I am a proponent of open science and sharing data and code to improve the reproducibility of archaeological analyses. I share data associated with many of my publications on the Digital Archaeological Repository (tDAR) which can be accessed via the link on the right. Data generated through the Southwest Social Networks and Chaco Social Networks projects are available to researchers via a user agreement with the University of Arizona, Archaeology Southwest, and Arizona State University which can be found here.
R-Scripts and Markdown Documents
The links below represent R-scripts and R Markdown documents associated with my research and teaching. I am slowly in the process of migrating old scripts to Markdown and putting several new analytical tools on GitHub complete with improved documentation, sample datasets, and error checking tools. Check back here in the coming months for updates.
Network science and statistical techniques for dealing with uncertainties in archaeological datasets
R Markdown document associated with the workshop at the Computer Applications and Quantitative Methods in Archaeology Meeting in Atlanta, GA (2017) hosted by Matt Peeples and Tom Brughmans. Network Science and Statistical Techniques for Dealing with Uncertainties in Archaeological Datasets.
Additional information for conducting bootstrapped correspondence analysis following the Peeples and Schachner (2012) Journal of Archaeological Science article. This approach uses a bootstrap procedure to estimate sampling error and assess the stability of CA solutions. GitHub link.
R Markdown document walking through the basic use of map data in R. This document was originally designed to accompany an assignment in my graduate course at ASU: ASB 568: Intrasite Research Strategies in Archaeology.
Script for calculating Brainerd-Robinson similarity coefficients based on tabular count or percent data. This script also estimates the probability that a given difference between two samples is the product of sampling error when count data are given. GitHub link.
This R Markdown document walks through a series of examples of basic point pattern analysis in R including Ripley’s K and related measures, quadrat and spatial tests of inhomogeniety, Poisson point process models, the Pair Correlation Function, and G-hat measures. This document was originally designed to accompany an assignment in my graduate course at ASU: ASB 568: Intrasite Research Strategies in Archaeology.
Script for calculating mean ceramic dates (following South 1977) for sites with data ceramic materials. This script is designed to run for a large number of sites in one csv file and produces MCD estimates as well as estimates of confidence intervals based on Monte Carlo procedure. GitHub link.
This R Markdown document walks through a series of LISA analyses in R including Moran’s I and Getis-Ord G* analysis. This document was originally designed to accompany an assignment in my graduate course at ASU: ASB 568: Intrasite Research Strategies in Archaeology.
This R script replicates the analyses presented in : Kintigh, K. 1984. Measuring Archaeological Diversity by Comparison with Simulated Assemblages. American Antiquity 49: 44-54.
This R Markdown document walks through an example of the pure locational clustering approach to spatial analysis of point patterns in R. This includes K-means cluster analysis, DBSCAN density based clustering, fuzzy K-means analysis, as well as additional analyses designed to help parameterize cluster models. This document was originally designed to accompany an assignment in my graduate course at ASU: ASB 568: Intrasite Research Strategies in Archaeology.
This R Markdown document walks through an example of the unconstrained clustering approach to spatial analysis of point or grid data in R (see Whallon 1984; Kintigh 1990). This includes procedures for obtaining grid counts from point located data, K-means cluster analysis, and a procedure for evaluating the statistical significance of homogeneity of artifact types by cluster assignment. This document was originally designed to accompany an assignment in my graduate course at ASU: ASB 568: Intrasite Research Strategies in Archaeology.
This R-script implements a means for statistically assessing the degree of co-occurrence between classes within a tabular dataset based on an idea origially suggested by James Allison and published by Keith Kintigh in 2006. The script produces a square matrix comparing each class to every other class with values representing the number of standard deviations more or less than expected (based on relative frequencies of occurrence) two classes co-occur.
This R script produces a exploratory visual for assessing site date ranges originally suggested by Wesley Bernardini. The plot consits of horizontal bars on an x-axis reprenting time with the width determined based on the production span for that type and with the height of each box determined based on the relative frequency of that type with types sorted vertically by start date. The script reads in csv files in the formatted of the example dataset here and produces a pdf containing results for each site.