Empirical Legal Studies

"Checking the Basics": Regression Assumptions

Too often researchers neglect the basics that they learned in grad school (or, perhaps, as undergrads) regarding regression assumptions. A recent Statalist discussion (here) highlights some of the basics when it comes to standard OLS regression "diagnostics." The commentary also underscores that there is no single "correct" approach and that level-headed judgment is at a premium. Finally, as an early comment notes, searching on "regress postestimation" and "regress postestimation plots" in Stata's "Help" tab uncovers a treasure-trove of helpful commands.

The Topography of Civil Litigation and Plaintiff Attorneys' Role

In Litigation Migrants, Charlotte Alexander (Georgia State) draws on data from FLSA litigation to help develop a theory of "litigation migration." Seeking to explain trends and changes in FLSA litigation over time (and its geographic movement over time) and exploiting a sample of FLSA cases from 2000-2016, the author finds that "personal injury lawyers have migrated across case types to take on FLSA cases, that lawyers have migrated across geographic boundaries to take on FLSA cases in other jurisdictions, and that both types of migration have influenced FLSA case filing numbers.” What the author argues that the "FLSA boom provides a remarkable opportunity to study what John Coffee has called 'entrepreneurial litigation,' in full flower. Indeed, the litigation migrants studied here have remade the wage and hour landscape, recovering many millions of dollars for workers and adding to the list of worries that cause employers to lose sleep." From the data the paper forward's an attorney-centric account. " Paraphrasing Hyman and Silver (“It’s the [plaintiffs’ lawyers’] incentives, stupid,”), Alexander's account emphasizes plaintiffs’ lawyers as a key litigation driver. Accordingly, "statutory or regulatory [efforts], such as tort reform, to shut down those lawyers’ income streams may result in a caseload shift, and a flight to more remunerative specialties. Admittedly, the tort reform variable in the two-part regression model described above produced no statistically significant results. However, because data availability limited the study period to 2000-2016, the model may have failed to capture the true impact of tort reform that occurred earlier, such as Florida’s set of reforms enacted in 1999.” The paper's abstract follows.

"Civil law is enforced primarily via private litigation. One characteristic of private enforcement is that litigation levels tend to cycle between booms and busts. This Article builds a theory for explaining this fluctuation, proposing that plaintiffs' attorneys can be understood as economic migrants. Just as workers cross borders to find jobs, lawyers 'move' across case types and jurisdictions to find profitable claims, and case-filing numbers increase as a result. Using the recent wage and hour litigation boom as a case study, this Article paints an empirical picture of attorney migration and its influence on case filing numbers. Drawing on these analyses, the Article concludes by considering the optimal litigation level within our private enforcement system.

Importing (and "Destring"-ing) Data From Excel (Again)

Practical questions on how to import Excel data sets into Stata seem to follow a five-year (or so) cycle. While invaluable (and relatively inexpensive) data conversion software (e.g., StatTransfer) can mange most situations, how Stata "reads" imported Excel data that may include a blend of "alpha" and "numeric" data can trip folks up. A recent discussion on the Stata Listserv (here) on the utility of Stata's "destring" command may assist.

Online Opt-In Surveys, Sample Weights, and Accuracy

For some obvious and less-than-obvious reasons, polling in general and online opt-in polling in particular have come under increased scrutiny. Despite their increased popularity (owing partly to cost factors), a recent Pew Research Center Report notes that such surveys have "raised some concern within the industry because, while low participation rates pose a challenge for all surveys, the online opt-in variety face additional hurdles." In many instances researchers turn to various sample weighting adjustments in an effort to off-set sample biases. While the brief Report can speak for itself, one clear take-away is that: "Even the most effective adjustment procedures were unable to remove most of the bias." Interestingly, "when it comes to accuracy, choosing the right variables for weighting is more important than choosing the right statistical [adjustment or 'sample weighting'] method." Andrew Gelman's (Columbia--Statistics) comments on the report are found here.

Substantive v. Statistical Factors in Model Selection

In a recent Stata list discussion, a grad student investigating firms' leverage ratios raised a question about whether to run a fixed-effects or OLS model on some panel data and cluster. To assess his results' robustness to model specifications, he ran the data across an array of models (below).

The advice offered, however, emphasized that "the answer to your ... question comes from substantive finance considerations, not statistics or Stata." That is, in some instances the precise nature of your research question and context, in combination with research design and data structure factors, inform specification selection where various options are, from a purely statistical perspective, "not necessarily wrong."