Detect statistical outliers in numeric CSV columns using IQR or Z-score

Outlier detection is a critical step before computing column statistics, training ML models, or building dashboards—a single data entry error (typing 15000 instead of 150) will skew averages dramatically. The IQR (Interquartile Range) method defines outliers as values below Q1 − 1.5×IQR or above Q3 + 1.5×IQR, which is resistant to the influence of the outliers themselves. The Z-score method flags values more than N standard deviations from the mean and is appropriate for normally distributed data. DataFix annotates each outlier row with the method used, the threshold value, and a severity level (mild vs. extreme). Statistical anomalies caused by sensor errors, misplaced decimal points, or faulty currency conversions distort statistical distributions, variance metrics, and regression models. Detecting outliers using interquartile range (IQR) and standard deviation fences pinpoints extreme observations for audit and cleaning. Statistical anomaly screening requires careful threshold configuration to prevent false positives in skewed distributions. While standard Z-score analysis assumes a Gaussian bell curve, real-world e-commerce prices, user session lengths, and transaction amounts typically exhibit heavy-tailed distributions. In these cases, Tukey's interquartile range (IQR) rule using 1.5x or 3x fences provides substantially more robust outlier boundaries, isolating true instrumentation errors or fraudulent spikes without erroneously discarding valid high-value orders. Screening numeric metrics using interquartile range (IQR) fences identifies anomalous transaction spikes, sensor errors, and calculation mistakes without assuming Gaussian normality. Identifying these statistical deviations helps analysts isolate data corruption before training predictive models or publishing financial reports.

What is the IQR method?

Values below Q1 - 1.5×IQR or above Q3 + 1.5×IQR are flagged as outliers. This is the standard Tukey fence.

What is the Z-score threshold?

By default, values with |Z| > 3 are flagged. You can adjust the threshold.

Validate

Validate CSV

Audit CSV files for structural errors, encoding problems, and row-length inconsistencies.

Open Tool →

Validate

Validate JSON

Parse and validate JSON strings to expose syntax errors at exact line and column positions.

Open Tool →

Validate

Find Duplicate Rows

Identify and count duplicate rows in CSV files by exact match or by key column selection.

Open Tool →