MODULE7_STUDENT_COMPLETION_SYNTHETIC
Binary Variables
Create and audit binary indicators from transparent rules.
Module 7
Qualitative Information and Binary Variables
Learn how to use qualitative information in regression with binary variables, categorical indicators, interactions, group-specific slopes, treatment evaluation, and linear probability models.
20 lessons9 to 13 hoursIntermediate
Skills
Distinguish quantitative variables from qualitative information that must be coded.Create binary variables and explain the omitted base group.Interpret intercept shifts, group mean differences, and adjusted group gaps.Avoid the dummy-variable trap when modeling multiple categories.Convert log-dependent dummy coefficients with exact percentage effects.Interpret binary-by-binary and binary-by-continuous interactions.Center continuous variables before interactions when a clearer reference point is needed.Test intercept, slope, and full-regression differences across groups.Estimate and interpret linear probability models with robust caution.Separate treatment comparisons from causal claims when self-selection is possible.
Progress tracking
Sign in to track your progress
Public visitors can preview the module, but personal progress is saved only after sign-in.
Module resources
20 quiz checks18 datasets12 notebooks
Lessons
1Qualitative Information in EconometricsHow do we use information like gender, region, industry, or program status in regression?Open lesson2Creating Binary VariablesHow can a yes/no condition become a regression variable?Open lesson3One Binary Independent VariableWhat does one dummy variable do in a regression line?Open lesson4Comparing Two Means with RegressionWhy does a dummy-only regression reproduce a difference in means?Open lesson5Controls and Conditional Group DifferencesWhy can an adjusted group gap differ from a raw difference in means?Open lesson6Dummy Variables in Log-Dependent ModelsHow do we interpret a dummy coefficient when the dependent variable is logged?Open lesson7Base Group and Dummy-Variable TrapWhy can we not include every category indicator with an intercept?Open lesson8Multiple CategoriesHow do we interpret a set of category dummies?Open lesson9Comparing Non-Base CategoriesHow can we compare two included categories when neither is the base group?Open lesson10Ordinal Information with IndicatorsWhen should ordered categories be coded as numbers or as indicators?Open lesson11Binary by Binary InteractionsWhat does a product of two dummy variables add to the model?Open lesson12Binary by Continuous InteractionsHow can a slope differ across groups?Open lesson13Centering Before Dummy InteractionsWhy does centering make dummy interactions easier to read?Open lesson14Testing Intercept and Slope DifferencesHow do we test whether groups differ in more than one way?Open lesson15Full Regression Differences Across GroupsWhat does a Chow-style comparison ask?Open lesson16Binary Dependent VariablesWhat changes when the outcome is zero or one?Open lesson17Linear Probability ModelHow can OLS model a probability?Open lesson18Limitations of the Linear Probability ModelWhy should LPM results be checked carefully?Open lesson19Policy Evaluation and Self-SelectionWhy is a treatment dummy not automatically causal?Open lesson20Module 7 CapstoneHow do we build a complete qualitative-variable analysis?Open lesson
Notebook labs
MODULE7_WAGE_GROUPS_SYNTHETIC
Group Mean Comparisons
Show how a dummy-only regression reproduces a two-group mean difference.
MODULE7_WAGE_GROUPS_SYNTHETIC
Dummy Variables in Log Models
Convert log-dummy coefficients using approximate and exact percentages.
MODULE7_CATEGORY_EFFECTS_SYNTHETIC
Multiple Categories
Encode categories with a base group and compare coefficient meanings.
MODULE7_WAGE_GROUPS_SYNTHETIC
Dummy Interactions
Interpret binary-by-binary interactions as conditional group differences.
MODULE7_WAGE_GROUPS_SYNTHETIC
Different Slopes and Centering
Estimate group-specific slopes and show how centering changes the reference point.
MODULE7_WAGE_GROUPS_SYNTHETIC
Chow-Style Group Difference Tests
Use interaction restrictions to test full group differences.
MODULE7_STUDENT_COMPLETION_SYNTHETIC
Linear Probability Model
Estimate an LPM and interpret coefficients as probability-point changes.
MODULE7_LOAN_APPROVAL_SYNTHETIC
LPM Robust Standard Errors
Inspect fitted probabilities and use HC1 robust standard errors.
MODULE7_PROGRAM_EVALUATION_SYNTHETIC
Policy Evaluation and Self-Selection
Compare randomized and self-selected treatment comparisons.
MODULE7_DISCRETE_OUTCOME_SYNTHETIC
Discrete Dependent Variables
Describe count outcomes and explain why discrete models may be needed later.
MODULE7_STUDENT_COMPLETION_SYNTHETIC
Module 7 Capstone
Complete a full qualitative-variable modeling checklist.