There is a running joke within the legal profession that lawyers choose to go to law school because they are bad at math.1 Facially, this proposition makes sense. The bread and butter of the legal profession is written and oral advocacy, not numbers and arithmetic.
But the legal profession has never existed outside of the realm of numbers. And in today’s world of big data where judges’ decisions and opposing counsel’s actions can be quantified and packaged into orderly statistics, basic statistical literacy is a critical part of legal research and practice.
Statistical literacy refers to the skills pertaining to reading and interpreting statistical data.2 This includes analyzing the meaning of the data represented by statistics and understanding the limitations of those statistics. Statistical literacy does not necessarily include knowing how to mathematically calculate statistics; instead, it is focused on effectively engaging with statistical data that has already been created.3 At its core, statistical literacy is about being able to effectively judge if statistical information is accurate and trustworthy.
The good news is that most attorneys are already using statistical literacy skills in their everyday lives. For example, let us assume a study found that 55% of Michiganders prefer coffee over tea. Most people would be able to read the data and conclude that coffee is slightly more popular in Michigan than tea because they know statistics generally add up to 100% and that 55% is more than half of 100. Based on that knowledge, people could also deduce that 45% of Michiganders prefer tea. These basic interpretations of a single statistic are examples of using statistical literacy skills to analyze the meaning of a number.
Statistical literacy goes beyond just reading the numbers, however; it also includes critically analyzing their validity. In the previous example, effective statistical literacy would also include examining the collection methods and presentation of the information. For example, if the survey only included people from Detroit, then the results could not accurately be applied to all Michiganders. Or, if only 20 people from around the state were surveyed, the survey pool would be too small to be accurately applied to the entire state.
It is this second part of statistical literacy — critically engaging with the validity or accuracy of a statistic — that can be the most important part of statistical legal research. Presenting or crafting statistics in a manner that privileges a certain view or result is not difficult.4 These statistics can distort the weight a judge or jury might put on certain pieces of evidence. The ability to critically research statistical claims is vital to being able to effectively counter their assertions.
STATISTICS IN LEGAL RESEARCH
Statistics can be effectively used in oral arguments, contract negotiation, trial planning, and many other areas of legal practice.5 But before an attorney can use a statistic, they must first critically examine the data set used to create it and understand the limits of its application.
The emerging field of litigation analytics provides a good example of how attorneys can effectively review statistics. Using litigation analytics software, attorneys can look up the individual judge they are appearing in front of to see the statistical likelihood of that judge granting a motion to dismiss.6 This statistic is calculated by analyzing data on that judge’s prior motions to dismiss.7
To analyze the accuracy of that statistic, a lawyer should first look at how the data was collected and the number of data points used to create it. In regard to analyzing data collection, it is important to know what data was used to create the statistic as well as what data was left out. Many litigation analytics software applications only have data from federal courts, so a prior state court judge’s record would be completely left out of the statistical analysis. The omitted data has the potential to affect the statistic’s accuracy. Furthermore, if the data points used to create the statistic are too few, the results would not be statistically significant and would not provide useful information. For example, if a judge was appointed to the bench within the last year, it is probable they simply have not ruled on enough motions to dismiss to create statistically significant data.
Secondly, a lawyer should examine how the data is being presented. A judge might generally grant a motion to dismiss 53% of the time across all practice areas but might specifically grant motions to dismiss 72% of the time in immigration cases. How this statistic is presented can shape its interpretation and effectiveness. If an immigration attorney is told that the judge only grants 53% of motions to dismiss, that attorney would be making a decision based on inaccurate information for their specific practice area. The statistics used to inform decisions must be both relevant and narrowly tailored to an attorney’s question.
CONCLUSION
For the statistically literate attorney, it is crucial to never take a statistic at face value. It is only by first critically examining the creation and presentation of a statistic that an attorney can effectively engage with it. Whether it is to employ a statistic to inform trial preparation or to counteract opposing counsel’s use of statistics, statistical literacy is a vital skill in both modern legal research and legal practice.