Reverse Exception Filter

A data selection bias where a researcher removes “exceptional” cases not because they are true outliers, but because they contradict the desired hypothesis. Unlike normal outlier removal (which may be justified), reverse exception filtering targets any data point that would make the result significant or interesting. The removed cases are often the most informative – extreme values, rare events, or real-world anomalies. The filter is “reverse” because it operates to eliminate signal rather than noise.

Reverse Exception Filter Example: “He deleted the three participants who had the strongest response to the treatment, then declared the drug ineffective – reverse exception filter, cutting the signal because it was too bright.”

Reverse Exception Filter by Dumu The Void April 25, 2026