Use the Metrics

Data Analytics: The Basics

Start with the question you want to answer.

The heart of data analytics is curiosity and the need to answer a specific question. It's usually some hypothesis you have or a desire to set benchmarks and determine how well (or not) something is functioning.

For example: is my licensed paralegal practitioner (LPP) program helping address the access to justice gap in my jurisdiction?

ℹ️ Explore the Metrics to discover the types of questions justice system participants are already asking.

Consider what individual data points you need to answer that question.

Metrics typically consist of multiple pieces of data (a.k.a. "measures") that, when taken together, reveal an important insight related to your question. It can be as simple as a percentage or as complex as a multi-factored formula you create to understand or test your hypothesis.

For example: The answer to our LPP question above most likely depends on a number of individual measures, including things like: client satisfaction, service affordability, or number of court orders entered by alternative legal professionals.

Source and prepare your data.

Finding and preparing your data will be your most resource-intensive exercise. For the most part, law-related data takes the shape of quantitative (numbers-based) figures or qualitative (narrative-type) information. If sourcing quantitative data from multiple sources, you'll need to ensure your data is normalized for consistency's sake. Qualitative data may need to be categorized or summarized in some way.

For example: client satisfaction for LPP providers will come in the form of qualitative surveys or client interviews, so you'll need to develop instruments and conduct your research. But (assuming its captured) the number of court orders entered by alternative legal professionals can be quantified by the courts and output as a number or percentage.

Analyze and interpret your results.

Data analysis can take many shapes and can lead to differing interpretations. For simple questions, a comparative analysis might be sufficient. For more complex inquiries (and larger quantities of data), visualization or more sophisticated data science approaches might be necessary. But ultimately, your data analysis should provide a signal through the noise and help you gain insights on your initial question.

For example: if only 30% of LPP clients are reporting satisfaction with the outcomes of their services and just 10% of LLP-led court filings are reaching a final judgment, there may be significant questions as to the effectiveness of the program.

ℹ️ Check out the Metrics in Action page to see how others are doing this in their own jurisdictions.

Determine your next steps.

Data should drive action. If problems are uncovered, your findings should inspire changes that will address the issue(s) and lead to forward progress. In many cases, the results may inform new questions as your understanding of the issue evolves. If your results are pleasantly surprising, it's worth sharing them with others or applying those learnings to other aspects of your work (as is relevant).

For example: low client satisfaction and through-put rate for LPP filings might lead to more robust training programs for program participants. It might also lead to new inquiries and hypothesis as to why judgment rates are low - with answers ranging from abandonment to settlement and everything in between.