Using R for Policy Data Analysis
Description: Data analysis is an essential part of quantitative policy analysis; however, focused application of statistical methods is beyond the scope of what can be taught in classes such as Cost Benefit Analysis (CBA) and Program Evaluation. In this course, students who have completed CBA and Program Evaluation will apply a variety of data analysis techniques using R, a free open source statistics and graphical analysis environment that is increasingly used by data miners and analysts. Class sessions will include a combination of instruction on data analysis techniques, in-class application using R, and critical evaluation and discussion of published CBA and Program Evaluation cases. Applications will focus on analyses that support the execution of CBA and Program Evaluation, including cases that focus on consumer protection, affordable housing, and public health.
Learning Outcomes: • Using microdata to estimate the size of a population impacted by a policy or program. • Estimating the per-unit impact of a policy change or program implementation. • Understanding the demographics of impacted populations, including demonstrating which populations are disproportionately impacted. • Accounting for uncertainty for sensitivity analysis. • Import data into R data structures and check for missing data • Produce scatterplots, histograms, boxplots and other graphs to identify data problems and test data assumptions (e.g. normality, linearity). • Produce descriptive statistics, including mean, median, mode, proportion, standard deviation and other measures of central tendency and dispersion. • Calculate confidence intervals. Conduct t-test for differences in sample means and chi-square test to test for differences in categorical variables across groups. • Conduct bivariate correlation and simple regression analysis.