Database Reference
In-Depth Information
Figure 7-23. 95% confidence intervals for fifth- and sixth-grade chess club participation
From these charts, it's apparent that based on these survey results, the president can be con-
fident about far less than he originally thought. In fact, nothing can be said about the relative
difference of chess club participation between the first and second graders of each city. The
intervals completely overlap one another.
Additionally:
▪ Bishop Village has been filtered out of the first and second grade results because the
sample size times the proportion (np) was less than 5, meaning that the normal approx-
imation isn't legitimate.
▪ While Castleborough seems to have much higher participation rates for third and fourth
graders, we can only be confident at the 95% level that rates are higher at Castleborough
than Gambitopolis and Bishop Village.
Search WWH ::




Custom Search