Scatter in dose-response data

2 Scatter in dose-response data The figure of the Stomoxys strains also illustrates the wide scatter expected for the response data. Each point is based on 20 individuals, i.e., more than 400 flies plus controls (60) were used in this small experiment. The scatter is formidable in spite of the great number of flies used. The reason is the stochastic nature of the outcome. For instance, the probability (P) of getting exactly 15 dead flies by using a dose that should kill 75% is only P = 0.203. It is much more probable that we get another wrong value. These results may be calculated from the binomial formula 

---