Popper and ‘diminishing returns from repeated tests'