Category: Science
-
PSLE Science Reality Lab Vol No.535 | “AUC = 0.90” — Is the Classifier 90% Accurate?
A massive Primary 5/6 evidence guide to ROC AUC: why AUC = 0.90 is not automatically 90% accuracy, and how thresholds, test data, errors and independent validation change the claim.
-
PSLE Science Reality Lab Vol No.539 | “Result: R — Rejected” — Does That Mean the Substance Was Not There?
A massive Primary 5/6 PSLE Science evidence guide to rejected laboratory data: why an unusable result is not a measured zero, a valid non-detect, or proof that a substance was absent.
-
PSLE Science Reality Lab Vol No.534 | “Six Check Results Keep Rising but Stay Inside the Limits” — Can We Ignore the Trend?
A Primary 5/6 evidence guide to scientific control charts: why a sustained trend inside the limits can still signal a changing measurement process and deserves investigation.
-
PSLE Science Reality Lab Vol No.533 | “Proficiency-Test z-Score = 0.2” — Is the Laboratory 99.8% Accurate?
A Primary 5/6 evidence guide to laboratory proficiency-test z-scores: why z = 0.2 is a standardised comparison, not 0.2% error or 99.8% accuracy.
-
PSLE Science Reality Lab Vol No.532 | “Water-Quality Geometric Mean = 40” — Were Most Samples About 40?
A Primary 5/6 evidence guide to geometric means in environmental reports: why one summary value does not show what most samples were, the maximum, or the full spread.
-
PSLE Science Reality Lab Vol No.531 | “The Radar Map Is Blank in This Wedge” — Does That Mean No Rain Fell There?
A Primary 5/6 evidence guide to blank weather-radar sectors: why a missing or weak echo may reflect beam blockage, coverage or attenuation rather than proof that no rain fell.
