the idea behind the test–retest approach to measuring the reliability of scales is that if random variations are present, they will be revealed by variations in the scores between the two sampled measurements.