Law School Admission Council Investigation
PIPEDA Report of Findings #2008-389
May 29, 2008
* See Postscript for update.
Complaint under the Personal Information Protection and Electronic Documents Act (the Act)
1. The complainant, Mr. Eike-Henner Kluge, objects to the Law School Admission Council’s (LSAC) requirement that Canadian students provide a finger/thumb print in order to be able to take the Law School Admission Test (LSAT). In Mr. Kluge’s view, such a requirement is an infringement of the student’s right to privacy. He is also concerned that these prints, while in the hands of LSAC, may be accessed by U.S. authorities under the USA PATRIOT Act.
Summary of Investigation
What is LSAC?
2. LSAC was founded in 1947 to coordinate, facilitate and enhance the law school admission process. It is a non-profit corporation whose members comprise more than 200 law schools in the United States and Canada. Fifteen Canadian law schools are members. LSAC provides a number of services for its member schools to facilitate the law school admission process. All law school admission assessments and decisions are made by individual law schools according to their own admission procedures and the information that LSAC collects for each applicant’s file. All LSAC member schools require applicants to take the LSAT.
3. LSAC was incorporated under the laws of Delaware. Each law school has a vote with respect to the selection of a Board of Trustees. The Board of Trustees has 18 voting members. According to LSAC, typically one of the 18 is from a Canadian law school. There is also a President of the Canadian Law School Admission Officers who sits with the LSAC Board of Trustees, but who does not have a vote. This individual does, however, participate in Board discussions.
4. LSAC has four permanent standing committees—at least one member of each is from a Canadian law school. These committees are involved in developing policies. According to LSAC, Canada has a fairly strong presence and LSAC is alert to its Canadian members and issues.
5. All Canadian common law schools, except the University of Moncton, McGill University and the French common law program at the University of Ottawa, require LSAT marks for admission purposes.
6. LSAC states that it has no physical presence in Canada in that it has no office in Canada. It sends boxes to law faculties with information books and posters regarding the LSAT. According to LSAC, it is up to the universities to post these (which normally occurs). An individual, typically a member of the university’s registrar’s office, agrees to serve as the exam administrator, or Test Centre Supervisor (TCS). There is a signed agreement between the TCS and LSAC under which the TCS agrees to make a site available on certain dates and procures the room for the examination. LSAC signs an agreement with the facility for the room and sometimes pays for the room. The TCS is required to obtain proctors (invigilators), who help with the in-person registration process on the day of the exam and monitor the room during the examination. While the TCS receives an honorarium from LSAC, as do the proctors, there is no signed agreement between LSAC and the proctors.
7. A student wishing to take the LSAT registers for it prior to the exam by mail, online or telephone and pays a fee of $155 CDN (there is a late registration fee, as well as a fee to write a test at a nonpublished test centre). If students in Canada are paying by cheque or money order, they must pay in Canadian funds. If paying by credit card, the fee is paid in U.S. funds.
8. LSAC does not believe that its activities fall with the scope of the Act because it is not engaging in commercial activities. In support of this position, LSAC notes that, while “commercial activity” in a privacy context has not yet been commented upon by a court, the concept has been judicially considered in a business context and found to require a preponderant purpose of making a profit. LSAC states:
If one were to objectively consider LSAC, specifically its mandate and preponderant purpose in administering the LSAT, its status as a non-profit, non-stock organization, its membership and governance structure and the public policy aspect in providing an education-related mechanism to assess individuals seeking to enter a regulated profession to practice law, then, given no specific guidance in Canadian personal information protection legislation or case law, we believe and submit to you that LSAC is not engaged in commercial activities for the purposes of PIPEDA.
9. While LSAC maintains that it is not subject to PIPEDA, it nevertheless cooperated with the Office’s investigation.
Collection
10. The LSAT is a standardized test administered four times a year at designated testing centres throughout the world. According to LSAC, an average of 135,000 prospective law students take the test each year. In 2005, just over 7,000 LSATs were administered in Canada, approximately five per cent of the total.
11. At the time of registration, the student is given an admission ticket, which the student brings to the test centre on the day of the exam. There are two boxes on the ticket that are reserved for the candidate’s thumbprints, and a place for the candidate’s signature.
12. On examination day, there is an on-site registration process, during which the student brings the admission ticket, provides his or her photo identification and thumbprints, and signs the admission ticket. Part of the ticket, the Admission Ticket Stub, is removed from the portion that contains the prints (and other personal information detailed above). This stub is taken into the testing room by the student. The stub contains the student’s name, LSAT account number and Social Insurance Number (SIN).
13. The testing staff (the TCS/proctors) collects the print, which is in clear ink form, not digital. The print is not matched at the time. Existing prints are not at the site during the examination.
Use
14. LSAC states that prints may be used in the event that there is an allegation of impersonation. For example, if a score gain of a repeat test taker is suspicious, LSAC could review the prints and the handwriting on the tests. LSAC indicates that since the chief purpose of thumbprints is as a deterrent to the use of impersonator/expert test takers, it does not have the occasion to match thumbprints very often. In fact, since the implementation of its thumbprint policy, we learned that there have been no cases in which LSAC found it necessary to review test takers’ thumbprints. LSAC believes that this is because the printing policy is an effective deterrent to those who may wish to use an impersonator who is an expert test taker.
Notification
15. If the student registers by mail, he or she receives an information booklet along with the LSAT registration forms. Under the section entitled “Day of the Test,” test takers are informed that they must provide one current valid government-issued (not expired) photo identification (ID) bearing their signature. The ID presented must contain a recent and recognizable photo and visible signature on the same original form of ID. Acceptable forms of ID include passport, driver’s license or other government-issued ID.
16. On the next page, there is information regarding test centre regulations. This is where notification is given to students that they will be thumbprinted. It states:
Thumbprinting. Examinees must be thumbprinted at every LSAT administration. Your thumbprint is retained only as long as needed to ensure the authenticity of test scores and to protect the integrity of the testing process.
17. All registrants are required to complete an authorization document which states in part:
I understand that I am responsible for the completeness and accuracy of all the information on this form. I certify that I have obtained and read the LSAT and LSDAS Information Book either in print or online and agree to be bound by its terms?. (emphasis added by LSAC).
18. On the day of the test, candidates are given a copy of the “LSAT Candidate Information Sheet,” which outlines information regarding the test, prohibited items, misconduct, complaints, and so on. On the second page of the sheet, it states the following with respect to thumbprinting:
Thumbprinting
Thumbprinting of examinees is required at all LSAT administrations. Your thumbprint is retained only as long as needed to assure the authenticity of test scores and to protect the integrity of the testing process.
19. Test takers are not provided with any options to thumbprinting. If they do not provide prints, they cannot write the test. To the best of LSAC’s knowledge, there have been no refusals. Any refusals would typically be noted by supervisory staff, who note other “incidents,” such as the removal of certain prohibited items from the test centre. However, LSAC states that one candidate had questions about printing and wrote to LSAC. LSAC responded to him and did not hear further about it.
Retention of information and Safeguards
20. After the exam, admission tickets and answer sheets are sent to LSAC, where they are stored. The documents are stored in the United States, at an off-site warehouse that is rented by LSAC as the sole tenant. The warehouse is locked and has a security system in place. Within the warehouse, the information occupies a caged secure area, with a key entry. The building is only accessed when records need to be retrieved.
21. LSAC states that all documents collected by it relating to candidates, including admission tickets, are destroyed through a confidential and secure destruction process after five years. LSAC has policies in place regarding the retention and confidentiality of data.
22. As for access to records by authorities, LSAC’s policy and practice when receiving a subpoena for test-taker records is to notify the subject of the subpoena to give that person an opportunity to challenge it, and for LSAC to challenge the subpoena itself if it believes it is improper or overbroad. It should be noted, however, that under the USA PATRIOT Act, the organization is not allowed to inform the individual whose information is being subpoenaed.
Background to decision to collect thumbprints
23. LSAC first began taking thumbprints of candidates in the 1974-75 testing cycle as one security measure to protect the integrity of the LSATs. Since then, wherever the test is administered, every test taker has been required to give his or her thumbprint. Although anecdotal evidence suggests that thumbprints have not always been collected since 1974, LSAC believes that the practice has been implemented consistently.
24. According to LSAC, at the time the decision was made, there were concerns about impersonation as there had been some allegations of it, and thumbprinting was considered the most foolproof method of deterrence. LSAC indicates that nothing since that time has caused it to change its position on this point. The organization had considered photographs, but it was not considered as reliable. At that time, it was considered more burdensome by candidates. Moreover, photograph identification was not as widely available as it is today.
25. LSAC states that privacy implications were considered. Concerns had been expressed by candidates, and LSAC felt that these were addressed in that assurances were given that the collection was for limited purposes, that the information was not to be disclosed beyond those purposes, and that the information would then be destroyed. LSAC states that it has resisted requests that it felt were improper, such as if a university wanted access to the prints. However, if a request were made under the USA PATRIOT Act, LSAC would have no option but to comply.
26. LSAC maintains that impersonation continues to be a concern. It cited an incident in which someone in Toronto had sent out a solicitation asking for another person to take the test on his behalf. According to LSAC, this incident demonstrates that, although it believes the thumbprint requirement is an effective deterrent to impersonation schemes, the temptation to cheat remains strong among some test takers.
Purposes/four-point test
27. LSAC’s stated purposes for collecting thumbprints are to assure the authenticity of test scores and to protect the integrity of the testing process. It acknowledges, however, that its primary purpose is one of deterrence—to prevent another individual from taking the test on behalf of the registered test taker. LSAC states that it will use the prints in cases of suspected impersonation.
28. LSAC states that thumbprints are not used to confirm the identity of the test taker at the time of the exam, but rather are used after the test if suspicious circumstances warrant their use. Thumbprints are considered to have a “fundamentally critical role” in deterring behaviour that would affect the test score. To LSAC, this purpose is key. LSAC believes that candidates who know in advance that they will have to provide a thumbprint will be less likely to participate in impersonation schemes. LSAC states that it also serves as incontrovertible evidence as to who actually took the test, should allegations of impersonation or malfeasance arise afterwards. LSAC states:
It is in the interests of all parties concerned – especially test-takers and law schools – that there be no suggestion of the possible compromise of test results. In the event of any allegation to the contrary, the thumbprints provide the highest possible assurance that test-takers are who they say they are.
29. When asked to explain how comparing two sets of prints could serve as evidence of who actually took the test, LSAC pointed to its check-in procedures used at each test administration. It stated that these procedures are thorough enough to allow it to be confident that the person who provides the thumbprint is the one who takes the examination. LSAC requires government-issued photo identification, and there is little opportunity between the check-in process and the beginning of the test administration to sneak in a substitute test taker. LSAC believes that by informing test takers in advance that they will have to provide positive photo identification and a thumbprint, they are deterred from concocting and executing schemes to deceive. It states that, “in fairness to those who take the LSAT without ?cheating and to meet our member schools’ need for a reliable score, we ? are compelled to have the best possible security measures in place.”
30. Although LSAC maintains that it is not subject to the Act, it nevertheless provided a rationale for its practice of collecting and using thumbprints based on a review of a previous finding issued by this Office in 2004, and the ensuing Federal Court case, as that particular set of complaints concerned the collection of biometric data, namely a voice print.
31. LSAC maintains that its purpose for collecting test takers’ thumbprints is a reasonable one. It submits that:
- The degree of sensitivity associated with thumbprints as personal information is as sensitive as that of voice prints.
- The security measures implemented by LSAC are narrowly tailored to meet its security needs and are appropriate.
- LSAC has bona fide business interests to which the collection of thumbprints is directed, namely, the preservation of the test’s integrity through a deterrence effect and the ability to identify test takers.
- The use of thumbprints is effective in meeting those objectives.
- The collection of thumbprints is reasonable as against alternative methods of achieving the same levels of integrity and identification at comparable cost and with comparable operational benefits.
- The potential for any loss of privacy is proportional to the costs and operational benefits given the security that LSAC provides.
32. LSAC states that it has a legitimate interest in preserving, at a minimum, the integrity of the testing process. It believes that the value of the LSAT to law schools as an aid in assessing applicants and the desire of test takers in having LSAT results that are not qualified by any claim of compromise are important factors for this Office to consider. LSAC goes on to state:
LSAC has adopted a very high identification standard so as to authenticate the identity of test-takers should it be required to do so. Any allegation of individuals taking the test for others would damage the trust placed in LSAC and in LSAT test scores by both test-takers and law schools. Any belief that such “substitute test-takers” could successfully take the test for others would arguably lead to an erosion of trust in the LSAT integrity. Whatever “intrusiveness” that may be argued by taking thumbprints is balanced by the compelling need to ensure test integrity. Given all of the circumstances surrounding the collection and use of thumbprints, we believe and would submit to your Office that a reasonable person would consider the collection of thumbprints appropriate in the circumstances. (emphasis added by LSAC)
33. As for limiting collection, LSAC states that, to the extent that the purpose of collecting a single thumbprint is intended to preserve the integrity of the testing process and permit the incontrovertible identification of test takers, it believes that the collection of personal information is limited to that which is “necessary.” Taking thumbprints is necessary to assure all interested parties that there is no suggestion of possible compromise of test results. Thumbprints provide the highest possible assurance that test takers are who they say they are.
34. In 2003-2004, 77.9 per cent of the total number of test takers took the LSAT just one time; 18.5 per cent took the test twice; and approximately 3.6 per cent took the LSAT more than twice. LSAC does not regularly report statistics on the number of repeat test takers for more than one test cycle. It did, however, provide some preliminary information on the 2005-2006 testing year. Its data indicated that 28 per cent of test takers who took the test during that cycle were exposed (at least once) to a previous test.
35. Our Office also reviewed the identification requirements for admissions tests administered in the United States; and in Britain and Australia, two jurisdictions with data protection laws. While the U.S.-developed admissions tests collected thumbprints, and in some cases, digital photographs at the time of the test, we could find no evidence that the British- and Australian-developed exams have such requirements. Typically, candidates were required to provide photo identification when presenting themselves for the test.
Application
36. In making our determinations, we applied paragraph 4(1)(a), subsection 5(3), and Principle 4.4. Paragraph 4(1)(a) states that Part 1 of the Act applies to every organization in respect of personal information that the organization collects, uses or discloses in the course of commercial activities. Subsection 5(3) states that an organization may collect, use or disclose personal information only for purposes that a reasonable person would consider are appropriate in the circumstances. Principle 4.4 states that the collection of personal information shall be limited to that which is necessary for the purposes identified by the organization.
Findings
37. The first matter to address is that of jurisdiction. LSAC contends that it is not engaged in any commercial activities for the purpose of the Act. In support of this position, LSAC relies on “its status as a non-profit, non-stock organization, its membership and governance structure and the public policy aspect in providing an education-related mechanism to assess individuals seeking to enter a regulated profession to practice law.”
38. LSAC’s status as a non-profit, non-stock, membership-based organization is not determinative. The Act applies to organizations, defined in section 2 as including “an association, a partnership, a person and a trade union.” There is no exemption for non-profit or member-oriented organizations. To the contrary, the definition of “commercial activity,” namely, “any particular transaction, act or conduct or any regular course of conduct that is of a commercial character, including the selling, bartering or leasing of donor, membership or other fundraising lists,” makes clear Parliament’s intention that the Act apply to commercial transactions that non-profit, membership-based organizations might engage in.
39. Moreover, I am unable to accept that LSAC’s activities are educational in nature. While LSAC’s purposes are related to ability measurement for the purposes of the law school admission process, this must necessarily be considered as separate from the actual process of training the mind or improving a branch of human knowledge. Ability measurement, in and of itself, does not constitute the provision of an educational service or reflect an education purpose. I am therefore unable to accept that there is an educational component to LSAC’s activities sufficient to place its activities outside the scope of the Act.
40. In my view, the LSAT assesses the pre-existing intellectual abilities of prospective law students for purposes associated with the administrative needs of LSAC’s member organizations and not the educational needs of prospective law students. By centralizing admissions testing, LSAC operates to alleviate administrative burdens that would otherwise fall to individual law schools that want assessed specific analytical abilities of prospective law students. LSAC members benefit from savings in time, cost and other resources, and the realization of economies of scale that centralized testing permits. The costs of LSAC’s activities are offset by the fees it charges to prospective law students who are required by most Canadian law schools to write the LSAT in order to pursue a professional career as a lawyer. The centralized administration of the LSAT by LSAC constitutes a valuable service to the professional schools that are LSAC’s members. In the circumstances, I am of the view that LSAC’s core activities serve, primarily, the administrative and organizational needs of its members and not educational or other public purposes. I am therefore satisfied that the administration of the LSAT by LSAC constitutes an activity with a commercial character, sufficient to satisfy the requirements of paragraph 4(1)(a).
41. The question of the reach of Canadian legislation to an organization such as LSAC, which is physically located in the US, must also be considered. A statute may apply to persons, property or transactions physically located outside the enacting body’s jurisdiction, where there is a sufficient link between the matter at issue and the jurisdiction of the enacting body. In such cases, the convergence of linking factors effectively brings the matter within the jurisdiction. The sufficiency of links is assessed by looking at real and substantial connections.
42. A variety of factors must be explored to determine whether such a connection exists, including: location in which the activity takes place; location to which profits flow; location of preparatory activities; residency of parties involved; location of contract; location of any potential related proceedings; jurisdiction where promotional efforts primarily targeted; location of content provider; location of host server; location of intermediaries; location of the end user.
43. A review of the investigation shows many links to Canada. By LSAC’s admission, Canada has a strong presence in LSAC. Fifteen Canadian laws schools are members. One of the 18 voting members on the Board of Trustees is typically from a Canadian law school. At least one member of the four permanent standing committees is from a Canadian law school. The President of the Canadian Law School Admission Officers sits on the Board of Trustees, participates in discussions, but does not have a vote.
44. LSAC accepts cheques and money orders in Canadian funds, when a student in Canada registers for the LSAT. LSAC may have no physical presence in Canada, but it contracts with the test centre supervisor (usually a member of the university’s Office of the Registrar). LSAC advertises its services directly to Canadians and in Canada. Tests are performed in Canada, and the thumbprints are collected in Canada. The LSAT is also integral to the Canadian law school admissions process.
45. Therefore, based on the foregoing, I am satisfied that a real and substantial connection exists between LSAC and Canada sufficient to bring the subject matter within jurisdiction of the Act.
46. Having thus established our Office’s jurisdiction, I will turn to the activity in question. In determining the appropriateness of thumbprinting LSAT candidates as put forth by LSAC, the four-point test needs to be considered. This is the test the Office relies on to assess the appropriateness of purposes with respect to subsection 5(3). The points are as follows:
- Is the measure demonstrably necessary to meet a specific need?
- Is it likely to be effective in meeting that need?
- Is the loss of privacy proportional to the benefit gained?
- Is there a less privacy-invasive way of achieving the same end?
47. LSAC states that it collects the thumbprints of candidates to assure the authenticity of test scores and to protect the integrity of the testing process by deterring would-be impersonators and by providing the means to identify the test taker in a suspicious situation.
48. Given the number of law schools that use the LSAT, it is understandable that LSAC wants to prevent impersonation and preserve the integrity of the test in fairness to honest test takers and to reassure member schools of the reliability of the test scores. In terms of evidence of impersonation problems, LSAC provided anecdotal information about impersonation being a problem prior to the introduction of thumbprinting in the 1970s. Some thirty years later, LSAC notes that impersonation continues to be a concern, and cited a recent solicitation in the Toronto area for someone to take the test on another’s behalf. While LSAC believes that the thumbprint requirement is an effective deterrent to impersonation schemes, the temptation to cheat nevertheless remains strong among some test takers.
49. However, the effectiveness of thumbprinting as a deterrent is difficult to measure. It is impossible to determine whether the thumbprint requirement in fact deters impersonation schemes or whether those inclined to cheat find a way around the requirement. LSAC claims that beforehand knowledge that thumbprints will be collected deters candidates from seeking out an expert test taker, yet it indicates that a flyer was recently found in Toronto in which an individual was looking for just that. Did that person go forward with the scheme? It is impossible to say.
50. Thumbprinting may be effective in those cases where impersonation is suspected (i.e. repeat test takers with suspicious results), but in over 30 years of this measure being in place, prints have never in fact been used for this purpose.
51. There is no question that some level of authentication is required to meet the need to ensure the test’s integrity, and LSAC does require test takers to provide photo identification. The prints are not used at the time of collection to ensure that a particular candidate is who she says she is. It is therefore possible that a one-time candidate, who has enlisted the assistance of an expert test taker and faked all identification, could “pass” the test without LSAC being any the wiser. Authenticity of the test score would thus not be assured.
52. Is the loss of privacy proportional to the benefit gained?
53. This answer lies in whether thumbprinting is considered privacy invasive. LSAC points to a case involving biometrics (voice prints) previously considered by this Office, to support its view that thumbprinting is not privacy invasive.
54. By their nature, all biometrics are privacy invasive to a certain extent. LSAC contends that thumbprints are as privacy invasive as the voice print, which our Office had determined was not very privacy invasive. However, unlike a voice print, a thumbprint carries with it a negative connotation, given its association with the criminal process. Given this association, the thumbprint takes on a more privacy-invasive character than the voice print, in my view.
55. Moreover, the measure needs to be considered within the proper context. Thumbprinting may be acceptable in certain circumstances. However, in this instance, it is being used in order to take an entrance exam, which is hardly a matter of public safety or national security.
56. The benefit of thumbprinting to LSAC is that it can assure its member law schools that it is doing what it can to ensure the security of the test. However, when one considers that the prints of thousands of individuals rest with the company, unlikely ever to be used, the balance shifts and the loss of privacy is not proportional to the benefit gained.
57. As for less privacy-invasive ways of achieving the same end, if test takers are properly authenticated at the time of testing, then there is no need to collect thumbprints that likely will never be used. We can contrast this lack of use with the case involving voice prints, in which the prints were to be used every time an employee wanted to enter the company’s computer system.
58. When thumbprinting was implemented in the mid-1970s, photo identification was less common. The current requirement to present valid photo identification should be, in my opinion, sufficient to minimize the possibility of impersonation.
59. With respect to Mr. Kluge’s concern about data stored in the United States being accessed by U.S. authorities under the provisions of the USA PATRIOT Act, I would note that, even if the thumbprints were stored in Canada, they would continue to be vulnerable to U.S. and Canadian law enforcement.
60. In sum, it is understandable that LSAC wants to ensure the authenticity of the test scores and to protect the integrity of the testing process by deterring impersonation and by providing the means to identify the test taker in cases of suspected fraud. Thumbprinting, however, does not effectively meet that purpose. It is impossible to measure how effective thumbprinting has been as a deterrent, and the prints have never actually been used in cases of suspected impersonation. Its questionable effectiveness and lack of use for its intended purpose shifts the balance, making the loss of privacy greater than the benefit gained. Finally, it is clear that this purpose can be appropriately met by properly authenticating candidates when they arrive to take the test. As a result, I am not satisfied that the collection and retention of thumbprints for the purpose of deterrence is appropriate under subsection 5(3). Furthermore, this practice is not limited to that which is necessary for this purpose, in contravention of Principle 4.4.
61. Before issuing findings in this complaint, former Assistant Commissioner Black recommended that LSAC cease collecting the thumbprints of students writing the LSAT in Canada. LSAC responded, indicating that it is prepared to explore alternative means to achieve its business objectives. It states that “without acquiescing to the jurisdiction of your office or the Act, and while respectfully disagreeing with your preliminary analysis and finding, LSAC will, effective…September 29, 2007…voluntarily suspend its policy requiring the collection of test takers’ thumbprints. Instead LSAC will use photographic evidence of test takers’ identities for the reasonable purpose of ensuring the security and integrity of the test.” The organization adds that “This step is without prejudice to LSAC’s ability to reinstitute its thumbprint policy at some future date and does not waive any ability LSAC may have to contest the jurisdiction of your office or the Act in future legal proceedings.”
62. LSAC clarified what it meant by using “photographic evidence.” Firstly, the organization asserts that, since this complaint pertains to the collection and retention of test takers’ thumbprints, LSAC is not precluded from collecting and retaining test takers’ photographic evidence. Secondly, LSAC’s instruction to candidates participating in testing as of July 2007 states, “effective with the September 2007 Administration of the LSAT, candidates testing at test centers in Canada MUST bring a photograph to the test center on the day of the test. These photographs will be retained by LSAC.”
63. The instruction to candidates details the specifications for the photograph. It must be “a recent, clear photograph of yourself” and it “must have been taken within the last six months and must show only your face and shoulders. Laminated copies and photocopies of your photo are not acceptable. … The photograph must be clear enough so that there is no doubt about your identity. Your face in the photo must show you as you look on the day of the test.” It also notes that the student’s name, test date, LSAC account number and test centre code should be printed on the back of the photograph before it is glued or taped to the admission ticket in case the ticket and photo become separated. Additionally, it stresses that a student without a recent photograph attached to her or his LSAT admission ticket will not be allowed to take the test. Each student must also bring to the testing location one form of government-issued identification that includes her or his signature.
64. In her preliminary report, Assistant Commissioner Black determined that the practice of collecting thumbprints was unacceptable as it does not meet the purpose of ensuring the authenticity of test scores and protecting the integrity of the testing process by deterring impersonation and by providing the means to identify the test taker in cases of suspected fraud. She noted that the current obligation to present government-issued positive photo identification should be all that is necessary to mitigate the possibility of impersonation.
65. In response to our recommendations, LSAC has chosen to replace its requirement for thumbprints with one for photographs. Has LSAC simply substituted one unacceptable collection and retention practice for another? As was the case with thumbprints, LSAC proposes to collect and retain photographs for the primary purpose of deterrence.
66. The organization wishes to prevent a registered test taker from cheating by getting someone else to take the test on his or her behalf, and it wants to be able to use the information collected should suspicious circumstances occur after the fact. However, it is not clear whether the collection and retention of photographs is appropriate for the purpose of deterrence. In order to determine whether this is the case, we have again applied the four-point test.
67. Firstly, is the measure demonstrably necessary to meet a specific need? As the former Assistant Commissioner noted, it is understandable that LSAC wants to prevent impersonation and preserve the integrity of the test in fairness to honest test takers and to reassure member schools of the reliability of the test scores. Although LSAC remains concerned that the temptation to cheat has not abated since the 1970s, it has no evidence of any impersonation attempts. It is not clear whether this means that the previous measure (thumbprinting) worked or whether impersonations have gone undetected. The proposed measure (photographs) will in all likelihood suffer the same uncertainty. The lack of proof one way or another makes it difficult to assess the need for a particular measure as a deterrent. Obviously, however, some level of authentication is required.
68. Secondly, is the measure likely to be effective in meeting that need? With such a limited amount of factual evidence available regarding the level of cheating involved among test takers, it is difficult to measure the effectiveness of any deterrent, including photographs. However, compared to the difficulty of matching thumbprints scientifically, there is an increased ease with which photographs may be visually matched to a test taker or to a publicly available image of a test taker. This might result in an improved ability to identify and catch cheaters after the fact, thus enhancing the deterrent effect. Also, as with thumbprints, photographs may be effective in instances for which impersonation is suspected, i.e. repeat test takers with questionable results. By and large, I am inclined to believe that photographs may be effective in meeting the need for deterrence.
69. Thirdly, as to whether the loss of privacy is proportional to the benefit gained, clearly there is a difference between the collection and retention of thumbprints and that of photographs. In my view, an individual’s expectation of privacy regarding her or his own photograph is weaker than the expectation of privacy regarding a thumbprint. Government ministries, employers, educational institutions and regulatory bodies regularly collect and retain photographs of individuals for authentication, identification and other acceptable purposes.
70. In this Office’s Preliminary Report, mention was made of the negative connotation of the association between thumbprints and the criminal law process. There is no such connotation with photographs. Additionally, we noted that, with thumbprints, there is a disproportionate invasion of privacy considering the circumstances, e.g. a law school entrance exam as opposed to national security. There is less concern with photographs as the personal information is not as sensitive. The collection and retention of photographs, in my opinion, is less privacy invasive than the collection and retention of thumbprints. The loss of privacy would appear to be low.
71. What is the benefit gained by LSAC? As with thumbprints, LSAC can assure its member law schools that it is taking active steps to ensure the security of the test and the reliability of the test results. On the one hand, if there is after-the-fact evidence that arouses suspicions of impersonation, LSAC would benefit from the availability of a photographic record for reference. On the other hand, there seems never to have been a need to refer to thumbprints after the fact, and the same situation could occur with photographs. However, considering that photographs are considerably less sensitive than thumbprints and that the loss of privacy, in the circumstances of this case, is minimal, I believe that, on balance, the loss of privacy is proportional to the benefit gained.
72. Lastly, is there a less privacy-invasive way of achieving the same end? In the case of thumbprints, less privacy-invasive alternatives undoubtedly exist. In the case of photographs, LSAC already has a requirement for government-issued photo identification to be presented (but not collected or retained) for authentication purposes at the time of the test. I remain of the opinion that this obligation should be all that is necessary to mitigate the possibility of impersonation. Clearly, this is less privacy-invasive than the collection and retention of photographs and would most likely achieve the same end. However, I am willing to grant that LSAC’s proposal of a new process requiring it to collect and retain test takers’ recent photographs, in addition to viewing their government identification, is only slightly more privacy invasive than simply viewing government identification. It also provides a means by which test takers suspected of fraud can be identified after the fact.
73. Overall, I am of the opinion that an appropriate balance has been achieved between the right of LSAC to collect personal information for purposes that a reasonable person would consider appropriate in the circumstances and the right of the individual to privacy. There is some evidence to support the effectiveness of collecting and retaining independent photographs; the loss of privacy is proportional to the benefit gained; and, while a less privacy-invasive measure exists, it is only marginally less invasive. I believe, therefore, that the collection and retention of such photographs for the purpose of deterring fraud meets the requirements of subsection 5(3) and does not contravene Principle 4.4 as it is limited to that which is necessary for LSAC’s purpose.
74. The collection of photographs appears to be substantially less problematic than the collection of thumbprints. I have determined, therefore, that the organization has not simply substituted one unacceptable collection and retention practice for another.
75. Nonetheless, I am concerned by LSAC’s assertion that, while it has suspended for the moment its policy of collecting and retaining thumbprints, it will keep in reserve its “ability to reinstitute its thumbprint policy at some future date.” I firmly believe that its thumbprint policy is not reasonable for the stated purposes and contravenes Principle 4.4. As LSAC maintains that it can reinstitute this policy whenever it deems it to be necessary, I cannot conclude that the complaint is resolved.
76. Accordingly, the complaint is well-founded.
77. In order to comply with the Act, I recommend that LSAC:
- Permanently cease its collection of thumbprints; and
- Limit the retention of independent photographs to five years, if it continues to collect them.
78. We will be pursuing the matter in accordance with our authorities under the Act.
Postscript
Following the issuance of the Commissioner's Report of Findings, both parties were prepared to pursue legal action. The matter was settled prior to either party issuing a Notice of Application.
LSAC committed to implementing all of the Office's recommendations. LSAC will permanently cease the collection of fingerprints from individuals taking the Law School Admission Test in Canada. LSAC will limit to five years its retention of the independent photographs it will use in place of fingerprints to verify identity and deter fraud. These photographs will be securely stored and confidentially and securely destroyed after five years.
The OPC agreed not to initiate any litigation and to issue this addendum updating its Report of Findings to reflect the resolution of this matter. As had been previously outlined in our May 29, 2008 Report of Findings, the OPC determined that LSAC's current practices in relation to photographic verification of identity for the purposes of deterring fraud meet the requirements of s. 5(3) of the Act and comply with Principle 4.4. The Office is pleased to consider the matter resolved.
Related Documents
For further information about related investigations involving college admissions examinations, please see:
- PIPEDA Case summary #2010-007: Test administrator revises measures aimed at preventing exam fraud
- PIPEDA Case summary #2011-012: GMAT Test-taker Objects to Palm-Vein Scanning
- Lessons Learned: Collection of Biometric Information for College Admissions Examinations; An Overview of Three OPC Investigations
- Date modified: