# Elementary statistics for jurors

I have been summoned for jury duty. I am conscious of the relevance of statistics to some jury trials. For example, the concept of "base rate" and its application to probability calculations is sometimes - perhaps always - relevant.

What statistical topics might a person in my situation usefully study, and what materials would be suitable for somebody with my background?

I have a "hard science" degree and so have some limited statistical knowledge, but my skills are rusty. I work full time and don't have a lot of time before my jury duty. So it would be appropriate for the answers to focus on elementary concepts, simple problem solving skills and their application to relevant problems (and the limitations of those concepts and methods, of course).

• There are 3 or so historical cases from the US and the UK presented very clearly in the book "The Drunkard's Walk: How Randomness Rules our Lives", by Leonard Mlodinow. They're mostly related to poor uses of conditional probability. One is the O. J. Simpson case, and another is the Sally Clark case, which is also covered at understandinguncertainty.org/node/545 Dec 10, 2012 at 10:56

I very much enjoyed reading Gerd Gigerenzer's book "Das Einmaleins der Skepsis" - I believe there are two English versions, Reckoning with Risk and Calculated Risks.

I think that could be a good brush-up in basic statistics which I'd recommend to everyone. What may be even more important in the context of a jury is that he gives examples of how to talk about statistical topics in a way that can be understood by (statistical) lay persons. And how to translate certain kinds of statements into something that can be understood by humans.

(I could point you to some other nice and relevant popular statistics books, but they are available in German only)

I don't think you should study anything, unless your goal is to be kicked off during the Voir Dire. Personally, telling lawyers that I am a psychometrician has gotten me removed from a few juries.

• I'd agree. It really depends on what kind of case you get. The only time I've been on jury duty and actually gotten picked for a jury, it was a case involving a traffic stop and the person claiming the policeman lied. No statistics at all, and we finally came to a conclusion based on an applicable law, which allowed us to make a judgement based on facts that both the person and the policeman agreed on. Aug 11, 2012 at 12:02
• You'd agree that you shouldn't study anything, but whether you should study anything depends on the kind of case you get? That sounds contradictory. The kind of case you get is not known in advance. All jurors on serious cases were, before the case started, far more likely to be on a traffic case. Once a case starts, it's too late to be trying to educate yourself. Aug 11, 2012 at 21:11
• +1 In fact, @CroadLangshan mentioning his degree in "hard science" or knowledge of some statistics (or even just know the jargon -- words like base rate, or Bayesian decision theory, or prosecutor's fallacy) when he has not been asked about it explicitly makes it very much more likely that he will be excused and returned to the jury pool room to await the next call. Aug 12, 2012 at 2:17
• Peter: I believe in England (where I live), the jury selection process is conducted by the Judge and court staff rather than barristers, and is very much "lighter" than in the United States, so my guess is that whether I have prepared for jury service in this way is not likely to be asked about -- en.wikipedia.org/wiki/Jury_selection Aug 16, 2012 at 20:09
• For the record: the case, back in 2012, was a relatively serious and long one -- about 3 weeks -- and neither the (British court) judge nor anybody else asked any juror any questions. He made it quite clear that in fact we'd better have good reasons if we thought we might be excused, which were to be explained to him personally and immediately at that moment (not to other lawyers). I'm not allowed to say what happened in the jury room, and will stick to that, but I can say that I do still hold the opinion that knowledge of elementary statistics is important in jury trials. Nov 27, 2018 at 11:24

I am not sure that specific statistical knowledge is crucial for jurors. Jurors need to understand the strength of evidence and decide what preponderance of the evidence and beyond a reasoanble doubt mean. These are subjective notions. It is up to the prosecution and the defense to present evidence and explain any statistical issues that affect the interpretation of the evidence.

• This is the wrong forum for debate, but I'll try to explain. I read in the press that some forensic scientists have said that fingerprints are infallible, DNA evidence has been misapplied in criminal cases as a result of failing to account for base rates, and expert legal evidence has been presented with the unjustified assumption that the events in question are independent. In that context, if the prosecution fails to address major issues that are relevant to a criminal case, so that "reasonable doubt" remains, the jury should not convict. To do that they must have the relevant concepts. Aug 11, 2012 at 21:08
• I don't think that's in contradiction with the subjectivity of "reasonable doubt" and the fact that it's up to the prosecution and defence to do the things you say they should. I think there's a line to be drawn somewhere between having an awareness and understanding of these issues and walking in with a statistics textbook and a determination to apply it to your traffic case. I'm already aware of the concept of a base rate, but I'd be surprised if you suggested that I should wipe it from my mind for the duration of any case -- but perhaps you would! Aug 11, 2012 at 21:10
• @CroadLangshan Jurors come into cases with whatever knowledge they have and they are free to use their knowledge to help judge a case. What I am suggesting is that there is not and there need not be a class in statistics given to every juror before they can serve on a court case. Aug 11, 2012 at 21:32
• Thanks for your comments. I had not considered the idea of requiring jurors to attend classes (in statistics or otherwise). The following is not sarcasm or anything similar: As with requiring voters to take very basic classes that anybody could pass with a little effort, I think it's an interesting and defensible idea. Aug 12, 2012 at 16:52
• While this is maybe meant to be reassuring for the OP, I don't see how it can be an on-topic answer. Also, IMHO reasonable doubt is not at all disconected from statistical concepts. Thirdly, getting off-topic as well: I'd say the main thing the defense has to do is look after the interests of the accused. And I don't see how this will necessarily mean explaining statistical issues. In contrast, I'd say that the mandate of the defense forbids a proper explanation of statistical issues if that would strenghen the case against the accused. (-1) Aug 20, 2012 at 22:34