Law 792, §A—Empirical Methods Fall 2017

University of Illinois

College of Law

Professor Robert M. Lawless

Professor Jennifer K. Robbennolt Fall 2017

Empirical Methods in Law

Law 792, §A

Overview

An increasing number of legal and policy issues are turning on accurate empirical information. A few examples will illustrate the point. In Exxon Shipping v. Baker, 554 U.S. 471 (2008), the Supreme Court used empirical findings about the variance in punitive damage awards to announce a presumptive cap on punitive damages in admiralty cases. Regulators and lawyers relied on sampling techniques to identify the extent of problematic documentation in the 2011 “robo-signing” scandal in the mortgage foreclosure industry. The Seventh Circuit has considered whether the knowledge of the “least sophisticated consumer” can be shown through survey evidence for purposes of the Fair Debt Collection Practices Act. See DeKoven v. Plaza Assocs., 599 F.3d 578 (7th Cir. 2010). In the recent Supreme Court decision about the constitutionality of a particular type of lethal injection as a method for the death penalty, the Court extensively discussed the empirical literature on the death penalty including three citations to articles from the Journal of Empirical Legal Studies alone. See Glossip v. Gross, 135 S. Ct. 2726 (2015).

In view of these developments, a working knowledge of empirical methods ought to be among the professional tools of a well-trained attorney. An attorney trained in empirical methods will be more persuasive in the courtroom and the boardroom. For example, empirical training gives attorneys an understanding of the statistical tools that an expert witness might use. Also, given the prevalence of empirical methods in the social sciences, anyone considering a career in public interest or policy work will be well served by acquiring some basic quantitative skills. Finally, empirical scholarship is placing an increasing role in the legal academy especially in the United States but also all over the world, with entire journals and conferences devoted only to empirically based projects. Anyone considering an academic career (not to mention current law review editors) needs to be familiar with empirical methods.

Our purpose in this course is to equip you with knowledge of various empirical methods useful in investigating legal and policy issues. That knowledge will be both theoretical and practical. That is, we want you to understand the theory of statistical inference, such as why samples need to be taken in particular ways or how experiments ought to be framed. But we also firmly believe that to learn these topics fully, you also need to wrestle with the practicalities of these techniques by engaging in your own empirical project. So, we are hoping that the class will make you skilled consumers of empirical methods and, possibly, set you on the road to becoming skillful practitioners. Several of our students have gone on to expand their learning in this class in independent studies and dissertations. Some now even do empirical work as law school professors.

Both of us use empirical methods in our own legal scholarship. We are team teaching the class because we come from different experiences and specialties, and we hope our combined experience will give an overview of empirical methods in a wide variety of legal settings. Perhaps most importantly, we find this topic important and exciting.

Readings

The primary reading material is our textbook, co-authored with Emeritus Professor Thomas S. Ulen: Empirical Methods in Law (2d ed., Aspen 2016) (ISBN: 9781454875802). We worked hard to convince the publisher to make the second edition of the book available in paperback at a much lower price than the first edition. You should bring your book with you to class as we will often refer to problems, tables, and figures in the book. There are many differences between the first and second editions, and you should have a copy of the second edition.

There will also be a course blog on which we will post materials and supplement class discussion. The class blog is available at http://www.creditslips.org/emil. You should check the blog periodically and will be responsible for the content posted there. The class blog will be used to post any documents connected with the course (such as this syllabus). More importantly, the blog will be used as a supplement to class discussion. We have tried to set up the blog so it will not get picked up by any search engines or public directories. (So make a note of the URL.) There is no such thing as a private web site. Be sure not to post anything on the blog you would not want the whole world to read.

We also want to draw your attention to the Empirical Legal Studies blog at http://www.elsblog.org. The law professors who write that blog summarize interesting research, books, and new trends. Professor Lawless’ Credit Slips blog, available at http://www.creditslips.org, often contains empirical content on the topics of credit and bankruptcy. A popular blog that covers very advanced statistical issues is Statistical Modeling, Causal Inference, and Social Science at http://andrewgelman.com/. The blog Junk Charts at http://junkcharts.typepad.com has interesting commentary and examples about the effective and ineffective use of charts and graphs to convey empirical information.

Computers and Software

You may not use a laptop (or other electronic device) in class unless explicitly authorized. You should, however, bring your laptop with you to class as we will sometimes ask you to use them for in-class exercises. [For those of you interested in the empirical research on laptops and classroom learning, see Pam A. Mueller & Daniel M. Oppenheimer, The Pen is Mightier Than the Keyboard: Advantages of Longhand Over Laptop Note Taking, 25 Psychol. Sci. 1159 (2014) and Susan M. Ravizza, David Z. Hambrick, & Kimberly M. Fenn, Non-Academic Internet Use in the Classroom Is Negatively Related to Classroom Learning Regardless of Intellectual Ability, 78 Comp. & Educ. 109 (2014)].

You are required to have a copy of Microsoft Excel and a license for either SPSS or Stata, which are specialized statistical software packages. During the semester, you may need to use Excel and either Stata or SPSS to perform analyses on example datasets before you come to class or to engage in exercises that we do in class. You may be able to do some of the analyses you will need for these exercises and your course project in Excel, but the specialized statistical packages offer versatility and sophisticated analyses that Excel does not have. Professor Robbennolt tends to use SPSS, and Professor Lawless tends to use Stata. We both strongly believe in the superiority of the software we use. One of us is right.

As we write the syllabus, the University of Illinois WebStore has a student license for Stata SE at $50 (expires on 6/30/2018) and for SPSS at $70 (expires 8/1/2018). Prices subject to change. Your mileage may vary. Offer not good in the state of Connecticut. (Why is it always Connecticut?)[1]

Class Meetings

The class will meet on Mondays and Tuesday from 10:30 AM - 11:45 AM in Room J.

At this time, we do not anticipate having to cancel any classes during the semester. We expect you to attend every class. If there is a valid reason why you cannot make a class, please let both of us know. During the semester, we intend to assign exercises (to be announced no later than the end of the preceding class) or short quizzes on the reading material (which will be unannounced and at the beginning of class). These exercises and quizzes are intended to give you an opportunity to engage with the material, and they are discussed below in further detail.

As explained below, we will spend the last part of the course listening to student presentations on their empirical projects. It is important that you attend each of these days to offer feedback and comments to your colleagues. It is also the professional and courteous thing to do. Your attendance and participation at these presentations will be particularly important in assessing your class participation grade.

Grading

We will determine your course grade by six factors: (1) class attendance, (2) class participation, (3) compliance with the deadlines mentioned in the next section, (4) periodic exercises and quizzes about the reading, (5) the class project, and (6) a presentation of that project to the class. The project and presentation will count for 70 percent of your final grade; the other factors will account for the remaining 30 percent.

We will grade the papers based on the thoroughness of the literature review, the quality of the research design, the appropriateness of the statistical analysis, the effective presentation of the data (in tables, figures, and/or text),[2] the quality of the discussion (including tying the results to the literature, recognizing and analyzing any limitations of the project), the organization of the paper, and the quality of the writing (no typos or grammatical errors please). We expect polished work. In addition to the time spent designing and carrying out the project, you should read, revise, and edit your own work multiple times before you turn it in to us. For suggestions about how to implement an effective revision process, see http://www.dansimons.com/‌resources‌/Simons_on_writing.pdf. All papers must be footnoted, and you should follow the Blue Book or another accepted social-science citation form. Please note that all else equal it is generally preferable to use books or journals rather than internet sources.

Quizzes and Exercises: At the beginning of class we may provide you with one or more multiple-choice or short answer questions based on the assigned reading. The quizzes will be designed such that they should be able to be answered by someone who has read the assigned materially carefully. In addition to the quizzes, you may be assigned to write a short answer to one of the problems in the textbook or to attempt an analysis on a sample dataset you have been given. Typically, these assignments will be announced at the end of class and will be due before the beginning of the next class.

Class Projects

You are encouraged to form yourselves into teams of two people for the purposes of doing the class paper and the presentation. You may work independently. We caution you, however, that conducting an empirical project in the course of a semester is a challenging undertaking and we think that you will be better equipped to get the work done if you work in pairs. A separate document lays out the requirements for the topic of the class paper. The members of the team will receive the same grade on team material, such as the paper and class presentations, but we will give separate grades to individual team members for the other components of the class. You should let us know via e-mail who your partner is or that you will be working alone no later than the end of the day Tuesday, September 5. If you plan to drop the class, please do so in the first week, before groups are formed. We recommend finding a partner right away and getting to work on your topic.

Please note that there are two weeks in the schedule during which you will be expected to report on the progress in your project. On Tuesday, September 26, we will expect each individual or team to make a brief presentation of their research topic. Before making this presentation and by September 12, you should have consulted with one of us regarding your paper topic and received tentative approval for your topic. By September 15, you should have given us a one- to two-page summary of your research proposal along with a copy of the paper you propose to replicate and a bibliography that cites at least five references. (You should not limit yourself to only five references. Good final papers always refer to more than five sources.) Then on Monday October 23, each team will describe its data-gathering efforts. That is, you will explain the data you are using, how they help to answer the questions you are investigating, and any difficulties you have encountered in collecting your data. At the time of this presentation you will turn in a preliminary, electronic copy of your dataset as it stands on that date. After this presentation, you must meet with one of us regarding your research project.

These updates will not be graded, but we will take them very seriously. Empirical work requires regular progress. These are not projects that can be started and completed at the last minute. The updates are an opportunity for you to get feedback on your project. Also, we will meet with you as much as our schedules permit and your needs require, but there must be at least two meetings as discussed here. We will try to divide the class projects evenly between us based on the nature of the projects and our own expertise.

The last two to three weeks of the semester will be given over to your presentations of your papers. We expect these presentations to last, on average, 15 to 20 minutes with a required 5 to 10-minute Q&A period and to be professional. You must leave time for a Q&A period in your presentation. If you do not leave time for Q&A, you will be penalized on your grade. The exact amount of time for each presentation will depend on the number of students in the class and will, of course, be announced before the weeks of the presentations. A draft of your final paper is due at the time of your final presentation. We will read the paper and provide you with feedback.

The final paper and an electronic copy of your final dataset will be due in our hands by 5:00 p.m. on Friday, December 15. We think the final paper will have to be at least twenty pages in length except in the most unusual of circumstances. If you have made arrangements with us to get writing credit for the class, it is likely we will have required a longer paper. Unless you instruct us to the contrary, we may make use of your dataset for future class presentations, in future editions of our book, and in teaching materials for this class. We will, of course, make full attribution to you if we use your dataset. Please let us know of any limitations on the use of your data.