Citing this Document

This document can be cited as follows:

Peeples, Matthew A.

2019 A Brief Introduction to Archaeological Networks in R. [online]. Available:


Getting Started

This markdown document provides code and detailed examples associated with a segment of an R Workshop held on March 26th, 2019 in Cambridge, UK at the “Big Data in Archaeology” conference hosted by the McDonald Instititue of Archaeological Research. This document assumes you have a basic knowledge of network science terminology and at least some basic familiarity with R and R Studio. If you are new to R, I recommend you first run through the brief Code School examples focused on R available here ( The examples below can be run by simply copying and pasting the code below but modifications will require a bit of additional experience. For those new to network science in general, I recommend you start by reading the Peeples 2019 and Brughmans 2013 review articles (see recommended bibliography for more details). For those with more background in R and network statistics, there is also a more indepth version of this tutorial that goes into many more statistical techniques for evaluating network metrics and sensitivity analyses posted here


Getting Our Data into R

For the purposes of this workshop we will be using a real archaeological dataset as an example pulled from the Digital Archaeological Record (tDAR). These data are derived from research previously conducted by me associated with my “Connected Communities” book with the University of Arizona Press (Peeples 2018). These data represent the results of an analysis which defined clusters of cooking pottery from the Zuni/Cibola region of Arizona and New Mexico (ca. AD 1150-1325) based on a series of technological attributes recorded for just over 2,200 individual vessels. In the workshop we’ll practice searching for and retrieving these data directly from tDAR but I am also providing them here for future reference. This dataset is divided into two .csv files, the first with the counts of ceramic technological clusters by site and a second with the locations of those sites (not on tDAR). To ensure the security of site locational information as required by the state data repositories charged with maintaining these data, I have randomly relocated each settlement between 7-10 kilometers from their actual locations.