Opinion
Opinion By: Gregory D. Stumbo,Attorney General;Amye L. Bensenhaver,Assistant Attorney General
Open Records Decision
The question presented in this appeal is whether the University of Louisville violated the Open Records Act in denying English Professor Marc Bousquet's request to inspect "[t]he raw data of the English department's doctoral student satisfaction survey, conducted in May-June 2003." For the reasons that follow, we affirm the University's denial of Dr. Bousquet's request.
On behalf of the University, Open Records Officer William J. Morison denied Dr. Bousquet's request by letter dated January 26, 2004. Dr. Morison explained:
I learned Prof. Wayne Usui, chair of U of L's sociology department, administered this survey in association with the English Graduate Students Organization. At some point he will complete a summary report, which I assume the EGSO will make public. Prof. Usui is understandably concerned about the matter of the privacy of the students who completed the questionnaires. No one has seen the questionnaires but him, and he believes they should not be released. At my request he sent copies of the completed questionnaires to me.
I share his concern. The students participated in this survey with the clear understanding that their identities would not be disclosed. Under the statute's personal privacy exemption, and under the university's obligations pursuant to the federal Family Educational Rights and Privacy Act, I am denying your application to inspect them. 1
In later correspondence, Dr. Morison described the National Doctoral Program Survey Instrument, and the additions to that instrument suggested by the English Graduate Students Organization, and provided Dr. Bousquet and this office with a blank copy of the instrument.
Each form contains eleven pages. The first page is entitled "The National Doctoral Program Survey Instrument." 2 It explains the form and defines the terms "doctoral program" and "advisor". The next eight pages give students an opportunity to respond to multiple-choice sections and allow them to write out, in longhand, explanations of their responses. . . . These occupy all but the last two pages of the form. The last two pages are headed "EGO Survey Questions." The EGO students created this section, which was added to the existing instrument. It contains eight questions each consisting of one or more parts. Each question is followed by enough space to write a response of a few sentences. Some students wrote in longhand. Three students typed their responses. A few did not respond to this section.
If Dr. Bousquet were allowed to inspect these forms, read these narrative responses in the students' handwriting, or even their typed responses, I believe he could ascertain the identities of at least some of the respondents. . . It wouldn't be a stretch to figure out who wrote what, for a professor familiar with one or more students to recognize handwriting, writing style, and or the particular context of the remarks. So in this sense they very much contain "identifying information. "
I understand the students intended that their opinions would eventually be made available in a form that would guarantee their anonymity. Prof. Usui has been working on a draft of such a summary.
I believe very strongly that these records, the survey forms filled out by the students, or rather the portions of the forms containing the students' narrative responses, are exempt under the Kentucky Open Records Act because inspection of them by Dr. Bousquet would constitute an unwarranted invasion of the personal privacy of the students. Furthermore, if these records could be linked to their authors, and I strongly believe they could, they constitute student education records under FERPA, and as such cannot be disclosed in the absence of the students' written consent.
Dr. Morison provided both this office, and Dr. Bousquet, with an Excel spreadsheet developed by Dr. Usui, containing "the data in such a way that [t]he students couldn't be personally identified." He expressed that belief that the spreadsheet, coupled with the blank survey form, "provide and reflect the 'raw data' of the graduate students' survey."
On appeal, Mr. Everett C. Hoffman challenges the University's position on behalf of his client, Dr. Bousquet, noting that Dr. Bousquet:
? disputes the applicability of these exemptions to the records he has requested. He does not believe that the raw data from the student survey constitute "education records" or "directory information" as those terms are defined in 20 U.S.C. § 1232g(a)(4) and (5). Nor does he believe that these data contain "personally identifiable information" within the meaning of 20 U.S.C. § 1232g(b)(2) or "information of a personal nature where the public disclosure thereof would constitute a clearly unwarranted invasion of personal privacy, " as provided in KRS 61.878(a) [sic].
To the contrary, the questionnaire sought anonymous responses from doctoral students - and they should contain no identifying information whatsoever about the students who answered the questionnaire. Moreover, although students understood that the process was designed to preserve their anonymity, they also understood that their anonymous responses would be shared - certainly within the Department and publicly as well. Indeed, students understood that this survey was based on other surveys, the anonymous responses to which were posted on the Internet.
Amplifying on his client's position, Mr. Hoffman later asserted:
[T]he survey responses are not "educational records" within the meaning of FERPA, much less educational records containing "personally identifying information. " Under 20 U.S.C. 1232g(a)(4)(A), educational records are records that "contain information directly related to a student." Unlike grades, transcripts, disciplinary records, and medical information contained in students' files, these anonymous survey responses do not contain "information directly related to a student."
Nor is any of the information contained in the responses "personally identifiable information." Regulations promulgated by the U. S. Department of Education state as follows:
Personally identifiable information includes, but is not limited to:
(a) The student's name;
(b) The name of the student's parent or other family member;
(c) The address of the student or student's family;
(d) A personal identifier, such as the student's social security number or student number;
(e) A list of personal characteristics that would make the student's identity easily traceable; or
(f) Other information that would make the student's identity easily traceable.
34 CFR Sec. 99.3. Dr. Morison appears to argue that the handwriting of those students who did not type their responses - and even the phraseology of the responses, handwritten or not - satisfies paragraph (f) above in that it is "information that would make the student's identify easily traceable. "
He noted that earlier survey results had been made public, directing our attention to a website containing the 2001 Graduate Student Survey results from the University of Louisville's English Department 3 and emphasized that "there is no indication that the students intended the forms themselves to be kept 'confidential. '" Respectfully, we disagree with Mr. Hoffman and his client, Dr. Bousquet.
It is the decision of this office that the University of Louisville properly relied on KRS 61.878(1)(l), incorporating 20 U.S.C. § 1232g, the Family Educational Rights and Privacy Act, and KRS 61.878(1)(a) in denying Dr. Bousquet's request. Having reviewed the survey responses in dispute, we find that the handwriting that appears in the disputed responses renders them easily traceable by Dr. Bousquet to their student authors. We acknowledge that this is a close question, and that there is no existing legal authority supporting either party's position. However, we have discussed our analysis with the Family Policy Compliance Office 4 of the United States Department of Education whose representative, Program Specialist Bernie Cieplak, confirms that a determination of what constitutes personally identifiable information in an education record, making a student's identity easily traceable, is a matter of University discretion, and that the Department is not inclined to substitute its judgment for that of the University in the reasonable exercise of that discretion.
KRS 61.878(l)(k) permits an agency to withhold "[a]ll public records or information the disclosure of which is prohibited by federal law or regulation." This provision incorporates 20 U.S.C. § 1232g, the Family Educational Rights and Privacy Act, a federal statute which regulates access to "educational records." FERPA precludes the disclosure of personally identifiable student information to third parties in the absence of a parent or eligible student's prior written consent. Both the Kentucky Supreme Court and this office have recognized that FERPA operates as a bar to disclosure of education records, and that FERPA is incorporated into the Open Records Act by KRS 61.878(1)(k).
Hardin County Schools v. Foster, Ky., 40 S.W.3d 865 (2001); 99-ORD-217, inter alia.
20 U.S.C. § 1232g(b)(1) thus provides:
No funds shall be made available under any applicable program to any educational agency or institution which has a policy or practice of permitting the release of education records (or personally identifiable information contained therein other than directory information [meaning information relating to a student not normally considered confidential including the student's name, address, telephone listing, and date and place of birth]) of students without written consent of their parents to any individual, agency, or organization, other than [to certain enumerated officials and organizations, or in connection with certain activities]. . . .
The term "education records" is expansively defined at 20 U.S.C. § 1232g(4)(A) as "those records, files, documents, and other materials which (i) contain information directly related to a student; and (ii) are maintained by an educational agency or institution or by a person acting for such agency or institution." FERPA restricts access to "education records (or personally identifiable information contained therein). . ." on pain of withdrawal of federal funds. As noted above, among the definitions of "personally identifiable information" found at 34 CFR § 99.3 is "[o]ther information that would make the student's identity easily traceable. "
We find unpersuasive Mr. Hoffman's argument that the disputed records are not education records because they do not contain "information directly related to a student." In our view, a record containing the thoughts and impressions of a student relative to the program in which he or she is enrolled, though submitted under clearly understood terms of anonymity, is replete with information directly related to the student author. Such a survey is the instrument by which the student evaluates his or her program and measures his or her success in the program. To the extent that these surveys are "maintained by an educational agency or institution, or by a person acting for such agency or institution," they fall within the broad parameters of the definition of "education records" found at 20 U.S.C. § 1232g(4)(A).
Moreover, we find that the disputed surveys contain personally identifiable information that would make the responding students' identities easily traceable. A review of those surveys discloses that all but one contains at least some handwritten material, and it is this handwritten material which constitutes the responding student's "fingerprint," rendering his or her response easily traceable. We again find unpersuasive Mr. Hoffman's argument that an individual's handwriting is not personally identifiable or capable of making his or her identity easily traceable. Although we do not normally factor into our analysis the identity of the requester, Dr. Bousquet's identity is, on these facts, relevant because he has instructed the responding students and no doubt had occasion to review handwritten coursework and examinations, thereby rendering their identities even more easily traceable. We hasten to note, however, that it would not require a professional handwriting expert to identify a student by comparing a sample of his or her handwriting on a non-education record and the handwriting appearing on the disputed education record. We share the University's view that a student's handwriting is as much a unique personal identifier as a student's address, social security number, or physical description, and defer to the University in that view. For these reasons, we affirm the University of Louisville's denial of Dr. Bousquet's request on the basis of KRS 61.878(1)(l) and 20 U.S.C. § 1232g.
Additionally, we find support in existing legal authority for the proposition that disclosure of the disputed surveys would constitute a clearly unwarranted invasion of the privacy of those students who responded, and whose identities could, consistent with the analysis set forth above, be discerned based upon a comparison of their handwriting. In OAG 79-348, OAG 86-15, and 92-ORD 1145, this office recognized that the privacy interests protected by KRS 61.878(1)(a) are as much those of the evaluator as those of the person being evaluated, since the evaluator generally makes his [or her] evaluation with the understanding that it will be kept confidential. " 92-ORD-1145, p. 6. While it is true that it is a program, not a person, that is the subject of the student survey, and that there is conflicting evidence in the record as to what assurances of confidentiality were expressly given to the responding students, we find that the underlying rationale in the cited decisions of this office support nondisclosure. The candor of the participants was almost certainly contingent on their assurance of confidentiality. On this additional basis, we affirm the University of Louisville's denial of Dr. Bousquet's request.
If, of course, Dr. Bousquet solicits and receives the written consent of any of the participating students to the release of their surveys, or establishes to the satisfaction of the University that he has a valid educational interest in the surveys, 5 within the meaning of 20 U.S.C. § 1232g(b)(A), he will be afforded access to the underlying surveys he seeks. Until such time, he is entitled to inspect only those summaries of the surveys already in existence, i.e., the Excel spreadsheet, or soon to exist, i.e., Dr. Usui's summary report that has been, or will be, made available to the English Department's students, faculty, and staff.
Alternatively, Dr. Bousquet may appeal this decision by initiating action in the appropriate circuit court pursuant to KRS 61.880(5) and KRS 61.882. Pursuant to KRS 61.880(3), the Attorney General should be notified of any action in circuit court, but should not be named as a party in that action or in any subsequent proceeding.
Everett C. HoffmanSegal Stewart Cutler LindsayJanes & Berry PLLC1400-B Waterfront Plaza325 West Main StreetLouisville, KY 40202-4251
William J. Morison, Ph.D.University Archives and Records CenterUniversity of LouisvilleLouisville, KY 40292
Angela KoshewaOffice of General CounselRoom 206Grawemeyer HallUniversity of LouisvilleLouisville, KY 40292
Footnotes
Footnotes
1 While it is clear that he intended to rely on KRS 61.878(1)(a) and KRS 61.878(1)(l), incorporating 20 USC § 1232g, in denying Dr. Bousquet's request, we remind Dr. Morison that KRS 61.880(1) requires public agencies to "include a statement of the specific exemption authorizing the withholding of the record . . .". The University's response did not contain any statutory reference.
2 We note that the May 6, 2003 memorandum that was directed to the doctoral students provided assurances that their "responses will remain anonymous. . .".
3 Our examination of this website discloses that the results were not posted by the University of Louisville, but by the National Association of Graduate Professional Students, and that the sampling of narrative responses provided are not handwritten. While it is possible that NAGPS obtained the survey results from which the data was extracted and posted from the University, we are not persuaded that the University is foreclosed from asserting the applicable statutory exemptions, and denying access, to the 2003 survey results.
4 The Family Policy Compliance Office is the federal agency charged with interpretation and enforcement of FERPA.
5 The determination of whether a requester has a valid educational interest also rests within the sound discretion of the educational institution. See 02-ORD-132.