A theory might postulate that the distribution of the numbers of sternopleural bristles on male Drosophila melanogaster follow a Poisson distribution due to a Poisson process regulating their expression. Use a goodness-of-fit test to determine whether or not the Poisson model is a reasonable model for the before-selection population. If the Poisson model is not a good fit, then are the frequencies over- or under-dispersed? Assume that you have a simple random sample of individuals from the population.
Table 5.15 of Manly (1985). The Survival of male Drosophila melanogaster with different numbers of sternopleural bristles.
Manly, B.F.J. 1985. The Statistics of Natural Selection. Chapman and Hall, London, 484pp.
| Drosophila Selection Experiment of O'Donald (1971) |
| Sternopleural Bristle Numbers |
Frequency Before Selection |
Frequency After Selection |
| 12 | 1 | 0 |
| 13 | 1 | 0 |
| 14 | 25 | 7 |
| 15 | 51 | 8 |
| 16 | 98 | 45 |
| 17 | 104 | 36 |
| 18 | 106 | 32 |
| 19 | 67 | 22 |
| 20 | 54 | 17 |
| 21 | 24 | 6 |
| 22 | 21 | 4 |
| 23 | 7 | 1 |
| 24 | 4 | 1 |
| 25 | 0 | 0 |
| 26 | 1 | 0 |
| 564 | 179 |