API-201 Learning Objectives
- Create effective visualizations (histograms, scatterplots, bar charts) that tell a clear story
- Calculate mean, median, and mode for a dataset and explain when each is most appropriate
- Compute variance, standard deviation, and percentiles to characterize data spread
- Identify when summary statistics are misleading (e.g., using mean with skewed data or outliers)
- Look beyond central tendency to understand full distributions using percentiles
- Choose the most appropriate statistic or analysis for a specific policy question
- Recognize when metrics or statistics might mislead rather than inform
- Calculate marginal, conditional, and joint probabilities using probability tables
- Apply Bayes' Rule to update beliefs when new information becomes available
- Use probability distributions (discrete and continuous, especially Normal) to model uncertainty
- Structure complex decisions as decision trees with choices, uncertainties, probabilities, and payoffs
- Avoid common probability errors (e.g., confusing P(A|B) with P(B|A), base rate neglect)
- Apply probability thinking to real-world policy problems (medical testing, risk assessment)
- Apply Bayesian thinking: update beliefs appropriately based on new evidence and prior knowledge
- Conduct sensitivity analysis to determine which assumptions matter most
- Explain technical statistical concepts to non-technical audiences without jargon
- Explain why estimates vary from sample to sample (sampling fluctuation)
- Calculate standard errors to quantify uncertainty in estimates
- Recognize and work with three different distributions: population, sample, and sampling distribution
- Construct confidence intervals for means, proportions, and differences between groups
- Interpret confidence intervals and p-values in the context of real studies
- Formulate appropriate null and alternative hypotheses for policy questions
- Compute p-values using two methods: direct calculation and confidence interval approach
- Conduct hypothesis tests and interpret results correctly
- Distinguish between Type I errors (false positives) and Type II errors (false negatives)
- Understand the relationship between statistical power and study design
- Recognize when "failing to reject" the null hypothesis may reflect insufficient power rather than true absence of effect
- Critically assess whether statistically significant results are practically significant
- Read tables of statistical results from published research and extract key findings
- Assess whether study designs support causal claims or only correlational findings
- Determine what a study can and cannot tell us based on its design and results
- Identify threats to generalizability (can these findings apply to other contexts?)
- Synthesize findings across multiple studies with conflicting results
- Manipulate and clean datasets in Excel for analysis
- Use pivot tables to efficiently calculate group statistics and test differences
- Implement statistical formulas (STDEV, NORMDIST, confidence intervals, hypothesis tests)
- Create clear, effective visualizations from data
- Conduct complete statistical analyses without becoming Excel experts