Outlier Detection Analyzer
Identify potential outliers in numerical research data using the IQR and Z-score methods. Upload your CSV, select a numerical variable, and examine unusual observations before statistical analysis.
Upload Your CSV Dataset
No file selected
Commonly, an absolute Z-score above 3
is considered potentially unusual.
🔒 Privacy:
Your CSV is processed locally in your browser.
It is not uploaded to ResearchUtility.
Statistical Summary
—
Valid Observations
—
Mean
—
Median
—
Sample SD
—
Q1
—
Q3
—
IQR
—
Potential Outliers
Outlier Detection Limits
| Lower IQR Fence | — |
|---|---|
| Upper IQR Fence | — |
| Z-Score Threshold | — |
| IQR Outliers | — |
| Z-Score Outliers | — |
Observation-Level Results
| Row | Value | Z-Score | IQR Status | Z-Score Status | Overall Status |
|---|
Important research note:
An observation flagged as a potential outlier is
not automatically an error and should not be
automatically deleted. Investigate the original
measurement, experimental conditions, data-entry
records, and study design before deciding how to
handle any unusual observation.