-specific characteristics As time and interest permit, we will also discuss:
- Complex Surveys
- Multilevel Models
- Clustered standard errors
- Random intercept models
- Random coefficient models
- Effect plots
- Bayesian model estimation with
- Panel/Longitudinal/TSCS models
- Time-series properties of data
- Fixed and Random effect models
- Between and within unit effects
- Using the
plmpackage to estimate TSCS models
- Specification and error tests for models.
- Measurement 1. Summated rating model and reliability analysis 2. Exploratory factor analysis and principal components analysis 3. Confirmatory factor analysis and structural equation models
To walk us through all of these topics, I’ve written a book - R: Learning by Example. It’s in very early draft format, but it covers the ground we need to cover now. You can find the book here.
I have developed a cheat sheet of sorts to help remind participants in the course what we talked about.