MOS confidence-interval calculator

A dependency-light Python script (standard library only) that turns raw per-viewer ratings into a defensible subjective-test result: per-condition Mean Opinion Score, standard deviation, and the ITU-R BT.500 95% confidence interval (ε = 1.96 · S/√N, with a Student-t multiplier auto-selected for small N). It flags when N is below the ~15 practical floor, reports whether two conditions' confidence intervals overlap (i.e. whether the test can separate them), and runs on a CSV or a built-in sample d