Skip to content

Graphical communication for legal scholarship

Following my talk on infovis and statistical graphics at the Empirical Legal Studies conference, Dan Kahan writes:

The legal academy, which is making strides toward sensible integration of a variety of empirical methods into its scholarship, is horribly ignorant of the utility of graphic reporting of data, a likely influence of the formative influence that econometric methods has exerted on expectations and habits of mind among legal scholars. Lee Epstein has written a pair of wonderful articles on graphic reporting —

1. Epstein, L., Martin, A. & Boyd, C. On the Effective Communication of the Results of Empirical Studies, Part II. Vand. L. Rev. 60, 798-846 (2007).
2. Epstein, L., Martin, A. & Schneider, M. On the Effective Communication of the Results of Empirical Studies, Part I. Vand. L. Rev. 59, 1811-1871 (2007).

— but her efforts haven’t gotten the attention they deserve, and reinforcement, particularly at a venue like CELS is very important.

But the main issue there is just that graphic reporting dominates tables as a means of communicating information– something you touched on at the end (I am constantly referring people, incluiding authors whose papers I’m reviewing for peer-reviewed journals in law & in risk/science communication to Gelman, A., Pasarica, C. & Dodhia, R. Let’s Practice What We Preach: Turning Tables into Graphs. Am. Stat. 56, 121-130 (2002); also King, G., Tomz, M. & Wittenberg., J. Making the Most of Statistical Analyses: Improving Interpretation and Presentation. Am. J. Pol. Sci 44, 347-361 (2000)).

The more difficult points involve selection and integration of graphic reporting methods given the array of audiences legal scholars are likely to be targeting. These include (A) other scholars doing empirical legal studies, (B) other scholars who don’t use empirical methods and (C) lawmakers, judges, and lawyers. You sensibly said to me that the style of graphic reporting one uses depends on one’s purposes. For purposes of communicating with *any* of these constinuencies, the info-whatever USA Today stuff is not appropriate. But I don’t think there’s any *one* graphic reporting method that will always suit the purpose of communicating with all three of these groups of readers, and we often *are* writing for all three at once.

The nub of the problem, as I see it, is that the sort of graphic reporting you are extremely good at is aimed at communicating with (A). There is a lot that empirical legal scholars, particularly those who have been have lived through decades of the industrial-grade econometrics, need to learn in order to communicate with each other more effectively (in part b/c they don’t necessarily know *what* they *should* be communicating, or essentially what the most relevant statistical information is to make sense of their study results; one of the beautiful things about learning the art of graphic reporting is that it forces researchers to be reflective about data reporting and not approach statistical analysis in a mindless, robotic fashion).

But the graphic reporting that is ideal for them is often *not* the best for readers in the (B) & (C) classes. This is in part b/c the best graphics for (A) communicate certain concepts that (B) & (C) likely don’t understand. But it is also true b/c those readers, particularly ones in (C), are likely to be modest in numeracy, and likely to fail to understand or comprehend the significance of information that is reported in the graphics that are ideal for (A). Your excellent recent analysis of public opinion on capital punishment is a good example; the excellent Figures in that paper, some of which featured today in your presentation, would bounce ineffectually off the minds of essentially all (C) & of too many (B)’s as well (& not b/c they are dumb; they are smart, & motivated, which is why USA Today “get there attention” is not relevant at all).

I know you get this issue– you negotiated it very well, e.g., in Red State, Blue State. I think it would have been great for CELS attendees to have the chance to explore these issues w/ you in a concrete way, one focused on examples you have deal with & that we have faced. Also, I’d love to know what you make of the sort of research on communicating statistical information that Spiegelhalter, D., Pearson, M. & Short, I. Visualizing Uncertainty About the Future. Science 333, 1393-1400 (2011), features (I referred to this in our brief exchange).

I don’t have any great reply here; really, any good answer would require experience in communication that I don’t have. But I do think that well-chosen dot plots and line plot can do well.

I think that what really might work are some examples of the graph and the explanation alongside. You don’t have to be Chris Ware to know that a picture plus 1000 words is better than two pictures or 2000 words. But all too often we seem to demand of our graphs that they stand on their own.


  1. Dr Soul says:

    I believe that there are different types of intelligence (spatial, verbal, mathematic, musical, design, emotion and empathy, etc.) and people have different levels of each intelligence. Certain fields of endeavor, it seems to me, attract individuals with a certain intellectual predisposition. The law, in particular, attracts people who are good with words and logic. If these people were good with numbers, they might become engineers, statisticians, financial mavens or computer programmers, but they aren’t: they go to law school. The institution of law relies on words; the people who practice law successfully and so rise in institutional power (e.g., to become judges, legislators, etc.) come to value verbal communication (words) and so tend to devalue or disregard the visual communication of data (charts and graphics.) This is all just a way of trying to explain the phenomena the Mr. Kahan describes above.

  2. K? O'Rourke says:

    Interesting, especially because my recent thinking about the line up plot and how this makes hypothesis testing in contexts where it might be sensible – just a legal analogue. (i.e. although no one is ever completely innocent of some wrong doing, most of us would not want lots of people convicted of things they did not do)

    This should provide the background quickly – Graphical Inference for Infovis Hadley Wickham, Dianne Cook, Heike Hofmann, and Andreas Buja.