Churn Without Fragmentation: How a Party-Label Bug Reversed My Headline Finding

# Data Quality Mistakes Can Flip Your Conclusions Upside Down A researcher analyzing English election data discovered that a simple labeling error in how political parties were categorized completely reversed what their data appeared to show. The mistake—using inconsistent party names instead of standardized codes—fragmented the data into artificial groups and made trends look opposite to reality. It's a stark reminder that before you trust any analysis or business decision based on data, someone needs to verify that the underlying categories are clean and consistent, not just take the raw labels at face value.