Case o' The Week: Ninth Dives into SoCal Pool -- Hernandez-Estrada and Jury Pool issues

Q:  What is it called when Chief Judge Kozinski peels away from the majority decision, pens a witty separate opinion, and points out the "absurdity" of the current Ninth Circuit law?

Bell curve, with each band equal to one standard deviation

A:  A Standard Deviation.

United States v. Hernandez-Estrada, 2012 WL 6054774 (9th Cir. Dec. 5, 2012), decision available here.

Players: Admirable challenge by San Diego Ass’t Federal Defender Michele McKenzie. Decision by Judge Andrew Hurtwitz. Concurrence by Chief Judge Kozinski, joined by Judge Watford.

Facts: Hernandez-Estrada moved to dismiss his 8 USC § 1326 indictment, alleging Fifth and Sixth Amendment violations and violations of the Jury Selection and Service Act of 1968 (“JSSA”). Id. at *1. Among other challenges, he argued that the juror source list underrepresented African-Americans and Hispanics, that the Southern District of California (“SD Cal”) improperly excluded jurors for insufficient fluency in English based on their responses to the jury questionnaire, and clerks improperly disqualified jurors whose English abilities were unclear. Id. The district court found no constitutional violations, found “technical” JSAA violations, and denied the motion to dismiss the indictment. Id. at *2. The district court did, however, recommend significant changes in the jury selection practices. Id.

Issue(s): “The question in this appeal is whether the United States District Court for the Southern District of California violated the [JSAA] or the Constitution in compiling its 2009 master jury wheel.” Id. at *1.

Held: “Although the Southern District departed from the requirements of the JSSA in several respects, we find no reversible error in the underlying conviction.” Id. at *1. “The Southern District Clerk's Office should not automatically disqualify individuals who express doubt about their English skills. Nor should it put off preparing AO–12s [ed. note: mandatory forms submitted by the clerk reporting on the composition of the master jury wheel] until litigation is filed. The district should take steps to remedy both of these issues, but neither merits relief in this case.” Id. at *5.

Of Note: If one used a standard deviation analysis, two or three standard deviations would make a social scientist worry that a district’s jury draw isn’t random. Id. at *8 (Kozinski, C.J., concurring). In Hernandez-Estrada, there are fourteen standard deviations for black jurors in the SD Cal. Id.

 “So there’s cause for worry.” Id. 

Why doesn’t the Court reverse, when there appears to be such a huge statistical flaw in the composition of SD Cal’s pool? Because the Ninth doesn’t use “standard deviation analysis” when examining pools – it uses “absolute disparity.” This, our Chief Judge persuasively explains, “makes no sense.” Id. The “absurdity” of a fixed, permissible disparity-deviation sum (here, 7.7%) is laid bare by the Chief, who suspects that a “statistician would laugh at our current methodology.” Id. CJ Kozinksi concedes a three-judge panel can’t fix this approach, “but an en banc court could, and perhaps should, take a fresh look at the issue.” Id. at *9.

How to Use: The SD Cal’s jury questionnaire included a question from an old statutory standard: whether a prospective juror could “read, write, speak and understand the English language.” Id. at *6. Jurors who answered “no” were kicked out of the pool by the clerk. Problem is, the current statutory requirement in the JSAA only requires a juror to be knocked if he or she “is unable to read, write, and understand the English language with a degree of proficiency sufficient to fill out satisfactorily the juror qualification form” or “if unable to speak the English language.” Id. at *5 (quoting 28 U.S.C. § 1865(b)(2),(3)). While the Ninth doesn’t find a “substantial” error that merits relief, it warns that “change is necessary” and “caution[s] other district to evaluate their own questionnaires, as this problem appears not to be unique to the Southern District.” Id. at *7. We should take up Judge Hurwitz’s invitation and dig out our district’s questionnaire – challenges may await.                                             

For Further Reading: Does the racial composition of a jury really make a difference in the outcome of a trial? 

Yep. 

A recent and extensive study found that all-white juries convict black defendants sixteen percent more often than white defendants. See description of Duke study here. 

Sobering study -- particularly when read in conjunction with Chief Judge Kozinski’s righteous concurrence in Hernandez-Estrada.

Illustration of standard deviation from http://upload.wikimedia.org/wikipedia/commons/thumb/8/8c/Standard_deviation_diagram.svg/325px-Standard_deviation_diagram.svg.png

Image, "Is Justice Color Blind" from http://today.duke.edu/showcase/mmedia/hires/jurystudy.jpg

Steven Kalar, Federal Public Defender N.D. Cal. Website at www.ndcalfpd.org

.