## Bernoulli or binomial random variable

A variable representing a trial or sample that can have only two possible outcomes, such as zero or one.

Applying the binomial formula to determine expected probabilities associated with particular sampling outcomes is useful, and there is an even broader lesson that can be learned from the binomial. The examples up to this point have focused on the expected value.

are also easy to obtain (see Appendix). The standard error is the standard deviation of a mean, and the mean in this case is the expected value or allele frequency p or q. For genetic drift under the Wright-Fisher model, equally frequent alleles will give the widest range of outcomes for a given sample size (Fig. 3.6). The variability in allele frequency caused by genetic drift decreases as a population approaches fixation or loss, causing pq to approach zero (Fig. 3.7). This result makes intuitive sense. When alleles are equally frequent, sampling error is equally likely to increase or decrease allele frequency and could produce an outcome anywhere along the spectrum of possible allele frequencies. At the other extreme,

0 0