National Ass'n of Criminal Defense Lawyers v. Chicago Police Department

JUSTICE GALLAGHER delivered the opinion of the court: This consolidated appeal involves requests for data made by plaintiff-appellant National Association of Criminal Defense Lawyers (NACDL) under the Illinois Freedom of Information Act (FOIA) (5 ILCS 140/1 et seq. (West 2006)). NACDL requested the data used to assemble a legislatively mandated study of eyewitness identification procedures by the Chicago police department, the Joliet police department and two other police agencies. The Chicago and Joliet Police departments tendered final administrative responses to the requests, declining to produce the majority of the data requested. NACDL filed suit against both agencies, seeking to compel production of the requested data. The parties in the Chicago case filed cross-motions for summary judgment. NACDL also filed a motion for summary judgment in the Joliet case. Both trial courts directed defendants to produce some of the requested data but ruled that the majority of the information sought was either protected from disclosure by FOIA’s law enforcement and privacy exemptions or was too burdensome for the agencies to produce. On appeal, NACDL contends that: (1) the law enforcement and privacy exemptions of FOIA do not bar disclosure of police data after all personal identifying information has been redacted, (2) the generic and conclusory statements in police affidavits are insufficient to satisfy the police agency’s burden of proof to show that an exemption applies, (3) the privacy exemption of FOIA does not bar the disclosure of photographs used in lineups after all personal identifying information has been redacted, and (4) the burden of redacting identifying information does not outweigh the public interest in obtaining the requested data. For the reasons that follow, we affirm the trial courts’ partial grant of summary judgment in favor of NACDL, reverse the trial courts’ grant of summary judgment in favor of the Chicago and Joliet police departments and remand for further proceedings consistent with this opinion. I. BACKGROUND In 2003, the Illinois General Assembly passed section 107A — 10 of the Code of Criminal Procedure of 1963 (725 ILCS 5/107A — 10 (West 2006)). This legislation directed the Illinois State Police to conduct a one-year pilot study in the field on the effectiveness of the sequential method for lineup procedures. Under the sequential method, a witness is shown lineup participants one at a time and must state whether the individual shown is the perpetrator of the crime before viewing the next lineup participant. Moreover, the lineup administrator must be someone who does not know which participant is the suspect. 725 ILCS 5/107A — 10(c)(2) (West 2006). The traditional lineup procedure involves the simultaneous viewing of all lineup participants, and the lineup administrator usually knows which participants are suspects and which participants are “fillers,” i.e., persons who are not suspects themselves but bear a physical resemblance to the suspect. The statute directed the Illinois State Police to select three police departments to participate in the study (725 ILCS 5/107A — 10(b) (West 2006)) and Chicago, Joliet and Evanston were selected. The study was conducted in 2004 and 2005. The program director of the Illinois study released a report of the study’s findings on March 17, 2006. The report concluded that the Illinois data did not bear out the research experiments that suggest that sequential, double-blind lineups produce a lower rate of known false identifications. Instead, the study found that the sequential, double-blind procedures resulted in an overall higher rate of known false identifications than did the simultaneous lineups. The five categories in particular for which the study concluded that the sequential, double-blind procedures may produce a higher rate of false identifications are: (1) child witnesses, (2) older witnesses, (3) cross-racial identifications, (4) multiple perpetrators, and (5) suspects who do not match the description because of a change in appearance. The Illinois study was one of the first to compare the two methods in the field. The results of the study received national publicity and have been criticized by social science researchers who have cited flaws in the study design and the failure of the report’s authors to submit the study for scientific peer review. NACDL sent Freedom of Information Act (FOIA) requests to the Illinois State Police and the three police departments designated as participants in the study. The FOIA requests sought the following information: (1) the procedures followed by investigating officers in the control group of cases which followed the traditional simultaneous lineup method, including documentation on whether each live lineup was a first viewing of a suspect by an eyewitness or if the eyewitness had previously identified the suspect in a photo lineup, (2) the training materials and records for police personnel participating in the study, (3) records regarding the retention of certain personnel in connection with the study, (4) the criminal court case numbers for each case included in the study together with the corresponding photos and recordings of lineups and all photos shown to an eyewitness during a photo spread or sequential photo array, and (5) the complete database of information used to generate the data tables in the report. The Chicago police department (CPD) denied the request, stating that it did not have some of the requested documents and claiming that the remaining documents were exempt from disclosure under the law enforcement exemption of FOIA (5 ILCS 140/7(1) (c), (l)(d) (West 2006)). The Joliet police department (JPD) disclosed some of the requested documents, stated that it did not have some of the requested documents, and claimed that the remaining documents were exempt from disclosure under the privacy and law enforcement exemptions of FOIA (5 ILCS 140/7(1)(b), (l)(c)(i), (l)(c)(viii) (West 2006)). The Evanston Police Department also denied the request, and the Illinois State Police produced some documents but declined to produce documents belonging to the various police departments. NACDL filed lawsuits in Cook and Will Counties against the four police agencies, seeking to compel production of the requested documents. The Evanston Police Department reached an agreement early on in the litigation with NACDL to provide the requested documents with redactions and is not a party to this appeal. The Illinois State Police remains a party in the Chicago case, but is taking no active role in the litigation and informed NACDL that it will comply with any applicable court order regarding the disclosure of the documents in its possession from the designated police departments. CPD filed a motion for summary judgment with supporting materials. CPD asserted that the requested documents were exempt from production because their release would (1) constitute an invasion of both public and personal privacy, (2) interfere with pending law enforcement proceedings, (3) obstruct ongoing criminal investigations, and (4) disclose information specifically prohibited from disclosure under federal and state law, citing sections 7(l)(a), (b) and (c) of FOIA (5 ILCS 140/7(l)(a), (l)(b)(v), (l)(c)(i), (l)(c)(vi), (l)(c)(viii) (West 2006)). Moreover, the research required to determine whether each matter remained pending would constitute an undue burden pursuant to section 3(f) of FOIA (5 ILCS 140/3(0 (West 2006)). In support of its motion, CPD submitted the affidavits of Officer Matthew Sandoval and Lieutenant James Gibson. Sandoval stated that he had conducted a random sampling of 10 cases used in the study and his sample research indicated that of the 250 investigations involved in the study, approximately half of the investigations were still open. Gibson submitted two affidavits. In the first affidavit, he stated that for open investigations, releasing the requested documents “could very well interfere” with the investigations. Suspects could become aware of the status of an investigation, what the police know and what evidence they have. Disclosure could also present a risk to the safety of witnesses and their families. In the second affidavit, Gibson stated that for open cases, even a redacted document could still possibly be linked to an ongoing investigation because although the crime may have occurred in a large city, the members of the small community in which the crime occurred could recognize the details of a specific incident and the safety of victims and witnesses could be compromised. Gibson further stated that if a redacted document was linked to an ongoing investigation, an at-large perpetrator could become aware of what evidence the police have against him and could learn about specific police investigative techniques. Gibson said there is no way to adequately redact these documents because no single employee possesses the knowledge and discretion to determine what information needs to be redacted. Gibson also stated that releasing photographs of lineup participants “could constitute a violation of privacy.” Finally, Gibson conceded that documents in closed cases could be released if the following information is redacted: (1) all personal identifying information of victims, witnesses and suspects, (2) the exact location of the incident, (3) the identities of individual lineup participants, and (4) the exact date, internal records division (RD) numbers, and other identifying codes. NACDL filed a cross-motion for summary judgment, explaining that the police data was necessary to advance an ongoing public debate about the problem of erroneous eyewitness identifications in criminal investigations. NACDL submitted the affidavits of Rob Warden, executive director of the Northwestern University School of Law Center on Wrongful Convictions, Norman Reimer, NACDL’s executive director, and professor Nancy Steblay, a social psychologist with expertise in eyewitness identification research whom NACDL retained as a consultant. Warden is a specialist in legal affairs, focusing primarily on wrongful convictions. He noted that in 59.3% of the documented wrongful convictions in Illinois since 1900, or 54 out of 91 known cases, eyewitness identification was the principal prosecution evidence. Warden stated that erroneous eyewitness identification is by far the most prevalent factor in wrongful convictions. Moreover, wrongful convictions entail enormous social costs, e.g., 601 years collectively behind bars for the 54 men and women known to have been wrongfully convicted and $39 million for taxpayers in just three civil rights actions stemming from these convictions. Warden explained that the results of the study have had a profound effect on public policy in Illinois, making it unlikely that the Illinois General Assembly will act to reform police identification procedures. Reimer provided examples of situations in which opponents of reforming lineup procedures have used the results of the study to support the use of traditional lineup procedures. Steblay stated that the results of the study contradicted decades of scientific research. The study has also been criticized in the scientific community for flaws in its design and methodology and a panel of researchers declared it unreliable as a basis for determining effective eyewitness identification procedures. Steblay further stated that she is acting as a consultant to NACDL without a consulting fee. Her role as a consultant is not to promote one lineup strategy over another, but to advocate for sound science particularly as it informs public policy. Oral arguments were scheduled on the parties’ cross-motions for summary judgment, and the court recommended that the parties consider settling the case under an agreement whereby the CPD would provide NACDL with the requested, redacted police records subject to a protective order prohibiting their dissemination. The parties were unable to reach an agreement; however, during settlement negotiations, NACDL learned that there had been a misunderstanding regarding the scope of its FOIA request. NACDL also learned that the researchers had not had access to all of the police records concerning the investigations that were included in the study but had been furnished with a form summarizing the outcome of the lineups and some limited documentation from individual investigations. CPD had interpreted the request to include only those documents to which the researchers had access (report data). This misunderstanding led to the filing of supplemental arguments and evidence by both parties. NACDL submitted a second affidavit of Steblay, in which she explained why a meaningful critique of the study required access to all of the records in order to determine whether important identification history had been excluded from the study. CPD submitted the affidavit of assistant corporation counsel Amber Ritter, in which she estimated that redacting the full police file for each investigation included in the study would take 170 hours. In the Joliet case, NACDL filed a motion for summary judgment that was identical to the motion it filed in the Chicago case. JPD responded, submitting the affidavits of Officer Robert Puleo and Deputy Police Chief Patrick Kerr. Puleo estimated that it would take 197 person hours to redact the 257 police files that had been included in the study. Kerr’s affidavit was similar to the Gibson affidavits that were submitted in the Chicago case. He stated that the disclosure of the requested documents “could very well” interfere with ongoing police investigations. He further stated that the release of the police reports would constitute an invasion of privacy of the victims and witnesses involved. In the Chicago case, the circuit court analyzed open and closed investigations separately, and also distinguished the report data from the remaining information contained in the police investigation files. With respect to the open files, the court determined that these files were exempt under section 7(1) (c) of FOIA (5 ILCS 140/7(1) (c) (West 2006)), relying on Gibson’s affidavits. For the closed files, CPD agreed to produce the report data with all personal identifying information redacted. This left three unresolved issues with respect to the closed files: (1) redaction of the faces from the photographs of lineup participants, (2) redaction of the RD numbers, and (3) production of the remaining information in the police investigation files. The court found that the privacy interests of the lineup participants outweighed the interests of NACDL and the public in viewing the photographs and ruled that CPD could redact the faces in the photographs. However, the court ruled that CPD could not redact the RD numbers because closed files are already in the public record so the release of a number that would allow NACDL to match a police file to a specific case would not constitute an invasion of privacy. Finally, the court ruled that the production of information in the police investigation files that was not part of the report data would impose a substantial burden on CPD, and that the public interest in the data did not outweigh the burden of redacting the documents. In the Joliet case, the circuit court held that the information in open cases was exempt from disclosure under sections 7(1) (b) and (c) ofFOIA (5 ILCS 140/7(l)(b)(v), (l)(c)(iii), (l)(c)(viii) (West 2006)). The circuit court granted NACDL’s motion for summary judgment in part and ordered JPD to produce the report data in the closed cases with all personal identifying information redacted. The court held that the photographs in the closed cases were exempt from disclosure under section 7(1)(b) of FOIA (5 ILCS 140/7(l)(b)(v) (West 2006)). Finally, the court held that JPD did not have to produce information from closed cases that was not part of the report data because the redaction of identifying information contained in those materials would impose an undue burden on the department. NACDL filed notices of appeal in the First District in the Chicago case and in the Third District in the Joliet case. NACDL then filed a motion in the Illinois Supreme Court to transfer the Joliet case to the First District and consolidate the appeals. The Illinois Supreme Court granted the motion on December 4, 2008. This consolidated appeal follows. II. ANALYSIS Summary judgment is proper when the pleadings, depositions, and affidavits demonstrate that no genuine issue of material fact exists and that the moving party is entitled to judgment as a matter of law. 735 ILCS 5/2 — 1005(c) (West 2004); State Farm Mutual Automobile Insurance Co. v. Coe, 367 Ill. App. 3d 604, 607, 855 N.E.2d 173, 176 (2006). We review de novo an order granting summary judgment. Jones v. Country Mutual Insurance Co., 371 Ill. App. 3d 1096, 1098, 864 N.E.2d 793, 795 (2007). CPD challenges this court’s jurisdiction over the Joliet case on the grounds that a denial of summary judgment is not a final and appeal-able order, citing Central Illinois Light Co. v. Home Insurance Co., 342 Ill. App. 3d 940, 795 N.E.2d 412 (2003). This case is distinguishable. In Central Illinois Light, the plaintiff filed an indemnification action against multiple excess liability insurers. The defendants filed nine motions for summary judgment, three of which were denied. The court concluded that it had no jurisdiction over the three denials because they were not final orders. Central Illinois Light, 342 Ill. App. 3d at 964, 795 N.E.2d at 433. “An order is final and appealable if it terminates the litigation between the parties on the merits or disposes of the rights of the parties, either on the entire controversy or a separate part thereof.” R.W. Dunteman Co. v. C/G Enterprises, Inc., 181 Ill. 2d 153, 159, 692 N.E.2d 306, 310 (1998). The only significant difference between the Chicago and Joliet cases is that the parties in the Chicago case filed cross-motions for summary judgment, while JPD merely filed a response to NACDL’s motion for summary judgment. Both lower courts ordered the disclosure of some of the requested documents and held that the remaining documents were protected from disclosure under various exemptions in FOIA. The order in the Joliet case states that “all matters in controversy have been resolved.” There was nothing else for the parties to litigate in the Joliet case. This court is not deprived of jurisdiction because the motion for summary judgment was partially denied and the court did not specifically enter judgment in favor of JPD. The denial of NACDL’s motion for summary judgment as to the majority of the requested documents operated as a judgment in favor of JPD. The court accepted JPD’s argument that the documents were exempt from disclosure and the order terminated the litigation between the parties. Indeed, appeal was the only recourse left to NACDL in its attempt to obtain the documents. Thus, we conclude that the order was final and appealable and this court has jurisdiction over the Joliet case. We now turn to the issues on appeal. FOIA declares that it is “the public policy of the State of Illinois that all persons are entitled to full and complete information regarding the affairs of government and the official acts and policies of those who represent them as public officials and public employees consistent with the terms of this Act. Such access is necessary to enable the people to fulfill their duties of discussing public issues fully and freely, making informed political judgments and monitoring government to ensure that it is being conducted in the public interest.” 5 ILCS 140/1 (West 2006). The legislature recognized that even with such a policy certain documents should not be disclosed, and it dedicated section 7 of FOIA to exemptions. This section provides, in relevant part: “(1) The following shall be exempt from inspection and copying: (b) Information that, if disclosed, would constitute a clearly unwarranted invasion of personal privacy ***. *** Information exempted under this subsection (b) shall include but is not limited to: (v) information revealing the identity of persons who file complaints with or provide information to *** law enforcement *** agencies; * * * (c) Records compiled by any public body for administrative enforcement proceedings and any law enforcement or correctional agency for law enforcement purposes or for internal matters of a public body, but only to the extent that disclosure would: (i) interfere with pending or actually and reasonably contemplated law enforcement proceedings conducted by any law enforcement or correctional agency; * * * (iv) unavoidably disclose the identity of a confidential source or confidential information furnished only by the confidential source; (v) disclose unique or specialized investigative techniques other than those generally used and known ***; (vi) constitute an invasion of personal privacy under subsection (b) of this Section; (vii) endanger the life or physical safety of law enforcement personnel or any other person; or (viii) obstruct an ongoing criminal investigation.” 5 ILCS 140/7 (West 2006). Finally, FOIA provides: “Requests calling for all records falling within a category shall be complied with unless compliance with the request would be unduly burdensome for the complying public body and there is no way to narrow the request and the burden on the public body outweighs the public interest in the information. Before invoking this exemption, the public body shall extend to the person making the request an opportunity to confer with it in an attempt to reduce the request to manageable proportions.” 5 ILCS 140/3(f) (West 2006). Public records are presumed to be open and accessible under FOIA and the exemptions are to be read narrowly. Lieber v. Board of Trustees of Southern Illinois University, 176 Ill. 2d 401, 408, 680 N.E.2d 374, 377 (1997). With that in mind, this court must address the following issues: (1) whether the circuit courts erred in determining that all data in the open investigations is protected from disclosure under the privacy and law enforcement exemptions and, if so, whether the undue burden exemption applies, (2) whether the circuit courts erred in determining that the faces in the photographic lineups should be redacted, and (3) whether the circuit courts erred in determining that redacting the documents from the police investigation files would impose an undue burden on the police agencies. NACDL first contends that the trial courts in both cases erred in holding that the affidavits submitted by CPD and JPD were sufficient to invoke the law enforcement and privacy exemptions in case files involving ongoing investigations. After the circuit courts in both cases ruled on this issue, this court issued an opinion in Day v. City of Chicago, 388 Ill. App. 3d 70, 902 N.E.2d 1144 (2009). In Day, this court held that affidavits submitted by Sandoval and Gibson (that appear to be nearly identical to the affidavits submitted by these same individuals in the instant case) were entirely conclusory and inadequate to sustain the city’s burden to show that the requested documents were exempt. Day, 388 Ill. App. 3d at 75, 902 N.E.2d at 1149. The case was remanded and the circuit court was directed to conduct an in camera review of the documents in order to determine whether the law enforcement or privacy exemptions applied. Day, 388 Ill. App. 3d at 80, 902 N.E.2d at 1153. CPD concedes that after Day, the affidavits submitted are insufficient to support its claim that the documents are protected under the law enforcement and privacy exemptions of FOIA. However, because the circuit courts ruled that the production of documents outside of the report data in the closed cases would be unduly burdensome, CPD asks this court to extend that rationale to all documents in the open investigations. If this court should conclude that production of the documents would not be unduly burdensome, CPD asks this court to remand for an in camera review of the documents to determine whether the law enforcement and privacy exemptions apply. To avoid undue burden on the circuit court, CPD suggests that the circuit court limit its review to those files that contain some marker indicating that production of the documents would obstruct the investigation, e.g., cases in which there is a living perpetrator still at large. CPD further suggests that the circuit court may limit its review to a sample of the documents. NACDL asks this court to reverse the circuit courts outright and order the police agencies to submit the requested data in open cases with redactions. NACDL contends that records in which all identifying information is redacted will not yield information that would threaten the safety of a witness or provide information to an at-large suspect. Alternatively, NACDL asks this court to require that the police agencies review the requested records, perform the redactions, and present to the court for in camera inspection any specific records for which they believe there is a reasonable possibility that disclosure even with redactions could obstruct or impede law enforcement. We agree with both parties that the affidavits submitted to the circuit courts were insufficient to satisfy the burden of showing that any statutory exemption applies. Moreover, for the reasons discussed below, we reject CPD’s argument that redaction of the open investigation files would be unduly burdensome. We do not agree that the order of the circuit court granting summary judgment in favor of the police agencies should be reversed outright, but we also do not think it is appropriate for the circuit court to conduct an in camera review of all documents in the open investigation files to determine whether an exemption applies. In Day, the police files in question involved a single criminal investigation. In the instant case, there are still potentially over 100 cases in which the investigations are ongoing. It is possible that there are some investigations in which production of the data could compromise an ongoing investigation, even with all identifying information redacted. However, we do not think such a scenario is likely in the majority of open investigations, nor will it be as easy to identify witnesses in redacted files as the Gibson affidavits suggest. The fact that witnesses, victims and perpetrators may all live in the same neighborhood and attend the same schools does not mean that the safety of a witness could be compromised if the documents are released. Both parties agree that the location of the crime and all personal identifying information and other unique identifiers will be redacted; therefore, it will be impossible to tell where the crime occurred and what individuals may have been involved in the majority of cases. On remand, the parties should presume that all requested documents in the open investigations are to be produced, subject to the agreed redactions. The burden is on the police agencies to demonstrate on a case-by-case basis specifically how a particular witness could be in danger or how an individual investigation could be compromised if a document is disclosed in which all identifying information has been redacted. Moreover, any affidavits submitted in support of a statutory exemption must conform to the level of specificity required in Day. However, we note that our supreme court has indicated that while summary judgment would be appropriate without in camera review if the affidavits show with reasonable specificity why the documents fall within the claimed exemption, in camera review is the most effective way for the public body to objectively demonstrate that the exemption claimed does, in fact, apply. Illinois Education Ass’n v. Illinois State Board of Education, 204 Ill. 2d 456, 470-71, 791 N.E.2d 522, 531 (2003). We reverse the orders of the circuit courts granting summary judgment in favor of the police agencies for documents in open investigations. We remand for the police agencies to review the files and determine which documents, if any, meet the statutory exemptions even after the agreed redactions are made. Those documents should then be presented to the circuit court for an in camera inspection to determine whether the exemptions do, in fact, apply. Next, NACDL contends that the circuit courts erred in holding that the faces in the photographic lineups are exempt from disclosure because their release would constitute an unwarranted invasion of personal privacy. NACDL argues that once all personal identifying information is removed from the photographs, the persons who typically act as fillers in lineups will have only a minimal privacy interest in the photographs. CPD argues that there is a stigma associated with “mug shots” and that the faces in the photographs could be recognized and identified even if all personal identifying information is redacted. To determine whether disclosure would constitute an unwarranted invasion of personal privacy, courts take into account (1) the plaintiffs interest in disclosure, (2) the public interest in disclosure, (3) the degree of invasion of personal privacy, and (4) the availability of alternative means of obtaining the requested information. Lieber, 176 Ill. 2d at 408-09, 680 N.E.2d at 378. NADCL’s interest and the public interest are essentially the same. As part of its analysis of the study, NACDL needs to assess the quality of the composition of the lineups in the study. This information is separate and distinct from the identification history and other information that will be analyzed from the report data and investigation files. A meaningful analysis of the study must include an assessment of the lineups that were actually used. Moreover, the actual photographs are the only means available to researchers to determine the composition of the lineups that were used in the study. On the other hand, the degree of invasion of personal privacy is insignificant. First, we reject CPD’s characterization of the photos as “mug shots.” The persons who act as fillers in lineups may be police officers, individuals in jail, or civilians. There is no way for anyone to tell by looking at the photographic lineup which category any given individual falls into when all identifying information has been redacted. Second, civilians and police officers who act as fillers have voluntarily consented to have their photographs viewed by strangers and possibly introduced as evidence in criminal cases and therefore can have no expectation of personal privacy in the photos. The only conceivable privacy interest at stake is that of the individuals in jail who have not necessarily consented to the use of their photographs in the lineups. However, even if it would be possible to determine which fillers are individuals who are or were in jail, the likelihood that someone viewing the photographs will recognize a photo and link that person to a separate case in which he or she may be a suspect or may have been charged is remote; therefore the degree of invasion of personal privacy does not outweigh the remaining factors. Thus, we conclude that the release of the photos with all identifying information redacted does not constitute an unwarranted invasion of personal privacy. We also note that the Evanston police department has released the photographic lineups with the photos intact and only the identifying information redacted. Finally, NACDL contends that the circuit courts erred in holding that the production of documents from the police files in closed cases outside of the report data would create an undue burden on the police agencies. NACDL argues that the circuit courts understated the public interest in disclosure of the records while overvaluing the burden that redaction would impose on the police agencies. CPD contends that NACDL is requesting the entire investigatory file for each case, including many documents that do not even relate to identification procedures, and states that redacting all of these documents would be a massive undertaking. Moreover, CPD argues that if it is required to produce the same information for the open investigations, the burden will include the redaction of the report data in addition to the investigative files. At oral argument, counsel for CPD estimated that redacting these files will take approximately 150 hours or 20 personnel days. Also, the individual responsible for making the redactions will need to consult with detectives on a case-by-case basis. NACDL responds that it is only requesting those documents in the investigation files that pertain to eyewitness identification, e.g., police records that directly describe the lineups and photo arrays or show prior contact between the eyewitness and the suspect. NACDL contends that these records will primarily be found in lineup reports and supplementary reports. As an initial matter, we note that FOIA contains a specific requirement for the party seeking to claim the undue burden exemption. “Before invoking this exemption, the public body shall extend to the person making the request an opportunity to confer with it in an attempt to reduce the request to manageable proportions.” 5 ILCS 140/3(f) (West 2006). Our review of the record does not disclose any evidence that the police agencies sought to confer with NACDL in an attempt to reduce the request prior to the circuit court’s ruling that the undue burden exemption applied to certain documents. There is also no evidence that the police agencies contacted NACDL to discuss ways to reduce the request before seeking to invoke the undue burden exemption before this court. CPD states in its brief that the burden on the agency “cannot be alleviated by narrowing the request,” however, it does not provide any support for this assertion. In fact, NACDL notes in its reply brief that the police agencies involved in this appeal never conferred with NACDL regarding their claim that compliance would be unduly burdensome and never attempted to explore whether there would be any way to satisfy NACDL’s request by producing only the relevant portions of the investigation files. For this reason alone, it appears that the police agencies have not complied with FOIA and cannot claim the undue burden exemption. However, we will still consider whether the circuit court erred in determining that the burden on the agencies outweighed the public interest. In order for the exemption to apply, compliance must be unduly burdensome, there must be no way to narrow the request, and the burden on the public body must outweigh the public interest in the information. 5 ILCS 140/3(f) (West 2006). As discussed above, the police agencies have never attempted to see if the request can be narrowed, and NACDL has stated in its brief that most of the additional information it is requesting from the police files can be found in two reports. NACDL further stated that, with dialogue between the parties, it believes agreements could be reached to narrow the range of documents to be redacted. Indeed, the Evanston police department was able to reach an agreement with NACDL to produce all of the requested documents with redactions. Moreover, the public interest at stake in obtaining these documents is significant, both for the people of Illinois and for people across the country who are considering the results of the Illinois study. Wrongful convictions on the basis of mistaken eyewitness identification impose a huge cost on society in addition to the cost imposed on the individual who is wrongfully convicted. One such cost that has not been highlighted in the affidavits or briefs is that if the wrong person is convicted, the actual perpetrator is still at large and continues to pose a danger to society. If an examination of the data used in the study confirms that the results were correct, it allows researchers to focus on other possible methods of reform for identification procedures. If, however, an examination of the data discloses flaws in the study methodology or design, reform based on the sequential, double-blind identification procedures is still a possibility in Illinois, as well as in other states that have been considering such reform. As emphasized in the affidavit submitted by Steblay, the controversial results of the study cannot be analyzed without access to the identification histories contained in the police investigation files. The importance of the public interest at stake is further demonstrated by an amicus curiae brief filed in support of NACDL by: The Innocence Network; the Northampton (Mass.) police department; Captain Kenneth Patenaude of the Northampton police department; Sergeant Paul Carroll (retired), formerly of the Chicago police department; Steven D. Penrod, Department of Psychology, John Jay College of Criminal Justice; D. Michael Risinger, Seton Hall University School of Law; Jon B. Gould, Chair, Innocence Commission for Virginia and Director, Center for Justice, Law & Society, George Mason University; Maurice Possley, freelance journalist; and Laura Spinney, freelance journalist. The brief argues that without the necessary scientific review of this study, which requires access to the requested documents, the development of research in this crucial field will be hampered, progress in dealing with the problem of mistaken eyewitness identification will be stalled, and, even more tragically, this stalled progress will lead to additional wrongful convictions based on erroneous eyewitness identifications. CPD cites several cases in support of its argument that the undue burden exemption applies. CPD’s reliance on these cases is misplaced. In Farley v. Worley, 215 W. Va. 412, 424, 599 S.E.2d 835, 847 (2004), the court held that in response to a proper FOIA request, “a public body has a duty to redact or segregate exempt from non-exempt information contained within the public record(s) responsive to the FOIA request and to disclose the nonexempt information unless such segregation or redaction would impose *** an unreasonably high burden or expense.” The court in Farley required the public body to provide the redacted material in response to the FOIA request. Farley, 215 W. Va. at 425, 599 S.E.2d at 848. In ACLU Foundation of Northern California v. Deukmejian, 32 Cal. 3d 440, 453, 651 P.2d 822, 830, 186 Cal. Rptr. 235, 243 (1982), the court held that the burden of segregating exempt from nonexempt material on 100 index cards was substantial and the utility of disclosure was questionable where the deletion of personal identifiers would make it impossible for the appellant to learn whether a particular person had been improperly listed as an associate of a criminal suspect. A request that is overly broad and requires the public body to locate, review, redact and arrange for inspection a vast quantity of material that is largely unnecessary to the appellants’ purpose constitutes an undue burden. American Federation of Government Employees, Local 2782 v. United States Department of Commerce, 907 F.2d 203, 208-09 (D.C. Cir. 1990). The requests at issue in American Federation would have required the public body to locate every chronological office file and correspondence file, internal and external, for every branch office, staff office, etc. and the information had little or nothing to do with the stated purpose. American Federation, 907 F.2d at 209. In the instant case, the request specifically targets the police files for those cases that were used in the study and the information requested is essential to a meaningful review of that study. We do not agree with the circuit court that “several weeks of full-time work by [CPD] personnel who need to possess a high level of knowledge and sophistication” is sufficiently burdensome to outweigh the important public interest at stake here. The results of this study have garnered nationwide attention on an issue of vital importance to our criminal justice system. Moreover, the circuit court’s order presumed that the police agencies would have to redact the entire investigatory file in each case. NACDL has stated that it does not need access to every document in each file. It has proposed ways to narrow the request and expressed a willingness to enter into dialogue with the police agencies in an attempt to further reduce the burden. We conclude that the burden on the police agencies of redacting any identifying information is not so excessive that it outweighs the vital public interest in the disclosure of these documents. Therefore, we reverse the orders of the circuit courts holding that production of the data from the investigative files in closed cases is unduly burdensome. We also reject CPD’s argument that production of the data from the investigative files in open cases would constitute an undue burden. We remand for the police agencies to produce the requested documents from the investigative files in open and closed cases with the agreed redactions after conferring with NACDL to reach a final agreement on narrowing the request to those documents that are relevant to its analysis of the study. III. CONCLUSION For the foregoing reasons, we affirm the circuit courts’ grant of summary judgment in favor of NACDL requiring production of the report data in the closed cases. We reverse the circuit courts’ grant of summary judgment in favor of the police departments protecting the report data and investigative files in the open cases and remand for further proceedings consistent with this opinion. We reverse the circuit courts’ grant of summary judgment in favor of the police departments protecting the investigative files in the closed cases. Finally, we reverse the circuit courts’ grant of summary judgment in favor of the police departments protecting the faces in the photographic lineups. No. 1 — 08—2073, Affirmed in part and reversed in part; cause remanded with directions. No. 1 — 08—3414, Affirmed in part and reversed in part; cause remanded with directions. O’BRIEN and NEVILLE, JJ., concur.