First off, having the review of everything that we learned last semester was very helpful. It wasn’t too redundant, and it was nice to see a highight of many of the important things that we learned last semester!

One thing that I was fuzzy on was the R work for working with proprotions. However, after our review I remember how to work through problems creating confidence intervals for proportions, such as the one I’ve created below.

Say we have 90 trials and 17 of them resulted in success. Let’s create a 99% confidence interval for the probability of success.

First, find a point estimate for the probability of successs.

p.est <- 17/90
p.est
## [1] 0.1888889

Next estimate the variability of the point estimator.

pvar <- p.est * (1-p.est) / 90
my.sd <- sqrt(pvar)
my.sd
## [1] 0.04125933

Now produce a 99% confidence interval for the probability of success.

p.est + c(1, -1) * qnorm(.005) * my.sd
## [1] 0.08261191 0.29516587

I am 99 percent confident that the true proportion of success lies between .0826 and .2952.