Services

Consultations

Whether you need a second opinion on your latest regression discontinuity design or need help understanding the difference between an independent and dependent variable, the Legal Data Lab is here to help! Our consultations cover everything from basic tutoring on key statistical and data management concepts, to troubleshooting runtime errors in your programming code, to advising clients on the tradeoffs between different research designs and modeling strategies. With roughly 20 years of data science experience, the Legal Data Lab is the place to go for statistical and programming advice!


Data Analysis & Interpretation

The nature of your research question and the data you have at your disposal generally shape which empirical method(s) to use. Even then, however, the appropriate research method to use in a given situation is not always obvious and interpreting the results of the analysis can be difficult. Let us do this hard work for you.

Here are some of the statistical and inferential methods in which we specialize:

  • Univariate descriptive statistics
  • Bivariate descriptive and inferential statistics
  • Linear regression for a variety of different dependent variables
  • Hierarchical and longitudinal analysis
  • Survival analysis
  • Exploratory and confirmatory dimensionality reduction techniques
  • (Quasi-)Experimental methods
  • Text analysis
  • Machine learning

Data Collection & Cleaning

With years of experience in Stata, SAS, R, and Python we can help optimize the collection and cleaning of research data so that you can focus on running your tests. Scraping websites, joining/merging/slicing/concatenating dataframes, or simply pulling data from canned resources like ICPSR or WRDS are daily and well-known activities for us. As librarians, finding the appropriate sources is also a common and familiar task.


Data Storage

Researchers have a few options for storing datasets. Those options are:

  • For research that includes a Library-created website, we regularly backup and make copies of the content on the web server, including local copies. If this is an option you would like to pursue, please contact us for more information.
  • For researchers who want a place to store their finished dataset, UVA hosts its own instance of Dataverse, a platform for preserving, sharing, citing, and exploring research data. The Law Library has its own "sub-Dataverse" for Law faculty and users can create their own accounts to organize and promote their research. File sizes are limited to 2GB per file and licensing restrictions lean towards broad, public use. For more information see: https://dataverse.lib.virginia.edu/

Visualization

Data visualization can often convey results in a more insightful manner than plain-text stats can. We have experience from hand-coding Javascript visualizations (with and without D3) to finessing the ever-growing list of visualization packages in R and Python, to outputting basic box plots and histograms. On top of this we occasionally play with ArcGIS when Google Maps and Leaflet don't cut it.

Training

We offer a series of applied trainings on different statistical software and programming packages, as well as specific topics in statistical modeling and inference. We've offered workshops on Natural Language Processing and Survival Analysis in association with UVA StatLab on Main Grounds, and we hope to offer these and other workshops at the Law School in the not too distant future.