Archaeological Data Analysis Cheat Sheet: Methods, Tools, and Best Practices

Introduction to Archaeological Data Analysis

Archaeological data analysis encompasses the systematic examination and interpretation of material remains to understand past human activities, behaviors, and cultural processes. It bridges quantitative and qualitative approaches to transform archaeological finds into meaningful insights about past societies. Effective data analysis is essential for rigorous archaeological interpretation, helping archaeologists move beyond descriptive inventories toward explanatory models and cultural reconstructions.

Core Data Types in Archaeology

Artifact Data

Spatial Data

Temporal Data

Quantitative Analysis Methods

Descriptive Statistics

Exploratory Data Analysis

# Example R code for basic exploratory analysis of lithic dimensions
# Load libraries
library(tidyverse)
library(ggplot2)

# Load data
lithics <- read.csv("lithic_assemblage.csv")

# Summary statistics
summary(lithics)

# Boxplot of length by raw material
ggplot(lithics, aes(x=raw_material, y=length_mm)) +
  geom_boxplot() +
  theme_minimal() +
  labs(title="Lithic Length Distribution by Raw Material",
       x="Raw Material Type", y="Length (mm)")

# Scatter plot of length vs. width with tool type
ggplot(lithics, aes(x=length_mm, y=width_mm, color=tool_type)) +
  geom_point(alpha=0.7) +
  theme_minimal() +
  labs(title="Length vs. Width by Tool Type",
       x="Length (mm)", y="Width (mm)")

Inferential Statistics

Multivariate Analysis

# Example R code for PCA of ceramic attributes
# Load libraries
library(FactoMineR)
library(factoextra)

# Load data
ceramics <- read.csv("ceramic_assemblage.csv")

# Select numeric variables for PCA
ceramic_vars <- ceramics %>%
  select(rim_diameter, wall_thickness, height, weight)

# Run PCA
ceramic_pca <- PCA(ceramic_vars, graph = FALSE)

# Visualize results
fviz_pca_biplot(ceramic_pca, 
                habillage = ceramics$vessel_type,
                palette = "jco",
                addEllipses = TRUE,
                title = "PCA - Ceramic Vessel Attributes")

Spatial Analysis Methods

Site-Level Analysis

Regional Analysis

GIS Operations for Archaeology

# Example Python code using ArcPy for archaeological site predictive modeling
import arcpy
from arcpy.sa import *

# Set workspace
arcpy.env.workspace = "C:/ArchaeologyProject/GIS"
arcpy.CheckOutExtension("Spatial")

# Environmental factors (input rasters)
slope = Raster("slope.tif")
dist_to_water = Raster("dist_to_water.tif")
elevation = Raster("dem.tif")
aspect = Raster("aspect.tif")

# Reclassify factors to suitability scores (1-10)
slope_reclass = Reclassify(slope, "Value", 
                           RemapRange([[0,5,10], [5,10,8], [10,15,6], 
                                      [15,25,3], [25,90,1]]))

water_reclass = Reclassify(dist_to_water, "Value",
                          RemapRange([[0,100,10], [100,500,8], [500,1000,5], 
                                     [1000,2000,2], [2000,10000,1]]))

# Weight and combine factors
predictive_model = (slope_reclass * 0.3) + (water_reclass * 0.5) + 
                  (elevation * 0.1) + (aspect * 0.1)

# Save output
predictive_model.save("site_prediction_model.tif")

# Validate with known sites
arcpy.MakeFeatureLayer_management("known_sites.shp", "sites_lyr")
ZonalStatisticsAsTable("sites_lyr", "OBJECTID", predictive_model, 
                       "validation_stats.dbf", "DATA", "ALL")

Chronological Analysis Methods

Seriation Techniques

Bayesian Chronological Modeling

# Example R code for Bayesian radiocarbon modeling with OxCal syntax
# This would typically be run in OxCal, but the syntax is shown here

# Define a sequence model in OxCal
"
Plot()
 {
  Sequence("Site A")
  {
   Boundary("Start of occupation");
   Phase("Early occupation")
   {
    R_Date("Sample 1", 5000, 30);
    R_Date("Sample 2", 4950, 35);
    R_Date("Sample 3", 4920, 40);
   };
   Boundary("Transition");
   Phase("Late occupation")
   {
    R_Date("Sample 4", 4800, 30);
    R_Date("Sample 5", 4750, 35);
    R_Date("Sample 6", 4700, 25);
   };
   Boundary("End of occupation");
  };
 };
"

Duration and Event Analysis

Compositional and Materials Analysis

Statistical Methods for Compositional Data

Visualization of Compositional Data

Dealing with Common Data Challenges

Missing Data Strategies

Small Sample Size Approaches

Taphonomic Bias Corrections

Integrated Approaches and Interpretation

Combining Multiple Data Types

Interpretive Frameworks

Digital Tools and Software

Statistical and Data Analysis Software

Spatial Analysis Tools

Visualization and Presentation Tools

Best Practices for Archaeological Data Analysis

Data Management

• Create comprehensive data dictionaries documenting all variables and coding systems
• Implement consistent measurement protocols to ensure comparability within and between projects
• Maintain original (raw) data separate from processed/analyzed data
• Use version control for analytical scripts and databases
• Follow data citation standards when using datasets from other sources
• Plan for long-term archiving in sustainable formats and repositories

Analytical Approach

• Begin with clear research questions rather than applying methods for their own sake
• Use exploratory analysis before confirmatory statistics
• Consider multiple working hypotheses rather than single hypothesis testing
• Acknowledge and quantify uncertainty in measurements and interpretations
• Combine quantitative and qualitative approaches for robust interpretations
• Document all analytical steps to ensure reproducibility

Reporting and Publication

• Provide access to raw data when possible through repositories or supplements
• Clearly describe methods including software versions and analytical parameters
• Visualize data effectively with appropriate chart types and clear labeling
• Report appropriate statistical details (sample sizes, p-values, effect sizes)
• Acknowledge limitations of data and methods
• Use open-access formats where possible for maximum accessibility

Resources for Further Learning

Textbooks and References

• Baxter, M. (2003). Statistics in Archaeology. London: Arnold.
• Conolly, J., & Lake, M. (2006). Geographical Information Systems in Archaeology. Cambridge: Cambridge University Press.
• Drennan, R. D. (2009). Statistics for Archaeologists: A Common Sense Approach. New York: Springer.
• Lock, G. (2003). Using Computers in Archaeology: Towards Virtual Pasts. London: Routledge.
• VanPool, T. L., & Leonard, R. D. (2011). Quantitative Analysis in Archaeology. Chichester: Wiley-Blackwell.

Journals with Strong Methodological Focus

• Journal of Archaeological Science
• Journal of Archaeological Science: Reports
• Archaeometry
• Archaeological and Anthropological Sciences
• Advances in Archaeological Practice

Online Resources and Communities

• Open Science Framework – Platform for sharing archaeological data and methods
• Archaeological Data Service – Repository and methodological guides
• Digital Archaeological Record (tDAR) – Digital repository for archaeological data
• GitHub Archaeological Repositories – Open-source archaeological code and tools
• Archaeology Data Analysis Forum – Community discussion forum

Training Opportunities

• Archaeological Institute of America Workshops
• Digital Archaeological Practice Online Courses
• Software Carpentry for Archaeologists
• Programming for Evolutionary Biology – Includes applicable methods for archaeological data

Remember that archaeological data analysis should always serve the broader goals of archaeological interpretation and understanding past human societies. The most sophisticated analyses still require thoughtful archaeological interpretation grounded in solid theoretical frameworks.