ANALYTICAL
METHODS
IN
ARCHAEOLOGY

Anthropology 588	Geostat Lab: Alderman
University of Virginia	Tuesday:5:00-7:30
Spring 2002	Fraser D.Neiman

Description This course examines quantitative analytical techniques used in archaeology. Topics include, regression, smoothing, correlation, measures of diversity and distance, spatial autocorrelation and Mantel methods, seriation, ordination, and clustering. The course features intensive analysis of real archaeological datasets, motivated by real archaeological problems. Highly recommended pre-requisite: an introductory course in statistics. Texts The following texts are required. They are available at the UVA Bookstore: Shennan, Stephen 1997 Quantifying Archaeology, 2nd Edition. University of Iowa Press, Iowa City. Manly, Bryan F. J. 1994 Multivariate Statistical Methods: A Primer. Chapman and Hall/CRC, Boca Raton Delwiche, L.D. and S.J. Slaughter 1998 The Little SAS Book, 2nd Edition. SAS Institute, Cary NC. Consult the Reading List for additional reading, all of which will be available on the Tookit Web Page for the course, or on reserve at Clemons Library. As you will soon discover, you can never have too many statistics books! To help you get started, here are a few recommendations. We'll be sampling all of them during the course. For a gentle review of basic descriptive and inferential statistics in archaeology, check out Robert D. Drennan's Statistics for Archaeologists: A Commonsense Approach (1996, Plenum, New York). Drennan stresses the Exploratory Data Analysis(EDA)approach developed by John Tukey and his students, starting in the 1970's. Visualizing Data , by William S. Cleveland (1993, Hobart Press, Summit New Jersey) offers beautifully illustrated treatment of both elementary and more advanced EDA methods (e.g. loess). M.J. Baxter's Exploratory Multivariate Analysis in Archaeology (1994, Edinburgh University Press, Edinburgh) offers first rate coverage of the standard multivariate methods: principal components, correspondence, cluster, and discriminant analyses. Sadly, it is out of print. But it's worth trying to find a copy. There is considerable overlap, not widely recognized, between analytical methods used in ecology and archaeology. Astoundingly complete coverage of the ecological side can be found in Numerical Ecology 2nd Edition, by Pierre and Louis Legendre (1998, Elsevier, Amsterdam). Brian Manly's Randomization, Bootstrap, and Monte Carlo Methods in Biology offers both an accessible introduction to and up-to-date coverage of techniques that figure importantly in archaeology as well (2001, Chapman and Hall/CRC Press, Boca Raton). Finally, if you do not already own one, you should acquire a good, comprehensive statistics text. I recommend Sokal, Robert R. Sokal and F. James Rohlf's classic, Biometry: The Principles and Practice of Statistics in Biological Research , 3rd Edition (1995, W.H. Freeman, San Fransico).

Requirements Written work for the course includes Problems Sets and a Final Project. Problem Sets will be assigned nearly every week. Completed problem sets will be due the following week, at the beginning of class, when we will discuss them. Your write up should include not only your numerical results, illustrated with appropriate graphics, but also a description of how you got them and what you think they mean, both in statistical and archaeological terms. The text should be no more than three pages, doubled spaced. You should also be prepared to offer the class a succinct summary of your work, as a contibution to the class discussion of the problem sets. Late problem sets will not be accepted. The Final Project is your opportunity to use the methods we have covered in the course to tease substantive meaning out of archaeological data. Note that the emphasis here is on meaning: the project should emphasize analytical methods as tools to make and evaluate defensible inferences about past dynamics from material evidence. You will need to identify an historical or anthropological issue and one or more sets of quantitive archaeological data that might be used to illuminate it. You are encouraged to choose issues and data in which you have a personal research interest and, therefore, basic familiarity with the current background literature. The data you use should contain information on several different variables that are of potential relevance to your problem. A key part of the project will be defending their relevance. The data should contain additional variables that independently document either the temporal or spatial contexts of the objects (assemblages, artifacts) characterized by the "relevant" variables. Over the first several weeks of the course, we will work together to identify suitable issues and datasets for everyone. The topic and dataset(s) for your project must be approved by the instructor. Your write up for the final project should not exceed 15 pages (double-spaced), exclusive of graphics. Grades of the course will be computed as follows: PS*.6 + FP*.3 + CD*.1, where PS=mean score for all problem sets, FP=score for the final project, and CD=mean score for contributions to class discussion.

Reading List Click here

Datasets Sherd thickness in Woodland-Period Illinois. Site size and population in 16th-century Peru. Site size and population in the 19th and 20th-century Southwest. Effective temperature, primary productivity, and variation in hunting, gathering, and fishing among forager groups. Shape variation in Poverty Point objects from the Lower Mississippi Valley. Stylistic Variation on ceramics from Woodland-Period Illinois Ceramic type frequencies from layers within room blocks at Los Muertos Pueblo Ceramic type frequency variation in plowzone quadrats at Site 7, Monticello Neutron Activation Analysis of Mississippian ceramics (MUUR Web Site) Neutron Activation Analysis of Woodland ceramics from Pinson Mounds, Tennessee (MURR Web Site)

Useful Bits of Code, etc. BinomialSamples.sas SAS code for binomial sampling distribution simulation (Project 3). NormalSamples.sas SAS code for normal sampling distribution simulation. SamplingDistributions.xls Excel spreadsheets with examples of binomial and Normal (Gaussian) sampling distributions. TimetechSubsetANOVA.xls Excel spreadsheet illustrating a one-way ANOVA on sherd thickness for two sites in the Timetech dataset. TimetechSubsetANOVA.sas Illustrates the use of SAS's Proc GLM to do the same one-way ANOVA on sherd thickness for the same two sites in the Timetech dataset. Timetech4SitesANOVA.sas Illustrates the use of SAS's Proc GLM to do an (unbalanced) ANOVA on sherd thickness and minimum diameter for four sites in the Timetech dataset. Peru.sas SAS code for exploratory and regression analysis of the Schreiber and Kintigh data on population and site size in 16th-century Peru. ForagerGroups.sas SAS code for exploratory and regression analysis of Kelley's data on foraging groups. TimetechReg.sas SAS code for linear regression and loess analysis of the Timetech data, summarized at the site level. Pueblo.sas SAS code for regression analysis of Dohm's data on Pueblo population size and room counts. TimetechCorr.sas SAS code for the within-site analysis of the correlation between sherd thickness and pot diameter. GraysonandDelpecheX2.sas SAS code for chi-square and adjusted residual analysis of long-bone shafts and ends from Le Flageolet. DistanceBootstrap.sas Bootstrap sampling of three inter-assemblage distance measures. BraunIn.sas Code to read and massage Braun's style data so we can analyse it. DistanceMatrix.sas Compute an inter-assemblage distance matrix using one of three measures. Mantel.sas Compare two distance matrices using Mantel methods. Diversity.sas Compute diversity measures -- Shannon and Simpson's diversity and evenness, richness, and Ewen's estimate of theta. Conkey's Magdalenian data are included as an example. BraunDiv.sas Compute diversity measures -- Shannon and Simpson's diversity and evenness, richness, and Ewen's estimate of theta -- for Braun's style data. BootDiversity1.sas Compute diversity measures --Shannon and Simpson's diversity and evenness, richness. Then bootstrap means and 95% C.L's. Conkey's Magdalenian data are again included as an example. PCAExample.sas IML code to explicate PCA. DeBoeretal.sas PCA and MDS of Megger's data from AM-MA-9 on the banks of the Rio Negro CAExample.sas IML code to explicate CA. PCAvsCASeriation.sas Illustrates -- with artificial data -- the "horseshoe effect" in PCA and CA ordinations of nonlinear data. LosMuertosCA.sas CA of type counts from Los Muertos Pueblo. Also includes code to compute mean ceramic dates to compare with the CA results. Site7CA.sas CA of ceramic type counts from plowzone quadrats at Site 7 at Monticello. Also includes code to smooth (filter) the data and to compute mean ceramic dates to compare with the CA results.

Software Tools ITC Home Directory Service Set up a home directory on Blue.unix. SAS Version 8 Tutorial A Web-based introduction to SAS. SAS Version 8 Documentation at UVA Complete documentation on the SAS system. SAS Version 8 Installation Instructions from ITC In case you want to get a copy for your own machine. Seriation Diagram Tool An Excel spreadsheet with VBA code that draws seriation diagrams. Beta version by Tim Hunt. Scatterplot Label Macro for Excel Allows you to label scattterplots(something that Lotus-123 could do in 1982, but Excel cannot, 20 years later).