Chicago Teachers Union v. Board of Education of the City of Chicago

UNITED STATES DISTRICT COURT

NORTHERN DISTRICT OF ILLINOIS

EASTERN DIVISION

CHICAGO TEACHERS UNION, LOCAL 1, )

AMERICAN FEDERATION OF TEACHERS, )

AFL-CIO; DONALD L. GARRETT JR.; )

ROBERT GREEN; and VIVONELL BROWN, )

JR., individually and on behalf of all similarly )

situated persons, )

)

Plaintiffs, )

) No. 12 C 10311

v. )

) Judge Sara L. Ellis

BOARD OF EDUCATION OF THE CITY )

OF CHICAGO, a body politic and corporate, )

)

Defendant. )

OPINION AND ORDER

In preparation for summary judgment briefing, both the Defendant Board of Education of

the City of Chicago (the “Board”) and Plaintiffs the Chicago Teachers Union, Local No. 1

(“CTU”), American Federation of Teachers, AFL-CIO, Donald L. Garrett Jr., Robert Green, and

Vivonell Brown, Jr., individually and on behalf of the class, have filed motions to exclude the

opposing party’s proposed expert testimony pursuant to Federal Rule of Evidence 702. The

Court assumes the reader’s familiarity with the background facts of this case, which the Seventh

Circuit’s class certification opinion more fully recounts. See Doc. 164. The Court certified the

following class in this case: “All African American persons employed by the Board of Education

of the City of Chicago as a teacher or para-professional staff, as defined in the labor agreement

between the Chicago Teachers Union and the Board of Education, in any school or attendance

center subjected to reconstitution, or ‘turnaround,’ in the 2012 calendar year.” Doc. 173. After

considering the parties’ arguments, the Court concludes the following. The Court allows

Trujillo’s opinions in full. The Court excludes Walker’s opinions in part and permits them in

part, as set forth in section II. Likewise, the Court excludes Blanchflower’s opinions in part and

permits them in part, explained more fully in section III. Finally, the Court allows Jacob’s

opinions in full.

LEGAL STANDARD

Federal Rule of Evidence 702 and Daubert v. Merrell Dow Pharmaceuticals, Inc., 509

U.S. 579 (1993), govern the admissibility of expert opinion testimony. See Bielskis v. Louisville

Ladder, Inc., 663 F.3d 887, 893 (7th Cir. 2011). Rule 702 provides that a witness qualified as an

expert by knowledge, skill, experience, training, or education may testify in the form of opinion

or otherwise provided that “(a) the expert’s scientific, technical, or otherwise specialized

knowledge will help the trier of fact to understand the evidence or to determine a fact in issue;

(b) the testimony is based on sufficient facts or data; (c) the testimony is the product of reliable

principles and methods; and (d) the expert has reliably applied the principles and methods to the

facts of the case.” Fed. R. Evid. 702. To admit expert testimony under this rule, the Court must

determine that (1) the witness is qualified, (2) the witness’ methodology is reliable, and (3) the

testimony will assist the trier of fact to understand the evidence or to determine a fact in issue.

Myers v. Ill. Cent. R. R. Co., 629 F.3d 639, 644 (7th Cir. 2010). The Rule 702 inquiry “is a

flexible one,” however. Daubert, 509 U.S. at 594. “Determinations on admissibility should not

supplant the adversarial process; ‘shaky’ expert testimony may be admissible, assailable by its

opponents through cross-examination.” Gayton v. McCoy, 593 F.3d 610, 616 (7th Cir. 2010).

The proponent of testimony bears the burden of proving that the proffered testimony meets these

requirements, and the Seventh Circuit grants the district court “wide latitude in performing its

gate-keeping function.” Bielskis, 663 F.3d at 894.

ANALYSIS

I. The Board’s Motion to Exclude Plaintiffs’ Expert Tina Trujillo [254]

Plaintiffs intend to present the testimony of Tina Trujillo, a professor of educational

policy at University of California, Berkeley, with a Ph.D. in education. Plaintiffs offer Trujillo’s

opinions to demonstrate that the Board could have implemented alternatives to turnarounds.

Trujillo’s work considers how educational reforms impact the educational environment in urban

schools. In her report, Trujillo analyzes the practice of school turnarounds and opines on

alternative reforms that are less discriminatory and more effective. Trujillo synthesizes

empirical evidence regarding turnarounds and related reforms and identifies five alternatives that

the Board could have selected.

The Board challenges Trujillo’s opinions on relevance and reliability grounds. The

Board does not take issue with Trujillo’s qualifications to testify as an expert; thus, the Court

will not address her qualifications. See United States v. Moore, 521 F.3d 681, 685 (7th Cir.

2008) (“A judge is not obliged to look into the questions posed by Rule 702 when neither side

either requests or assists.”). The Board argues that the Court should exclude Trujillo’s opinions

for the following reasons. First, Trujillo’s opinions are not based on sufficient facts or data

because she did not evaluate Chicago Public Schools (“CPS”) data. Second, Trujillo did not

apply reliable principles and methods because she synthesized other experts’ opinions without

setting forth her own methodology. Third, Trujillo proposed alternatives without indicating how

the Board could implement such alternatives. Lastly, the proposed alternatives are theoretical

and will confuse the jury.

A. Reliability

The Court first addresses the reliability of Trujillo’s opinions. Although “the district

court’s admissibility determination is not intended to supplant the adversarial process,” proposed

testimony must be “based on sufficient facts or data,” use “reliable principles and methods,” and

“reliably apply the principles and methods to the facts of the case.” Fed. R. Evid. 702. Daubert

provides a non-exhaustive list of factors for courts to use in this reliability analysis: “(1) whether

the theory can be and has been verified by the scientific method through testing; (2) whether the

theory has been subjected to peer review; (3) the known or potential rate of error; and (4) the

general acceptance of the theory in the scientific community.”1 Chapman v. Maytag Corp., 297

F.3d 682, 687 (7th Cir. 2002) (citing Cummins v. Lyle Indus., 93 F.3d 362, 368 (7th Cir. 1996)).

“To succeed on a disparate impact claim, plaintiffs bear the burden of showing that a

particular employment practice causes a disparate impact on the basis of race.” Allen v. City of

Chicago, 351 F.3d 306, 311 (7th Cir. 2003). Once Plaintiffs make this showing, the burden

shifts to the Board to show the practice is “job related” and “consistent with business necessity.”

Id. (quoting 42 U.S.C. § 2000e–2(k)(1)(A)(i)). If the Board makes this showing, the burden

shifts back to Plaintiffs to demonstrate that there was an alternative “which was equally valid and

less discriminatory that the [Board] refused to use.” Adams v. City of Chicago, 469 F.3d 609,

613 (7th Cir. 2006) (citation omitted). For Plaintiffs to succeed with this claim, the alternative

“must be available, equally valid and less discriminatory.” Id. (quoting Allen, 351 F.3d at 312).

Plaintiffs seek to rely on Trujillo’s opinions to establish this alternative.

The Board challenges Trujillo’s report for synthesizing other researchers’ opinions

without setting forth her own methodology. Throughout her report, Trujillo relies on studies by

other professionals in forming her opinions. Trujillo analyzes studies on the effectiveness of

1 Although Daubert was initially framed as applying only to scientific evidence, it applies broadly to all

“testimony based on ‘technical’ and ‘other specialized’ knowledge,” with Daubert’s reliability factors

applying flexibly depending on the specific issues presented by the testimony under consideration.

Kumho Tire Co. v. Carmichael, 526 U.S. 137, 141 (1999); see also id. at 150 (“[T]he factors identified in

Daubert may or may not be pertinent in assessing reliability, depending on the nature of the issue, the

expert’s particular expertise, and the subject of his testimony.”).

turnarounds and potential alternatives before discussing whether the Board could have

implemented those alternatives. The Seventh Circuit has explained that “review of experimental,

statistical, or other scientific data generated by others in the field” may suffice as a reasonable

methodology for an expert’s opinion. Clark v. Takata Corp., 192 F.3d 750, 758 (7th Cir. 1999)

(quoting Cummins, 93 F.3d at 369); See Walker v. Soo Line R.R. Co., 208 F.3d 581, 588 (7th Cir.

2000) (“[C]ourts frequently have pointed to an expert’s reliance on the reports of others as an

indication that their testimony is reliable.”). Here, Trujillo analyzes research that explains the

negative effects of turnaround schools, including research on Chicago’s reforms. Doc. 255-1 at

8–9. In proposing alternatives, Trujillo discusses how other school districts have implemented

her proposed alternatives. See, e.g., Doc. 255-1 at 18 (explaining how the Berkeley Unified

School District designed an integration plan to maintain balanced school diversity). The Board

argues that Trujillo includes data from school districts not comparable to CPS and criticizes

Trujillo’s reliance on case-study and opinion pieces by policy advocacy groups.

The Court will exclude expert testimony only if an expert’s reliance on the opinions of

others is “too speculative . . . or the underlying basis is faulty.” Walker, 208 F.3d at 588

(citations omitted). Trujillo’s reliance on other professionals’ opinions is not too speculative.

Indeed, Trujillo critically addresses turnaround policies, their impact on students, and more

effective alternatives. Although the Court agrees that much of Trujillo’s report relies on the

opinions of others, that reason alone is insufficient for excluding her report under Daubert. See

In re Yasmin & Yaz (Drospirenone) Mktg., Sales Practices & Prod. Liab. Litig., No. 3:09-CV-

10012-DRH, 2011 WL 6740363, at *6 (S.D. Ill. Dec. 22, 2011) (explaining that the expert was

“permitted to base his opinion in part on what other experts believe and [was] allowed rely on

the reports and studies of other experts”); Sanders v. City of Chicago Heights, No. 13 C 0221,

2016 WL 4398011, at *8 (N.D. Ill. Aug. 18, 2016) (permitting an expert’s testimony where the

expert relied on “a multitude of studies generated by scientific professionals” and “his own

experimental work and professional studies”). Additionally, Trujillo’s proposed alternatives are

more than mere speculation, as she explains the basis for each and includes studies that support

the feasibility of alternatives. See Smith v. Ford Motor Co., 215 F.3d 713, 719 (7th Cir. 2000)

(noting that although an expert’s opinion may take the form of hypotheticals, hypotheticals

“must themselves have ‘analytically sound bases’ so that they are more than mere ‘speculation’

by the expert”).

Moreover, throughout her report, Trujillo cites a substantial amount of professional

research, including her own. Plaintiffs contend that a majority of this research is peer-reviewed.

C.f. id. at 720 (“[L]ack of peer review will rarely, if ever, be the single dispositive factor that

determines the reliability of expert testimony.”). Trujillo extensively cites research on

turnarounds and alternatives for schools to implement. Trujillo evaluates the research about

turnarounds and their effectiveness before assessing whether the alternatives were available in

this case. Trujillo proposes five alternatives and explains why they were preferable and how the

Board could have implemented them. Although the viability of her proposed alternatives is not

completely apparent, the Court need not decide their viability for Daubert purposes. Such

arguments go to the weight of her testimony, not its admissibility. The Board will have an

opportunity to cross-examine Trujillo about her proposals’ viability and her use of data from

schools dissimilar to CPS. Daubert, 509 U.S. at 596 (“Vigorous cross-examination, presentation

of contrary evidence, and careful instruction on the burden of proof are the traditional and

appropriate means of attacking shaky but admissible evidence.”)

B. Assistance to the Trier of Fact

The Board next argues that Trujillo’s opinions will confuse the jury. Rule 702 “requires

that the evidence or testimony ‘assist the trier of fact to understand the evidence or to determine

a fact in issue,’” which “goes primarily to relevance.” Id. at 591 (citation omitted); Smith, 215

F.3d at 721. Under the relevancy determination, the Court is “limited to determining whether

expert testimony is pertinent to an issue in the case.” Stuhlmacher v. Home Depot U.S.A., Inc.,

774 F.3d 405, 409 (7th Cir. 2014).

The Board argues that Trujillo’s proposed alternatives were not viable options to the

Board between 2012 and 2014, and these theoretical alternatives will therefore not assist the trier

of fact. Plaintiffs respond that Trujillo’s opinions will assist the trier of fact in understanding the

effects of turnarounds and less discriminatory alternatives that also could improve academic

performance.

Here, whether the Board had available alternatives is a fact in issue. Trujillo identifies

the following less discriminatory and more effective alternatives: (1) expanding the district’s

turnaround framework to include multiple measures of effectiveness over time; (2) implementing

a district-wide desegregation plan; (3) reducing class size; (4) investing in early childhood

education; and (5) implementing full-service community schools. Trujillo’s opinions will assist

the trier of fact in assessing the availability of alternatives. The Board can challenge the viability

of these alternatives on cross-examination, as it is not clear whether these alternatives were

indeed available and feasible. See id. (“The question of whether the expert’s theory is correct

given the circumstances of a particular case is a factual one left for the jury to determine.”

(citation omitted)). Moreover, the cases that the Board cites to support its argument that

alternatives must be feasible are inapplicable. In Watson v. Fort Worth Bank & Trust, the Court

explained that the cost or other burdens of proposed selection devices were relevant for purposes

of the proposed alternative, but the Court did not discuss what an expert must establish. 487

U.S. 977, 998 (1988). Similarly, Gillespie v. Wisconsin only discussed the plaintiff’s burden in

establishing an alternative, not what an expert must demonstrate. 771 F.2d 1035, 1045–46 (7th

Cir. 1985).

The Court therefore denies the Board’s motion to exclude Trujillo’s opinion.

II. The Board’s Motion to Exclude Plaintiffs’ Expert Jonathan Walker [256]

Plaintiffs also retained Jonathan Walker to opine on whether the Board’s turnaround

policies had a disparate impact on African American employees in CPS. Walker has a Ph.D. in

Economics and is the President and Chief Executive Officer of Economists Incorporated.

Walker has experience as an expert in opining on alleged discriminatory conduct and other

employment-related topics. Here, Walker analyzes employee rosters from 2012 that included

employees’ school identification number, race, and union status. Walker also analyzes school-

level data for CPS elementary and high schools from 2008 through 2012, which include

information about each school’s performance metrics and probation status. Walker opines that

the Board’s application of its turnaround policies had a disparate impact on African American

employees compared to white employees. Specifically, Walker concludes that African American

employees were twice as likely to be impacted by turnaround as white employees at a

statistically significant level.

The Board challenges Walker’s anticipated testimony on relevance and reliability

grounds. The Board does not take issue with Walker’s qualifications to testify as an expert, thus,

the Court will not address his qualifications. In support of its motion to exclude Walker’s

opinions, the Board attaches a declaration of its economics expert, David Blanchflower, prepared

for this purpose. Plaintiffs challenge this declaration as offering new analyses and opinions after

the close of expert discovery in violation of Rule 26(a). Therefore, the Court proceeds to address

whether Walker’s opinions are relevant and based on reliable methodology and whether to

exclude Blanchflower’s declaration.

A. Blanchflower’s Declaration

The Court first evaluates whether it should exclude Blanchflower’s declaration. Plaintiffs

argue that the Court cannot consider Blanchflower’s declaration because it introduces new

analyses and opinions after the close of expert discovery in violation of Federal Rule of Civil

Procedure 26(a)(2)(B). Plaintiffs also contend that this declaration is prejudicial because they no

longer have an opportunity to conduct discovery into its analyses and opinions. The Board

responds that Blanchflower’s declaration is necessary to uncover Walker’s acts and omissions

for the trier of fact. The Board further contends that Plaintiffs are not prejudiced by the

declaration because they could have sought leave to depose Blanchflower regarding this

declaration.

“Under Rule 26(a)(2), a party that intends to rely upon an expert witness’s testimony is

required to furnish by a date set by the district court a report containing, among other

information, ‘a complete statement of all opinions’ the retained expert will provide, ‘and the

basis and reasons for them.’” Ciomber v. Coop. Plus, Inc., 527 F.3d 635, 641 (7th Cir. 2008)

(quoting Fed. R. Civ. P. 26(a)(2)(B), (a)(2)(C)). If a party fails to disclose information required

by Rule 26(a), it cannot use that information to “supply evidence on a motion, at a hearing, or at

a trial, unless the failure was substantially justified or is harmless.” Fed. R. Civ. P.

37(c)(1); see also David v. Caterpillar, Inc., 324 F.3d 851, 857 (7th Cir. 2003) (“[T]he sanction

of exclusion is automatic and mandatory unless the sanctioned party can show that its violation

of Rule 26(a) was either justified or harmless.” (quoting Salgado v. Gen. Motors Corp., 150

F.3d 735, 742 (7th Cir.1998)). “The determination of whether a Rule 26(a) violation is justified

or harmless is entrusted to the broad discretion of the district court.” Mid-America Tablewares,

Inc. v. Mogi Trading Co., 100 F.3d 1353, 1363 (7th Cir. 1996). The Court evaluates the

following factors to determine whether a Rule 26(a) violation is justified or harmless: “(1) the

prejudice or surprise to the party against whom the evidence is offered; (2) the ability of the

party to cure the prejudice; (3) the likelihood of disruption to the trial; and (4) the bad faith or

willfulness involved in not disclosing the evidence at an earlier date.” David, 324 F.3d at 857

(citing Bronk v. Ineichen, 54 F.3d 425, 428 (7th Cir. 1995)).

First, Blanchflower’s declaration prejudices Plaintiffs. The declaration introduces

analysis absent from Blanchflower’s rebuttal report. Walker wrote three reports in this case,

dated August 12, 2016, April 24, 2017, and May 24, 2018. Blanchflower submitted a rebuttal

expert report on October 9, 2017, addressing Walker’s first two reports. Expert discovery closed

on June 30, 2018. Through Blanchflower’s declaration, the Board attempts to introduce new

opinions—absent from Blanchflower’s rebuttal report—nearly a year after the close of expert

discovery. Blanchflower’s arguments regarding the flaws in Walker’s opinions relate to

Walker’s expert reports filed in 2016 and 2017. Walker’s 2018 report replies to Blanchflower’s

report. The Board does not provide any explanation as to why Blanchflower could not include

this analysis in his rebuttal report. Blanchflower had ample opportunity to raise these concerns

in his rebuttal report. See Blue Book Servs., Inc. v. Amerihua Produce, Inc., 337 F. Supp. 3d

802, 817 n. 16 (N.D. Ill. 2018) (excluding expert declarations and explaining that “[b]y

presenting new opinions in summary judgment declarations, neither side has an effective

opportunity to respond”). The Board sought no extension of the expert discovery deadline, and it

does not explain why its untimeliness is justified. The Board provides no reason for its delay,

and the Court finds the delay prejudicial to Plaintiffs.

Second, Plaintiffs cannot cure this prejudice. The Board suggests that Plaintiffs could

have sought leave to re-depose Blanchflower after he filed his declaration. This matter has been

ongoing since 2012. The parties have engaged in substantial discovery. Blanchflower wrote

eight reports in this case, one of which responded directly to Walker’s analysis. There is no

reason why Plaintiffs should have sought leave during Daubert motions to depose Blanchflower.

See Finwall v. City of Chicago, 239 F.R.D. 494, 501 (N.D. Ill. 2006) (“Late disclosure is not

harmless within the meaning of Rule 37 simply because there is time to reopen discovery.”);

Bowman v. Int’l Bus. Mach. Corp., No. 1:11-cv-0593-RLY-TAB, 2012 WL 6596933, at *3–4

(S.D. Ind. Dec. 18, 2012) (concluding that the plaintiffs’ untimely expert rebuttal reports were

not harmless because “reopening discovery to re-depose Plaintiffs’ experts would create

significant costs and would delay the resolution of the motion for class certification”).

Third, the Court has not yet set a trial date, which weighs against exclusion. However,

there is no reason to further delay this case. Hard Surface Sols., Inc. v. Sherwin-Williams Co.,

271 F.R.D. 612, 617 (N.D. Ill. 2010) (“If [no set trial date] were the determining factor, no court

could preclude expert or other testimony that was unseasonably disclosed contrary to the

discovery deadline dates set by the Court.”); see also Doe 1 v. City of Chicago, No. 18-cv-3054,

2019 WL 5290899, at *4 (N.D. Ill. Oct. 18, 2019) (explaining that the expenditure of

unnecessary time and resources did not render late disclosure harmless, regardless of whether a

trial date was set); Finwall v. City of Chicago, 239 F.R.D. 504, 507 (N.D. Ill. 2006) (rejecting the

party’s argument that there was more time for expert discovery since no trial date had been set).

As discussed, there was significant discovery in this case, especially with respect to

Blanchflower. Summary judgment briefing is forthcoming. There is no reason to delay this case

further to depose an expert. Finally, although it does not appear that the Board acted in bad faith,

the Board failed to justify its untimely disclosure. Accordingly, the Court excludes

Blanchflower’s declaration, except for Part III.2

B. Walker’s Opinions in Reports One and Three

Next, the Court addresses whether it should exclude Walker’s opinions. The Board

argues that Walker’s first and third reports are fundamentally flawed and misleading for three

reasons: (1) Walker erred by conducting a simple regression analysis; (2) Walker created an

unreliable hypothetical because the data did not support Plaintiffs’ desired result; and (3) Walker

erred by adopting Plaintiffs’ litigation position that all class members suffered an adverse action.

1. Regression Analysis

First, the Board argues that Walker’s methodology is flawed because he employed a

simple regression analysis that only controlled for race and ignored potentially explanatory

variables. “Regression analysis permits the comparison between an outcome (called the

dependent variable) and one or more factors (called independent variables) that may be related to

that outcome.” Manpower, Inc. v. Ins. Co. of Pa., 732 F.3d 796, 808 (7th Cir. 2013). Therefore,

an expert’s selection of independent variables for a regression analysis impacts the probative

value. Id. “While the omission of variables from a regression analysis may render the analysis

less probative than it otherwise might be, it can hardly be said, absent some other infirmity, that

an analysis which accounts for the major factors ‘must be considered unacceptable as evidence

of discrimination.’” Bazemore v. Friday, 478 U.S. 385, 400 (1986) (citation omitted). “[T]he

Supreme Court and this Circuit have confirmed on a number of occasions that the selection of

the variables to include in a regression analysis is normally a question that goes to the probative

weight of the analysis rather than to its admissibility.” Manpower, 732 F.3d at 808.

2 Part III of Blanchflower’s Declaration repeats explanations from his Rebuttal Report. This is not new

analysis, so the Court will consider these aspects of his rebuttal report. However, the Court will not

consider any analysis absent from the rebuttal report.

Here, the Board argues that Walker’s reports suffer from omitted variable bias and rely

on improper regression models because Walker did not consider the potentially explanatory

variable of academic performance. Instead, Walker evaluates whether there was a correlation

between race and selection for turnaround. Walker did not complete a multiple regression

analysis that considered academic performance. Plaintiffs contend this is because academic

performance was the facially neutral policy that disproportionately impacted African Americans.

Therefore, Plaintiffs claim, including the performance index in the regression model would

cancel a separate supposed racially disparate effect. Walker evaluates the rate that employees

and CTU members would be selected for turnaround based on race. Walker determines the

selection rate for employees based on race at the three stages of the Board’s decision-making

process. It is for the trier of fact to determine whether Walker’s failure to account for academic

performance in his regressions renders them less probative. See Adams v. Ameritech Servs., Inc.,

231 F.3d 414, 425 (7th Cir. 2000) (affirming admissibility of statistical analyses despite the

expert’s failure to run a multiple-regression analysis that would have isolated an independent

variable). Any limitations to Walker’s methodology will be comprehensible to the jury and the

Board can expose these through cross-examination.

The Court denies the Board’s motion to exclude Walker’s regression analyses in his first

and third reports.

2. Hypothetical

The Board next argues that Walker constructed an unreliable hypothetical that does not

use the data in this case. The Board contends that Walker built a model with a pre-determined

result based on a fixed and pre-determined equation. The Board also challenges Walker’s failure

to disclose the process by which he created this hypothetical. Plaintiffs respond that Walker

conducted the hypothetical analysis in response to Blanchflower’s conclusions, not to meet

Plaintiffs’ burden.

The proponent of expert testimony has the burden of proving that the proffered testimony

satisfies Rule 702. See Lewis v. CITGO Petroleum Corp., 561 F.3d 698, 705 (7th Cir. 2009).

The Court has “wide latitude in performing its gate-keeping function and determining both how

to measure the reliability of expert testimony and whether the testimony itself is reliable.”

Bielskis, 663 F.3d at 894 (citation omitted). Based on the Court’s review of Walker’s

construction of the hypothetical, the hypothetical is not the product of reliable principles and he

did not reliably apply it to the facts of the case.

In the hypothetical, Walker selects schools for turnaround at three stages. Walker applies

different criteria at each stage of the selection process to result in ten turnaround schools.

Walker creates his own method of selecting schools at various stages. Walker assigns each

school a randomly generated number. He then ranks schools based on the sum of the school’s

randomly assigned number, its 2010 percentage of possible points, and its 2011 percentage of

possible points. Then, Walker selects the ten lowest-ranked schools for turnaround. Plaintiffs do

not contend that Walker’s methodology is accepted amongst economists nor that his method is

peer-reviewed. Walker’s methodology is unreliable. See Bielskis, 663 F.3d at 894 (“An expert’s

opinion must be reasoned and founded on data.”); see also Smith, 215 F.3d at 718 (noting that in

evaluating whether expert testimony is reliable, the district court should determine if it “was

appropriate for [the expert] to rely on the test that he administered and upon the sources of

information which he employed.” (citation omitted)).

Moreover, this hypothetical will not assist the trier of fact. Walker does not replicate the

Board’s selection process. Instead, he creates his own selection process, which differs at every

stage, to discredit Blanchflower’s conclusion. Therefore, the hypothetical has nothing to do with

the Board’s actual selection process. Walker contends that he used an explicitly discriminatory

process to select schools to dispute Blanchflower’s model that controls for the percentage of

possible performance points in 2010 and 2011. Plaintiffs will have an opportunity to reveal any

flaws in Blanchflower’s analysis during cross-examination. A hypothetical that does not closely

reflect the Board’s selection process or clearly explain the decisions made at each level will not

assist the jury in its assessment of Plaintiffs’ claim or Blanchflower’s model. Moreover, any

minimal assistance it might provide would be substantially outweighed by its tendency to

confuse or mislead the trier of fact.

Accordingly, the Court excludes Walker’s hypothetical in Report 1, Figure 10.3

3. Adverse Employment Actions

The Board next argues that Walker’s assumption that all class members suffered an

adverse employment action when they received a termination notice following turnaround was

incorrect. Specifically, the Board argues that a materially adverse action should be limited to

whether a class member experienced a gap in employment, lost pay or benefits, or lost seniority

or pension service credit. The Board’s briefs seek to rebut Walker’s analysis by pointing to

Blanchflower’s conclusions on this point. Plaintiffs respond that the Court agrees with

Plaintiffs’ conclusion that the entire certified class suffered an adverse action when the Board

terminated them from their positions at turnaround schools.4 Plaintiffs also argue that the

3 Because the Court excludes this hypothetical, the parties’ arguments as to whether Walker disclosed the

process by which he created the hypothetical are moot.

4 Plaintiffs misrepresent the Court. The Court never defined adverse action for purposes of this case. At

the December 9, 2015 hearing, the Court rejected the Board’s argument that class members had to suffer

an adverse employment action. The Court explained that “issues regarding class members’ post-

displacement rights were not a problem with respect to certification for liability purposes, but, rather,

issues to be addressed when dealing with damages.” Doc. 259-13 at 4:4–9.

Board’s terminations of class members from their jobs in turnaround schools were adverse

actions as a matter of law because the terminations triggered the statute of limitations.

“[A]n adverse action must materially alter the terms or conditions of employment to be

actionable under the antidiscrimination provision of Title VII.” Porter v. City of Chicago, 700

F.3d 944, 954 (7th Cir. 2012). “A materially adverse employment action is something ‘more

disruptive than a mere inconvenience or an alteration of job responsibilities.’” Nichols v. S. Ill.

Univ.-Edwardsville, 510 F.3d 772, 780 (7th Cir. 2007) (quoting Rhodes v. Ill. Dep’t of Transp.,

359 F.3d 498, 504 (7th Cir. 2004)). A materially adverse action may exist where there has been

“a termination of employment, a demotion evidenced by a decrease in wage or salary, a less

distinguished title, a material loss of benefits, significantly diminished material responsibilities,

or other indices that might be unique to a particular situation.” Porter, 700 F.3d at 954 (quoting

Crady v. Liberty Nat’l Bank & Tr. Co. of Ind., 993 F.2d 132, 136 (7th Cir. 1993)). But “not

everything that makes an employee unhappy is an actionable adverse action.” O’Neal v. City of

Chicago, 392 F.3d 909, 911 (7th Cir. 2004) (quoting Conley v. Vill. of Bedford Park, 215 F.3d

703, 712 (7th Cir. 2000)). For instance, a purely lateral transfer or a transfer that involves

neither a pay reduction nor more than a minor change in working conditions is not actionable

under Title VII. See id. at 911–12.

The Board argues that class members only suffered an adverse action if they experienced

a gap in employment, lost any pay or benefits, or lost seniority or pension service credit due to a

turnaround. Plaintiffs contend that the class members’ receipt of a termination notice upon

turnaround was an adverse action, thus, all class members suffered an adverse action. “In

discriminatory discharge cases, the plaintiffs’ injury coincides with the decision to layoff the

plaintiffs, not the actual termination date.” Draper v. Martin, 664 F.3d 1110, 1113 (7th Cir.

2011). Here, in selecting schools to turnaround, the Board decided to terminate all

corresponding teachers and paraprofessionals from their positions. The collective bargaining

agreement between the CTU and the Board provides that if a school is selected for turnaround,

tenured teachers move to a reassigned teachers’ pool and continue to receive a full salary and

benefits for one school year. A tenured teacher who does not find a new position within that

year receives an honorable termination, unless the Board extends that teacher’s time in the pool.

Paraprofessionals are eligible for the cadre pool where they can receive substitute assignments

and are paid per assignment.

In Ricks, the Supreme Court concluded that the alleged discrimination occurred at the

time the decision to deny tenure was made and communicated to the plaintiff. Del. State Coll. v.

Ricks, 449 U.S. 250, 258 (1980). The Court explained that “[t]he proper focus is upon the time

of the discriminatory acts, not upon the time at which the consequences of the acts became most

painful.” Id. (alteration in original) (quoting Abramson v. Univ. of Hawaii, 594 F.2d 202, 209

(1979)). Although Ricks made this distinction for purposes of evaluating when a claim accrues

for statute of limitations purposes, it is also instructive here. The Second Circuit determined that

“[t]he Supreme Court’s conclusion that a discrimination claim accrues upon notice of

termination, rather than upon the implementation of that decision, necessarily implies that the

notification of termination qualifies as an adverse employment action.” Shultz v. Congregation

Shearith Israel of City of New York, 867 F.3d 298, 305 (2d Cir. 2017). There, the court

concluded that a notice of termination qualified as an adverse employment action even if the

employer later rescinded the termination. Id. at 304–05. The court explained that “[e]ven under

the most optimal circumstances . . . termination of an employee is likely to give rise to bad

feelings and anxiety.” Id. at 307 (quotation omitted); see also Singletary v. Howard Univ., 939

F.3d 287, 300 (D.C. Cir. 2019) (“[T]he mere notice of termination is a cognizable adverse

employment action regardless of whether the employer follows through.”).

Another court in this district recently reached the same conclusion in a case involving

nearly identical parties, as well as expert reports by Blanchflower and Walker. See Chicago

Teachers Union, Local 1 v. Bd. of Educ. of City of Chicago, No. 12 C 10338, 2020 WL 43009

(N.D. Ill. Jan. 3, 2020). The court agreed with the plaintiffs’ argument that layoff notices were

adverse employment actions, even if class members found other positions and did not experience

an interruption of pay or benefits. Id. at *5. Although the court did not rule definitively on the

issue, it explained that the fact that some class members found new positions concerned damages

and did not nullify the adverse action. Id. Similarly, here, class members assert they were

discriminated against when the Board terminated their positions. Class members received

termination notices and were removed from their positions, even though they continued to

receive their full salaries and benefits. The Board placed class members in a reassignment pool

where they had to apply for and be hired into a new position with the Board. Therefore, class

members suffered an adverse action when they received termination notices, and the Board’s

assessment that only thirty-one class members suffered an adverse action is erroneous. See

Ricks, 449 U.S. at 258 (explaining that the filing limitations period commenced when the

plaintiff was denied tenure where his employment would expire a year from that date). Class

members’ subsequent employment is only relevant for damages.

The cases that the Board cites in support of its arguments are inapplicable here. In Ajayi

v. Aramark Business Services, Inc., the employer informed the plaintiff that she was going to be

demoted, which never occurred. 336 F.3d 520, 531 (7th Cir. 2003). The Seventh Circuit

determined there was no adverse action because it was an “unfulfilled threat, which result[ed] in

no material harm.” Id. In Lanza v. Postmaster General of the United States, the court concluded

there was no adverse action because the employee only received a warning of potential

discipline, and she could have remained at work without a demotion, suspension, or change in

pay. 570 F. App’x 236, 239 (3d Cir. 2014). Here, class members received termination notices

and moved to a reassignment pool. There was material harm because they did not know whether

the Board would hire them and they had to reapply for positions.

The Court denies the Board’s motion to exclude Walker’s assumption regarding when an

adverse employment action occurred.

C. Walker’s Opinions in Report Two

The Board also argues that the Court should exclude Walker’s second report because it is

based on unreliable principles and methods. Specifically, the Board alleges that Walker

excluded 85% of the data from his analysis and failed to disclose results contradictory to his

opinion. Plaintiffs respond that the Board misrepresents Walker’s analysis, and Walker’s

findings support Plaintiffs’ pattern or practice of discrimination claims.

As discussed, in determining whether an expert’s testimony is reliable, the Court focuses

on the expert’s methodology rather than the quality of the data used or the expert’s conclusions.

See Manpower, 732 F.3d at 806. “The critical inquiry is whether there is a connection between

the data employed and the opinion offered; it is the opinion connected to existing data ‘only by

the ipse dixit of the expert’ that is properly excluded under Rule 702.” Id. (quoting Gen. Elec.

Co. v. Joiner, 522 U.S. 136, 146 (1997)).

Here, Walker completed probit regressions to estimate the probability of a class member

receiving a termination notice or being terminated, based on Blanchflower’s definition, after

controlling for performance metrics. Walker prepared these figures in response to

Blanchflower’s opinions. According to Walker, figures X and XI demonstrate that the

performance metrics that Blanchflower states the Board relied upon to make turnaround

decisions were statistically insignificant. The Board criticizes Walker’s methodology as only

considering about 15% of the relevant data. Walker submitted a declaration in which he explains

that STATA automatically excludes a variable that perfectly predicts a sufficiently large number

of observations as well as observations that are perfectly predicted. Walker explains that, as a

result, STATA automatically excluded observations from high schools in 2008 and 2010 that

were not equal to twelve or fourteen years on probation, respectively, because they were perfect

predictors of being spared turnarounds. By criticizing the n value involved in Figures X and XI

of Walker’s second report, the Board challenges the data that Walker used, rather than his

methodology. This determination is best left to the jury. See Manpower, 732 F.3d at 807

(“[T]he selection of data inputs to employ in a model is a question separate from the reliability of

the methodology reflected in the model itself.”); see also Wipf v. Kowalski, 519 F.3d 380, 385

(7th Cir. 2008) (explaining that “in a case of dueling experts . . . it is left to the trier of fact . . . to

decide how to weigh the competing expert testimony”). Further, Walker includes revised figures

X and XI as an attachment to his declaration in which the n value shows all observations, that is,

there is no exclusion of variables. The Board responds that Walker erred in attempting to fix the

dropped observations because “he failed to report how ASIS reached exactly the same regression

coefficients and results as his probit regressions.” Doc. 272-1 at 9. The Board can address

these concerns on cross-examination.

Finally, the Board argues that the Court should exclude Walker as an expert because he

did not disclose the results that contradict his opinions relative to the 2010 high schools. The

Board accuses Walker of concealing results that supported the Board’s position, in violation of

Rule 26(a)(1)(B). The Board also cites one Seventh Circuit case, discussed below, in support of

its point. Rule 26(a)(2)(B)5 requires a party to disclose to its opponent any information

“considered” by its expert. See Fed. R. Civ. P. 26(a)(2)(B); Fid. Nat’l Title Ins. Co. of N.Y. v.

Intercounty Nat’l Title Ins. Co., 412 F.3d 745, 750 (7th Cir. 2005). In Fidelity, the court

explained that a testifying expert “must disclose and therefore retain whatever materials are

given to him to review in preparing his testimony . . . [b]ut he is not required to retain every

scrap of paper that he created in the course of his preparation.” Id. at 751. Here, the Board does

not allege that Walker failed to disclose materials that Plaintiffs provided him for review.

Instead, the Board argues that Walker did not disclose the results that report all of his

observations. However, these results relate to analyses that Walker does not put forth in support

of his opinion and are therefore not covered by Rule 26.

Therefore, the Court denies the Board’s motion to exclude portions of Walker’s second

report.

III. Plaintiffs’ Motion to Exclude the Testimony of the Board’s Expert David

Blanchflower [258]

The Board intends to present the testimony of David Blanchflower regarding whether the

turnarounds had a disparate impact on class members when accounting for academic

performance. Blanchflower has a Ph.D. in economics and is a professor of economics at

Dartmouth College. Blanchflower reviewed CPS employee data regarding school, race, age, and

gender and analyzed these data while considering school academic performance. The Board also

provided Blanchflower with information about the school characteristics that the Board relied

upon in selecting schools to turnaround, including school performance, graduation rates,

probation status, and percentage of students meeting or achieving ISAT composite standards.

5 The Court assumes that the Board intended to cite this Rule, as Rule 26(a)(1)(B), the rule it cites, lists

the proceedings exempt from initial disclosure.

Blanchflower prepared eight reports in this matter. In his first four reports, Blanchflower opines

that there was no statistically significant impact on African American teachers or

paraprofessionals when controlling for schools’ academic performance. Blanchflower also

opines that the data are consistent with the Board using school performance as its main criterion

in selecting schools to turnaround in 2012. In his fifth report, Blanchflower opines that when he

controls for performance points, there is no evidence of any race effects in selecting schools for

turnaround. Blanchflower also opines that there is no evidence of a significant difference in the

proportion of African Americans who were laid off, including when he accounts for rehiring

data. In his sixth report, Blanchflower opines that there was no pattern of discrimination against

African American CTU members in selecting schools for turnaround between 2008 and 2010.

Blanchflower’s seventh report is a rebuttal report to Plaintiffs’ expert, Walker. In his eighth

report, Blanchflower concludes that there is no statistically significant evidence of race

discrimination in selecting schools for turnaround in 2013 and 2014.

Plaintiffs do not take issue with Blanchflower’s qualifications to testify as an expert, thus,

the Court does not evaluate his qualifications. Plaintiffs challenge various aspects of

Blanchflower’s reports on reliability and relevance grounds. Specifically, Plaintiffs argue the

following: (1) Blanchflower makes impermissible legal determinations in opining on what

constitutes an adverse employment action, and those determinations conflict with the definition

of the certified class; (2) Blanchflower’s conclusion that there was no race discrimination is a

legal conclusion and is therefore inadmissible; (3) Blanchflower is a statistics expert and

therefore cannot opine on educational issues without statistical analysis; (4) Blanchflower cannot

testify regarding the turnaround hearing officer reports; and (5) Blanchflower’s statistical

analyses purporting to find no adverse impact improperly control for academic performance.

The Court proceeds to address whether Blanchflower’s opinions are relevant and based on

reliable methodology.

A. Blanchflower’s Opinions Regarding Adverse Employment Actions

Plaintiffs first argue that Blanchflower makes impermissible legal determinations by

defining and opining on what constitutes an adverse employment action. Plaintiffs also contend

that Blanchflower’s determination conflicts with the definition of the certified class in this case

because his adverse action analyses are too narrow.

Federal Rule of Evidence 704(a) provides that “[a]n opinion is not objectionable just

because it embraces an ultimate issue.” Fed. R. Evid. 704(a). Rules 702 and 704 “prohibit

experts from offering opinions about legal issues that will determine the outcome of a case.”

Roundy’s Inc. v. N.L.R.B., 674 F.3d 638, 648 (7th Cir. 2012) (quoting United States v. Sinclair,

74 F.3d 753, 757 n. 1 (7th Cir. 1996)); see also Good Shepherd Manor Found., Inc. v. City of

Momence, 323 F.3d 557, 564 (7th Cir. 2003) (“[E]xpert testimony as to legal conclusions that

will determine the outcome of the case is inadmissible.”). “[A]llowing a witness to testify as to a

legal conclusion may cause the jury to accord too much weight to that testimony, and may infer

that the jury should look to that witness for legal guidance.” Naeem v. McKesson Drug Co., 444

F.3d 593, 610 (7th Cir. 2006).

There is a fine line between legal opinions that impermissibly intrude on the jury’s role

and those that assist the jury in reaching its decision. The Advisory Committee Notes

accompanying Rule 704 explain that Rules 702 and 403 “afford ample assurances against the

admission of opinions which would merely tell the jury what result to reach” and “stand ready to

exclude opinions phrased in terms of inadequately explored legal criteria.” Fed. R. Evid. 704

advisory committee’s notes to the 1972 proposed rules. The Advisory Committee Notes explain

that the question of whether an individual had capacity for a will would be an impermissible

legal conclusion, whereas, the question of whether an individual “had sufficient mental capacity

to know the nature and extent of his property and the natural objects of his bounty and to

formulate a rational scheme of distribution” would be permitted. Id.

The Court agrees that Blanchflower cannot opine on whether Plaintiffs suffered an

adverse employment action. Adverse action has a specific legal definition and expert testimony

as to whether class members suffered an adverse action would qualify as an impermissible legal

conclusion. But the phrase “adverse employment action” only appears in Blanchflower’s sixth

report when he suggests that it is challenging for Plaintiffs to argue that class members suffered

an adverse employment action if they were rehired within a year. Plaintiffs instead focus on

Blanchflower’s evaluation of whether any class members incurred economic loss following the

turnarounds. For the reasons previously discussed, class members suffered an adverse action

when they received a termination notice. Blanchflower does not opine otherwise.6 Instead,

Blanchflower evaluates whether class members experienced a monetary loss following the

turnarounds. Blanchflower conducted multiple statistical analyses to conclude that there was no

statistically significant adverse effect on African American teachers in the rehiring and

continuing employment process. See Doc. 295-5 at 19. Blanchflower created a model to

determine the probability that an individual was laid off and not rehired. Blanchflower then

evaluates whether there was a statistically significant effect on African American CTU members

who were displaced from the turnaround schools in the rehiring and retention process. Based on

this analysis, Blanchflower concludes that when rehiring is considered, there are no race effects.

6 In its motion to exclude Walker’s assumptions regarding the meaning of adverse employment action, the

Board relies on Blanchflower’s report and states that “[t]he data shows that the percentage of class

members who suffered an adverse employment action as a result of the 2012 turnaround was 31/215, or

14.4%.” Doc. 257-1 at 16. The Board therefore incorrectly defines adverse action, but that is not relevant

for purposes of evaluating whether Blanchflower makes an impermissible legal determination.

These opinions are not improper legal conclusions. Blanchflower does not opine as to what

constitutes an adverse employment action but instead assesses whether there was an adverse

impact on African American class members in the rehiring process.

Plaintiffs also argue that Blanchflower used an improper data set in this analysis.

Plaintiffs claim that Blanchflower should have been limited to analyzing the certified class,

rather than a subset of that population. “The district court usurps the role of the jury, and

therefore abuses its discretion, if it unduly scrutinizes the quality of the expert’s data and

conclusions rather than the reliability of the methodology the expert employed.” Manpower, 732

F.3d at 806. Here, Plaintiffs criticize Blanchflower’s selection of data by arguing that he omitted

employees based on his legal conclusions. However, Blanchflower evaluates the effect of

turnarounds on the entire class in his first three reports. Blanchflower explains the results of

regression analyses, which first show that there is a significant and positive effect on the

probability of turnaround based on the school’s percentage of African American employees.

Blanchflower then adds a control variable to the regression—the percentage of possible

performance points the school achieved—to demonstrate that once he controls for school

performance, race has no impact. Blanchflower repeats the regressions to consider employee

data, rather than school data. Therefore, contrary to Plaintiffs’ suggestion, Blanchflower does

consider the entire class.

In his reports, Blanchflower proceeds to consider whether employees at turnaround

schools experienced an economic loss and whether race was a statistically significant variable.

For example, Blanchflower’s fifth report provides a table that displays the proportions by race of

individuals at turnaround schools who were laid off and not rehired. Doc. 259-5 at 18, Table 1.

Based on this table, Blanchflower concludes that “the regressions show no statistically

significant adverse effects on African-American teachers in the rehiring/continuing employment

process.” Doc. 259-5 at 19. Blanchflower did not omit class members’ data, as Plaintiffs

suggest. Rather, Blanchflower defined “economic loss” and evaluated it with respect to the

class. Additionally, in their First Amended Complaint, Plaintiffs allege that “Defendants

intentionally subjected Plaintiffs and the class of similarly situated employees who were

terminated in the summer of 2012 to unequal and discriminatory treatment and denied them the

right to re-apply or be reinstated because of their race.” Doc. 8 ¶ 212. Therefore, it is a fact in

issue whether the Board did not rehire class members because of their race. Blanchflower’s

rehiring analysis will assist the trier of fact in determining whether the class members impacted

by the turnarounds were subject to discriminatory treatment in the rehiring process due to their

race. This analysis is also relevant for damages purposes.

Therefore, the Court denies Plaintiffs’ motion to exclude Blanchflower’s opinions

regarding adverse employment actions.

B. Blanchflower’s Opinions Regarding Race Discrimination

Plaintiffs also argue that the Court should exclude as improper legal conclusions

Blanchflower’s opinion that there was no race discrimination with respect to the Board’s

turnaround decisions. The Board responds that Rule 704(a) permits an expert to opine on an

ultimate issue in the case.

As already noted, “expert testimony as to legal conclusions that will determine the

outcome of the case is inadmissible.” Good Shepherd, 323 F.3d at 564. The Advisory

Committee Notes to Rule 704(a) explain that expert opinions cannot “merely tell the jury what

result to reach.” Fed. R. Evid. 704 advisory committee’s notes to the 1972 proposed rules; see

Sinclair, 74 F.3d at 757 n.1 (“[Experts] cannot testify about legal issues on which the judge will

instruct the jury.”). Expert testimony is relevant if it helps the jury determine any fact at issue in

the case. Stuhlmacher, 774 F.3d at 409.

Here, Plaintiffs bring discrimination claims against the Board. As part of their prima

facie case, Plaintiffs must show that the Board discriminated against them because of their race.

Discrimination has a specific legal meaning. See Torres v. Cty. of Oakland, 758 F.2d 147, 151

(6th Cir. 1985) (“[T]he term ‘discrimination’ has a specialized meaning in the law and in lay use

the term has a distinctly less precise meaning.”); Burkhart v. Wash. Metro. Area Transit Auth.,

112 F.3d 1207, 1212 (D.C. Cir. 1997) (discussing the explanation in Torres that “the term

‘discrimination’ has a specialized legal meaning that is more precise than the lay understanding

of the term”). The Court will instruct the jury on the meaning of discrimination, so the jury can

conclude whether the Board discriminated against class members. Therefore, Blanchflower

cannot opine on whether the Board discriminated against employees nor whether there is

statistically significant evidence of racial discrimination. See, e.g., Doc. 259-3 at 7 (“[T]here is

no evidence that African-American employees were discriminated against in the Chicago

Board’s decision to select the ten schools that were approved for turnaround.”); Doc. 259-7 at 26

(“The evidence shows there is no pattern of discrimination in the Board’s turnarounds.”); Doc.

259-8 at 27 (“There is no statistical evidence, here, once again, of discrimination against

African-Americans.”); see also United States v. St. Bernard Par., No. CIV.A. 12-321, 2013 WL

1563242, at *6 (E.D. La. Apr. 12, 2013) (preventing an expert from providing a blanket opinion

on whether discrimination occurred due to an ordinance); Sheridan v. E.I. DuPont de Nemours &

Co., 100 F.3d 1061, 1071 (3d Cir. 1996) (“The role of determining whether the inference of

discrimination is warranted must remain within the province of the jury, because a finding of

discrimination is at bottom a determination of intent.”).

The Court will permit, however, Blanchflower’s opinions that omit this legal term and

instead explain the statistical significance of his findings. See, e.g., Doc. 259-4 at 4 (“[N]one of

the African American coefficients is significantly different from zero. Put it another way there is

no evidence that the African American coefficient is significantly different from the coefficient

for whites among CTU Members.”); Doc. 259-3 at 5 (“[A]ll nine of the coefficients on the

African-American variables in this table fail to achieve statistical significance. Hence we must

conclude that, ceteris paribus, which means all other things being equal, race has no impact once

school performance is controlled for.”). “The Supreme Court has long noted the importance of

statistical analysis ‘in cases in which the existence of discrimination is a disputed issue.’”

Chavez v. Ill. Police, 251 F.3d 612, 637–38 (7th Cir. 2001) (quoting Int’l Bd. of Teamsters v.

United States, 431 U.S. 324, 339 (1997)); Ameritech Servs., Inc., 231 F.3d at 423 (“statistical

evidence can be very useful to prove discrimination” in disparate treatment and disparate impact

cases). The Supreme Court “has accepted statistics in the form of multiple-regression analysis to

prove statutory violations under Title VII of the Civil Rights Act of 1964.” McClesky v. Kemp,

481 U.S. 279, 294 (7th Cir. 1987). Blanchflower’s analysis is helpful to the Board, as it weakens

Plaintiffs’ prima facie case. See Coates v. Johnson & Johnson. 756 F.2d 524, 544 (7th Cir.

1985) (“By including the factor and showing that the statistics no longer indicate discrimination,

the defendant responds to the particular proof used by the plaintiffs to establish their prima facie

case and thus raises a genuine issue of fact as to whether the apparent disparate treatment is in

fact due to discrimination.”).

There is an important difference between these two types of opinions. In the first,

Blanchflower provides a legal conclusion that will influence the jury. See Good Shepherd, 323

F.3d at 564 (explaining that the court properly excluded an expert witness’s testimony where

“[t]he proffered testimony was largely on purely legal matters and made up solely of legal

conclusions, such as conclusions that the city’s actions violated the FHAA.”); see also United

States v. Perkins, 470 F.3d 150, 158 (4th Cir. 2006) (“[C]onclusory testimony that a company

engaged in ‘discrimination’ . . . nearly always invades the province of the jury.”).

Comparatively, Blanchflower’s second opinion would assist the jury in reaching its conclusion

by explaining the statistical significance of race without including a legal conclusion about

discrimination.

Therefore, Blanchflower cannot opine as to whether the Board discriminated against class

members, but he can explain the statistical significance of his analyses.

C. Blanchflower’s Opinions Regarding the Educational Efficacy of

Turnarounds

Plaintiffs argue that Blanchflower is not an educational expert and therefore cannot opine

on whether turnarounds work without statistical analysis. Defendants respond that Blanchflower

did not opine on the value of the turnaround process from an educational standpoint but made

observations based on the data published by the Board.

Under Rule 702 and the Supreme Court’s instructions in Daubert, “a district court is

required to determine (1) whether the expert would testify to valid scientific knowledge, and

(2) whether that testimony would assist the trier of fact with a fact at issue.” Smith, 215 F.3d at

718 (quoting Walker, 208 F.3d at 586). It is the Court’s role to “keep experts within their proper

scope.” Id. (quoting DePaepe v. Gen. Motors Corp., 141 F.3d 715, 720 (7th Cir. 1998)).

Plaintiffs challenge Blanchflower’s ability to testify regarding the effectiveness of turnarounds,

not his qualifications overall. Therefore, the Court will evaluate Blanchflower’s conclusions

regarding the effectiveness of turnarounds to assess whether “he has the adequate education,

skill, and training to reach them.” Hall v. Flannery, 840 F.3d 922, 926 (7th Cir. 2016) (quoting

Gayton, 593 F.3d at 617).

Plaintiffs allege that Blanchflower broadly opines that turnarounds “work.” However,

Blanchflower’s analysis is more analytical than Plaintiffs suggest. Blanchflower evaluates

academic performance at schools before and after turnarounds, based on the criteria the Board

used to select schools for turnaround. Blanchflower opines that turnarounds seem to improve

performance. To reach this conclusion, Blanchflower relies on data published by the Board.

Blanchflower evaluates the data to assess whether there are changes in reading, math, and

science scores, attendance rates, and drop-out rates. Blanchflower also compares this

information to district-average improvement. Blanchflower has background and training as an

applied economist. He does not purport to have a background in education. Blanchflower’s

opinions are consistent with his qualifications, as he interprets the data before and after

turnarounds to assess whether changes occurred. Blanchflower does not opine generally about

the effectiveness of turnarounds. Rather, he looks specifically at the data relevant here to reach a

conclusion. Moreover, the fact that Blanchflower is not an education expert impacts the weight,

not the admissibility of his opinion. See Hall, 840 F.3d at 929 (“The fact that an expert may not

be a specialist in the field that concerns her opinion typically goes to the weight to be placed on

that opinion, not its admissibility.”).

Plaintiffs next argue that the Court should exclude Blanchflower’s opinions regarding

turnarounds’ effectiveness because he did not conduct any statistical analyses. However, there is

no requirement that an expert employ a certain method, so long as it is reliable. Here,

Blanchflower relied on his expertise to evaluate trends in the data. Thus, the Court must assess

“the soundness and care with which the expert arrived at [his] opinion.” Timm v. Goodyear Tires

N. Am., Ltd., 932 F.3d 986, 993 (7th Cir. 2019). In evaluating the effectiveness of turnarounds,

Blanchflower assessed raw numbers provided by the Board to evaluate academic performance.

Although a jury is capable of observing scores before and after turnarounds to determine whether

there has been a change, Blanchflower’s opinion will assist the trier of fact in this analysis.

Blanchflower does not merely republish raw data, as Plaintiffs suggest. Blanchflower interprets

the data and explains what the data show. Although it may have been preferable for

Blanchflower to conduct additional analysis on the data, this shortcoming does not render his

methodology flawed. The opinion is connected to the data by more than Blanchflower’s ipse

dixit. C.f. Joiner, 522 U.S. at 146 (“Trained experts commonly extrapolate from existing data.

But nothing in either Daubert or the Federal Rules of Evidence requires a district court to admit

opinion evidence that is connected to existing data only by the ipse dixit of the expert.”).

Blanchflower’s opinions are restricted to data analysis. See, Doc. 259-6 at 7–9. The only

exception is Blanchflower’s statement that “[t]urnarounds appear to work.” See Doc. 259-7 at 5.

This conclusion is divorced from his methodology. For these reasons, the Court denies

Plaintiffs’ motion to exclude Blanchflower’s opinion on turnarounds and academic performance,

except that he may not opine that “[t]urnarounds appear to work.”

Plaintiffs also argue that the Court should exclude Blanchflower’s proffered testimony

related to employee turnover at turnaround schools. Plaintiffs contend that Blanchflower does

not perform any scientific analysis but only sets out tables of raw data. The Board does not

respond to this argument, and it therefore waives its ability to contest this point. See Ennin v.

CNH Indus. Am., LLC, 878 F.3d 590, 595 (7th Cir. 2017) (noting that “[f]ailure to respond to an

argument generally results in waiver”); see also Cincinnati Ins. Co. v. E. Atl. Ins. Co., 260 F.3d

742, 747 (7th Cir. 2001) (finding that where a party fails to respond to a non-frivolous

dispositive argument, it “acquiesces, rightly or wrongly” to that argument). The Court excludes

Blanchflower’s opinions with respect to employee turnover at turnaround schools.

D. Blanchflower’s Review of the Hearing Officer Reports

Plaintiffs argue that the Court should exclude Blanchflower’s opinions relating to the

hearing officer reports. The Board responds that the hearing officer reports are the type of

records that experts reasonably consider, and hearing officer reports are independently

admissible under the business record exception to the rule against hearsay.

An expert may base an opinion on otherwise inadmissible evidence so long as “experts in

the particular field would reasonably rely on those kinds of facts or data in forming an opinion

on the subject.” Sansone v. Brennan, 917 F.3d 975, 982 (7th Cir. 2019) (quoting Fed. R. Evid.

703). But “[e]xpert testimony does not assist the trier of fact when the jury is able to evaluate the

same evidence and is capable of drawing its own conclusions without the introduction of a

proffered expert’s testimony.” Viamedia, Inc. v. Comcast Corp., 335 F. Supp. 3d 1036, 1064

(N.D. Ill. 2018) (quoting Matter of the Complaint of Ingram Barge Co., No. 13 C 3453, 2016

WL 3763450, at *10 (N.D. Ill. July 14, 2016)); see also Dhillon v. Crown Controls Corp., 269

F.3d 865, 871 (7th Cir. 2001) (“An expert must testify to something more than what is ‘obvious

to the layperson’ in order to be of any particular assistance to the jury.”); Taylor v. Ill. Cent. R.R.

Co., 8 F.3d 584, 585–86 (7th Cir. 1993) (explaining that it was proper for the court to exclude

expert testimony because “any lay juror could understand th[e] issue without the assistance of

expert testimony”).

Here, the Board proffers Blanchflower as an economic expert who concludes that race

has no impact on the Board selecting a school for turnaround once academic performance is

controlled for. The Board relies on Blanchflower’s opinion to demonstrate that the Board used

school performance, not race, in deciding which schools to turnaround. Blanchflower examined

the hearing officer reports to evaluate why the Board selected schools for turnaround in 2013 and

2014. Specifically, Blanchflower’s eighth report looks at the reasons the Board provided for

selecting schools to turnaround in 2013 and 2014. In discussing the hearing officer reports,

Blanchflower quotes extensively from the reports. Blanchflower lists a series of conclusions

based on his review. For example, Blanchflower repeats that turnaround schools had a long

history of being on probation and that there was a lack of progress. Blanchflower also states that

none of the transcripts mention discrimination against African American teachers or

paraprofessionals, despite ongoing lawsuits. Blanchflower’s report proceeds to summarize the

hearing officer reports before concluding that “the schools selected performed below the district

and network averages for percentage of performance points.” Doc. 259-8 at 11.

Blanchflower summarizes the hearing officer reports without independent analysis.

Therefore, Blanchflower’s expertise is unnecessary to evaluate the Board’s process in selecting

schools for turnaround. This is especially true where this is not an area of his expertise. See

Viamedia, 335 F. Supp. 3d at 1065 (excluding testimony where the expert interpreted documents

“as a matter of economics” but did not “undertake any expert assessment in arriving at [his]

conclusion”); Elorac, Inc. v. Sanofi-Aventis Canada Inc., No. 14 C 1859, 2017 WL 3592775, at

*27 (N.D. Ill. Aug. 21, 2017) (excluding expert testimony because “[n]o expert economic

analysis is necessary on [a] basic point, nor, in truth, did [the expert] perform any; his opinion

appears to be based simply on his review of [evidence], not a scientific analysis”). The trier of

fact is capable of assessing the hearing officer reports to determine how the Board selected

schools to turnaround. Additionally, the Board’s witnesses can testify about the criterion used.

Blanchflower’s opinion on this matter is unnecessary. In summarizing the hearing officer

reports, Blanchflower acts outside his proper scope of expertise.

The Court excludes Blanchflower’s opinions relating to the hearing officer reports, see

Doc. 259-8 at 5–11.7

E. Blanchflower’s Statistical Analysis Purporting to Find No Adverse Impact by

Controlling for the Board’s Stated Reason for Turnarounds

Plaintiffs argue that the Court should exclude Blanchflower’s statistical analyses that

control for performance points. Plaintiffs contend that Blanchflower’s analyses purport to find

no adverse impact by improperly controlling for school performance. Plaintiffs also claim that

Blanchflower “data mined” and improperly relied on facts outside the record because he

considered data that the Board did not consider in its turnaround decision.

“Multiple regression analysis is sometimes well suited to the analysis of data about

competing theories for which there are several possible explanations for the relationships among

a number of explanatory variables . . . . In a case alleging sex discrimination in salaries, for

example, a multiple regression analysis would examine not only sex, but also other explanatory

variables of interest, such as education and experience.” See generally Daniel L. Rubinfeld,

Reference Guide on Multiple Regression, in Reference Manual on Scientific Evidence 303, 305

(3d ed. 2011). In his first three reports, Blanchflower evaluates whether race was a significant

factor in selecting schools for turnaround. Based on the results of these regressions,

Blanchflower concludes that “there is a significant and positive effect on the probability of

turnaround based on a school’s percentage of African-American employees.” Doc. 259-3 at 4.

Blanchflower then proceeds to add an additional control variable: the percentage of possible

performance points a school achieved in 2011. Based on the subsequent regression,

7 If the Board intends to introduce the reports under Rule 803(6) during trial, the Court will address

arguments as to their admissibility at that time.

Blanchflower concludes that upon introduction of the control variable, the African American

variable never achieves significance. In this case, the Board argues that it used performance

points in deciding which schools to turnaround. Blanchflower adds the performance points

variable as a variable of interest in support of this position. Therefore, Blanchflower utilized

reliable methodology by including an explanatory variable of interest.

Further, an expert’s selection of variables to include in a regression analysis goes to the

probative weight of the analysis rather than its admissibility. See Manpower, 732 F.3d at 808.

Therefore, Plaintiffs can challenge Blanchflower’s use of the performance points variable on

cross-examination. See EEOC v. DHL Express (USA), Inc., No. 10 C 6139, 2016 WL 5796890,

at *3 (N.D. Ill. Sept. 30, 2016) (refusing to exclude an expert’s testimony where the expert used

certain control variables that the opposing party challenged because it went to the probative

weight); Kleen Prod. LLC v. Int’l Paper, No. 10 C 5711, 2017 WL 2362567, at *10 (N.D. Ill.

May 31, 2017) (explaining that although the exclusion of certain control variables raised

concerns about the accuracy of the expert’s analysis, it was for the jury to assess).

Second, Plaintiffs argue that Blanchflower erred by using 2010 performance points

because the Board did not rely on 2010 performance points in making its turnaround decision in

2012. Blanchflower adds performance points from 2010 as a control variable, after evaluating

the significance of 2011 performance points. The Board argues that it turned schools around due

to consistently failing academic performance. Therefore, the 2010 points variable could be

relevant to that argument. Again, Plaintiffs’ argument relates to Blanchflower’s selection of

variables, which goes to probative weight. Plaintiffs can challenge Blanchflower’s use of this

variable on cross-examination.

The same is true for Plaintiffs’ argument that Blanchflower engaged in data mining.

Plaintiffs argue that Blanchflower used a subset of data not pertinent to the facts of this case in

order to reach a result that there was no race discrimination. Plaintiffs indicate that Blanchflower

should not have controlled for attendance or ISAT/PSAE 2009 because the Board did not utilize

those metrics to select turnaround schools between 2008 and 2010. Plaintiffs cite

Blanchflower’s sixth report in support of this argument. Blanchflower explains that the purpose

of his sixth report is to determine whether there is any statistical evidence of a pattern of

discrimination against African American CTU members between 2008 and 2010. In the

regression to which Plaintiffs likely refer,8 Blanchflower’s analysis follows multiple steps. In

part ii, Blanchflower adds a variable for 2009 ISAT/PSAE to determine its effect. It is not

apparent that an economics expert would not reasonably rely on this variable. ISAT and PSAE

scores measure students’ achievement in reading, mathematics, and science. The scores

therefore relate to academic performance and academic trends, which are related to this case.

Plaintiffs can challenge this variable on cross-examination.

Therefore, the Court denies Plaintiffs’ motion to exclude Blanchflower’s testimony

purporting to find no adverse impact.

IV. Plaintiffs’ Motion to Exclude the Board’s Expert Brian Jacob [258]

The Board intends to present testimony of Brian Jacob, an economist with a Ph.D. in

public policy and a professor of education policy and economics at the University of Michigan.

Jacob evaluates the relationship between CPS’ school rating measures and other measures of

school effectiveness. Jacob opines that the CPS academic performance measures that best reflect

8 Notably, Plaintiffs do not cite a specific portion of Blanchflower’s report that involves “data mining.”

Instead, Plaintiffs broadly cite his sixth report and provide one example of the alleged facts outside the

record. Based on the example provided, the Court assumes that Plaintiffs refer to pages 42 and 43 of

Blanchflower’s sixth report, reflected in Table 6. Doc. 259-6.

a school’s productivity are positively correlated with CPS academic performance, measured by

student academic achievement. Jacob also opines that the summary measure of academic

performance on which the Board partially relied to select schools for turnaround is strongly

correlated with other measures of school quality. Jacob further opines that the schools selected

for turnaround in 2012 scored substantially lower than other schools on various measures of

school quality. Additionally, Jacob analyzes the reports of Plaintiffs’ Experts, Trujillo and

Baker.

Plaintiffs do not take issue with Jacob’s qualifications to testify as an expert, thus, the

Court will not evaluate his qualifications. Instead, Plaintiffs challenge Jacob’s anticipated

testimony on relevance and reliability grounds. Specifically, Plaintiffs argue that Jacob’s use of

school-level measurements is improper because the case is about the termination of teachers and

paraprofessionals. Plaintiffs also argue that Jacob’s opinion partially relied on a report that used

unreliable statistical methods, requiring exclusion. Therefore, the Court proceeds to address

whether Jacob’s opinions are relevant and based on a reliable methodology.

A. Jacob’s Use of School-Level Measurements

Plaintiffs argue that Jacob’s use of school-level measurements do not relate to the class

members’ employment consequences and will not assist the trier of fact. In his analysis, Jacob

examines the relationship of school rating measures on which CPS partially relied to select

schools for turnaround and other measures of school effectiveness. Jacob explains that he

selected variables to reflect aspects of school quality. For instance, Jacob evaluates teacher

attendance, teacher experience, and teacher evaluation data for all schools, schools placed on

probation, schools considered for turnaround, and the schools ultimately selected for turnaround.

See Doc 259-9 at 14–15. Based on this analysis, Jacob concludes that the schools selected had

substantially lower scores for instructional quality and supportive environment, as well as lower

scores for ambitious instruction and teacher evaluations. Jacob also evaluates the correlation

between the accountability measures used by CPS and alternative school quality measures (i.e.,

ambitious instruction and supportive environment). Jacob concludes that CPS’ accountability

measures reflect important dimensions of school quality and effectiveness.

Plaintiffs argue that Jacob erred by evaluating school-level data, rather than assessing the

individual teachers’ performance. In their complaint, Plaintiffs allege that in selecting the ten

schools for turnaround, Defendants targeted schools with large African American staff and

teaching populations, which resulted in the termination of a disproportionate number of African

American tenured teachers and staff. Doc. 8 ¶ 170. The Board contends that its turnaround

decisions were focused on improving students’ academic performance in chronically failing

schools. Here, the Board selected schools for turnaround based on school-level data and

terminated teachers at the schools selected. Plaintiffs challenge the Board’s selection of certain

schools for turnaround. Therefore, Plaintiffs’ claims in this case involve the Board’s school-

level decisions. Although individual class members felt the effects of the Board’s decisions, the

Board terminated class members due to its school-level decision. Therefore, Jacob’s analysis of

school-level data is entirely appropriate.

Moreover, Jacob’s analysis will assist the trier of fact. Plaintiffs broadly allege that

Jacob’s opinions will not assist the trier of fact because the Board’s turnaround policy did not

evaluate teachers and paraprofessionals’ performance in terminating them. Plaintiffs suggest that

the Board must provide a job-related reason for terminating individual class members rather than

looking to the overall school performance. However, it is for the trier of fact to determine

whether the Board’s decisions were job-related and consistent with business necessity. Jacob’s

report involves several regressions, which lead him to conclude that the schools selected for

turnaround were less effective than other schools. Plaintiffs will have the opportunity to

challenge Jacob’s use of school-level data on cross-examination.

The Court denies Plaintiffs’ motion to exclude Jacob’s opinions for using school-level

data.

B. Jacob’s Opinions Relying on the Consortium on Chicago School Research

Report

Plaintiffs next argue that the Court should exclude Jacob’s opinion regarding the

Consortium on Chicago School Research (“CCSR”) Report, disseminated to the Board prior to

its turnaround decisions. Plaintiffs contend that the CCSR Report uses unreliable statistical

methods and depends on unverified facts.

Plaintiffs seek to discredit the CCSR Report through their expert, Trujillo. However, as

explained with respect to the materials that Trujillo relies on throughout her report, the Court

only excludes expert testimony if an expert’s reliance on the opinions of others is “too

speculative . . . or the underlying basis is faulty.” Walker, 208 F.3d at 588 (citations omitted).

Here, Jacob critically analyzes the CCSR Report. In section VI of his report, Jacob discusses the

CCSR Report’s analysis and conclusions. In responding to Trujillo’s report, Jacob reviews the

CCSR Report to evaluate how student achievement would have differed absent turnarounds.

Jacob explains that the CCSR Report examined turnarounds in twenty-two CPS elementary

schools and fourteen CPS high schools between 1997 and 2010. Jacob details the research

design and how the authors applied it. Jacob even acknowledges that while the study is not a

“gold standard” randomized control trial, “it provides strong evidence to suggest that the earlier

turnarounds in Chicago were at least moderately successful.” Doc. 259-9 at 27. Therefore,

Jacob’s reliance on the CCSR Report was not speculative and the underlying basis is not faulty.

Plaintiffs will have the opportunity to challenge Jacob’s analysis of the CCSR Report on cross-

examination. See Walker, 208 F.3d at 589 (explaining that “[t]o the degree that [the expert]

might have relied on faulty information, the matter certainly could be explored on cross-

examination”); Manpower, 732 F.3d at 809 (“[A]n expert’s reliance on faulty information is a

matter to be explored on cross-examination; it does not go to admissibility.”).

Plaintiffs also argue that the Court should exclude Jacob’s statements that the Board

received the CCSR Report prior to the turnaround decisions because those statements are based

on what a Board attorney told him. However, Jacob makes no such statements. Jacob only

opines that “[h]ad I been a CPS administrator reading this report in January 2012, I would have

felt optimistic that school turnarounds had the potential to lead to meaningful changes in student

achievement.” Doc. 259-9 at 27. Jacob does not suggest that he has any knowledge as to

whether the Board reviewed the CCSR Report.9

The Court denies Plaintiffs’ motion to exclude Jacob’s opinions relating to the CCSR

Report.

C. Jacob’s Discussion of Meyer’s Development of the Value-Added

Measures

Finally, Plaintiffs argue that the Court should exclude Jacob’s discussion of Dr. Meyer’s

development of the value-added measures. Specifically, Plaintiffs move to exclude Jacob’s

opinion that “the process used to develop school value-added measures for Chicago was

evidence-based and adhered to ‘best practices’ in the field.” Doc. 259 at 26 (quoting Doc. 259-

9 The Court also rejects Plaintiffs’ contention that the Board failed to address Plaintiffs’ argument that

Jacob cannot testify with respect to dissemination of the CCSR Report, requiring exclusion of Jacob’s

opinions. Plaintiffs cite Cincinnati Insurance Company in support of this argument. As discussed,

Plaintiffs misrepresent Jacob’s statements. Their arguments are frivolous, and exclusion is therefore

unwarranted. See Cincinnati, 260 F.3d at 747 (explaining that failure to respond to an argument is not

automatically forfeiture if the argument is nondispositive or frivolous).

10 at 10). The Board responds that Jacob is qualified to testify about value added measures, he

relied on work conducted by the Value-Added Research Center at the University of Wisconsin-

Madison (“VARC”),10 and he applied his expertise to the facts of the case to reach this

conclusion.

Plaintiffs criticize Jacob for failing to disclose that “he relied on his own knowledge of

school value added measures generally and his ‘prior knowledge of the VARC’ to come to his

conclusion.” Doc. 259 at 26. It is well established that an expert can rely on his own knowledge

in forming an opinion. “Rule 702 specifically contemplates the admission of testimony by

experts whose knowledge is based on experience.” Smith, 215 F.3d at 718 (quoting

Walker, 208 F.3d at 591); see also Kumho, 526 U.S. at 156 (“[N]o one denies that an expert

might draw a conclusion from a set of observations based on extensive and specialized

experience.”). Therefore, Jacob was not required to disclose that he relied on his own

experience.

Plaintiffs next argue that Jacob did not cite the correct literature when opining that “the

baseline model chosen by VARC is standard in the literature.” Doc. 259 at 27 (quoting Doc 259-

21 at 77:21–78:26). In his deposition, Jacob explained that the literature to which he referred is

“literature in the field of educational statistics in creating value added measures.” Doc. 259-21 at

77:25–78:1. Again, Plaintiffs improperly challenge Jacob’s ability to opine based on his

expertise. The Court therefore rejects Plaintiffs’ argument.

The Court denies Plaintiffs’ motion to exclude Jacob’s discussion of the value-added

measure.

10 VARC developed value-added measures for CPS.

CONCLUSION

The Court denies the Board’s motion to exclude Trujillo’s testimony [254]. The Court

also denies Plaintiffs’ motion to exclude Jacob’s testimony [258]. The Court grants in part and

denies in part Plaintiffs’ motion to exclude Blanchflower’s testimony [258]. The Court excludes

Blanchflower’s legal conclusions about race discrimination, his opinions regarding employee

turnover, and his opinions regarding the hearing officer reports. The Court grants in part and

denies in part the Board’s motion to exclude Walker’s opinions [256]. The Court excludes

Walker’s opinions regarding the hypothetical in Report 1, Figure 10. The Court also excludes

Blanchflower’s declaration, which the Board filed in support of the its motion to exclude

Walker’s opinions, Doc. 257-1, Exhibit 7.

My

Dated: February 25, 2020 Ry

SARA L. ELLIS

United States District Judge

42