Can you elaborate? As a statistical noob, this feels like a huge effect (1 of 50, or 2% - vs - 13 of 26, or 50%).