Data Cleaning & Outlier Removal

Data Cleaning & Outlier Removal (SPSS / Statistical Analysis)

Overview:

Performed thorough data cleaning and preprocessing on a 300-observation dataset to handle extreme outliers and normalize the distribution. By identifying and removing skewed data points, I restored statistical validity to ensure accurate population-level modeling and reliable linear regression metrics.

Key Process & Deliverables:

Outlier Detection & Removal: Screened baseline metrics to detect extreme variance and eliminate influential outliers that skewed overall population estimates.

Residual & Distribution Analysis: Evaluated standardized regression residuals using histograms and normal probability curves to confirm a bell-shaped, normal distribution ($N = 300$, $\mu \approx 0$, $\sigma \approx 1$).

Data Integrity & Reporting: Produced clean, unskewed variables ready for downstream predictive modeling, regression analysis, and executive reporting.

Outcome:

Transformed raw, heavily skewed survey/financial data into a balanced, statistically sound dataset, improving model accuracy and preventing misleading conclusions.