Graduate students often lose points—or entire thesis chapters—because of avoidable statistical errors. This guide highlights the most common mistakes and how to fix them before they become a problem.
1. Using the Wrong Regression Model
OLS is not always appropriate. Check assumptions before choosing your model.
2. Ignoring Multicollinearity
Highly correlated predictors inflate standard errors and distort results.
3. Treating Categorical Variables as Numeric
Always convert categories into dummy variables or factors.
4. Failing to Check Residuals
Residual plots reveal heteroscedasticity, nonlinearity, and outliers.
5. Misinterpreting p‑values
A p‑value is not the probability your hypothesis is true.
6. Overfitting the Model
Too many predictors reduce generalizability and inflate R‑squared.
7. Not Documenting Your Workflow
Your advisor will ask how you produced your results—be ready.
Conclusion
Avoiding these mistakes will strengthen your thesis and improve your confidence. For more support, explore the Graduate Toolkit Hub or visit the Statistics Tutoring NYC page.

Leave a Reply