C. Project Checklist

Use this checklist before submitting an empirical project.

Research Question

• The research question is specific.
• The dependent variable is clear.
• The main explanatory variable is clear.
• The question is feasible with available data.
• The question is economic, not only technical.

Data Source

• The data source is identified.
• The data source can be verified.
• The collection date is recorded.
• The raw dataset is preserved.
• Any limitations are stated.

Dataset Integrity

• The unit of observation is defined.
• Missing values are checked.
• Duplicate rows are checked.
• Outliers are checked.
• Categorical levels are checked.

Variable Definitions

• Each variable has a clear definition.
• Units are reported where relevant.
• Constructed variables are explained.
• The dependent variable is justified.
• Control variables are justified.

Data Cleaning

• Cleaning steps are documented.
• Raw and cleaned files are separate.
• No unexplained observations are deleted.
• Code can reproduce the cleaned dataset.
• Cleaning choices are economically reasonable.

Descriptive Statistics

• Numeric variables have summary statistics.
• Categorical variables have frequency tables.
• Unusual values are discussed.
• Descriptive statistics match the research question.

Graphs

• Graphs are readable.
• Axes are labeled.
• Each graph has a purpose.
• Graphs support the research question.
• Graphs are not used as causal proof.

Econometric Model

• The regression equation is written clearly.
• Variables in the equation match the dataset.
• The model choice is justified.
• Coefficients are interpreted correctly.
• Causality claims are avoided unless the design supports them.

Diagnostics

• Residuals are checked.
• Heteroskedasticity is considered.
• Multicollinearity is considered.
• Functional form is considered.
• Robustness checks are included where appropriate.

Interpretation

• Results are explained in words.
• Statistical significance is reported.
• Economic significance is discussed.
• Limitations are stated.
• Conclusions follow from the evidence.

Tables and Figures

• Tables are clean and readable.
• Figures are numbered or clearly referenced.
• Regression results include key statistics.
• Tables and figures are discussed in the text.

Writing Quality

• The paper has a logical structure.
• Paragraphs are concise.
• Terms are used consistently.
• The paper answers one main question.
• The conclusion does not introduce new results.

Reproducibility

• Code runs from top to bottom.
• File paths are documented.
• Packages are listed.
• Random seeds are set when needed.
• Output can be recreated.

Final Submission

• Dataset is included or linked.
• Code file is included.
• Variable table is included.
• Results tables are included.
• Final report is complete.

Minimum Acceptable Project Package

• dataset
• code
• variable table
• graphs
• regression results
• diagnostics
• short empirical report