Defensible landscapes in archaeology. The defensibility of a given landscape in the Northwest United States was studied in the Journal of Anthropological Archaeology (May 2014).

Archaeologists typically define “defensive” locations as those that are hidden with high elevation and escape routes. A question arose regarding the defensibility of trench embankments.

Each in a sample of 1,914 archaeological sites was classified as

“highly defensible” or not. In addition, whether the site was located in a trench embankment was determined. The results of the categorizations (number of sites in each category) are summarized below.

Highly Defensible Not Highly Defensible Total Trench Embankment 9 14 23 Non-trench Embankment 311 1,580 1,891

a. Is this sufficient evidence to conclude that the proportion of trench embankment sites that are “highly defensible”

differs from the corresponding proportion of non-trench embankment sites? That is, is the defensibility status of a site dependent on site type (trench embankment or not)?

b. Fisher’s exact test (see Exercise 13.45) resulted in a p-value of p = .008. Interpret this result, and then explain why this test is preferred over the x2 test in part a.