The discriminating capacity of a measuring instrument: Revisiting Bloom (1942)’s theory and formula