Number of happy doctors correctly identified by the 96 assessors, each of whom made a comparative judgement on 20 different pairs of doctors. The horizontal axis shows the number of times the doctor in a pair who was actually the happier was correctly identified as such by an assessor for each of the 20 pairs assessed by an assessor. The blue bars shows the actual numbers of correct judgements out of twenty for each assessor, and the yellow bars the expected distribution under a binomial distribution (left-hand vertical axis). The dashed red line shows the expected cumulative binomial distribution, and the solid line the actual cumulative binomial distribution (right-hand vertical axis). The light blue bars show pairs within the 95% range of the binomial distribution, whereas the dark blue bar show a pair which is outside the approximate 95% range, and hence is significant at about the 5% level.