Subgroup splits, held to the survey's own noise

Given each question's 2024 national share, lives fitted to records from 2016 to 2022 split five fresh 2024 questions on public spending (crime, education, defence, social security, space) by party and age, and all five landed inside the survey's own sampling noise, where four were needed. Levelled only to the 2022 national share, using nothing from 2024, all five passed too. Registered before any question was scored; each bar set at the error the true shares themselves would stay under 95 times in 100. The model was built on six earlier questions alone, where it passed 12 of 12.