ORDER PO-3859

Appeal PA16-303

The University of Waterloo

June 20, 2018

Summary: The engineering program at the University of Waterloo (the university) considers a number of factors in its admissions process, including a number known as an “adjustment factor” that it assigns to various Ontario high schools. The appellant, who is a reporter, submitted an access request to the university under the Freedom of Information and Protection of Privacy Act  (the Act ) for a record that shows the adjustment factor that the university has assigned to each high school. The university located a responsive record and ultimately denied access to it under the mandatory exemption in section 21(1) (personal privacy) of the Act . In this order, the adjudicator finds that the information in the record does not qualify as the “personal information” of any of the university’s engineering students. In addition, he does not accept the university’s argument that if the record were made public, the information in it could be combined with other publicly available information, which would reveal the personal information of its engineering students. Given that the section 21(1) exemption only applies to “personal information,” the information in the record cannot be withheld under that provision. The adjudicator orders the university to disclose the record to the appellant.

Statutes Considered: Freedom of Information and Protection of Privacy Act , R.S.O. 1990, c. F.31 , as amended, ss. 2(1) (definition of “personal information”) and 21(1).

Orders Considered: Orders P-230, P-1880, PO-2240 and PO-2518.

Cases Considered: Ontario (Attorney General) v. Pascoe, 2002 CanLII 30891 (ON CA), aff'g 2001 CanLII 32755 (ON SCDC).

OVERVIEW:

[1]  The engineering program at the University of Waterloo (the university) has a highly competitive admissions process. In assessing high school students, the university considers a number of factors in its admissions process, including a number known as an “adjustment factor” that it assigns to Ontario high schools. [1] The adjustment factor represents the university’s experience with past applicants from each high school as to how much the grades of those admitted will drop by the end of first year in the engineering program.

[2]  The appellant is a reporter who submitted an access request under the Freedom of Information and Protection of Privacy Act  (the Act ) to the university for the following record:

When considering admission applications, the University of Waterloo’s engineering department uses an adjustment factor to compare grades from different high schools.

I request a copy of this document, including the name of the high schools and the adjustment factor used, for the most recent year available, released as an electronic record in a sortable format, such as .txt, .csv, or .xlsx.

[3]  In response, the university located a two-page record entitled, “Adjustment Factors for 2016 Engineering Admission Cycle.” This chart has three categories, including a list of various Ontario high schools and the adjustment factor for each high school. The university then sent a decision letter to the appellant stating that it was denying him access to this record under the discretionary exemption in section 18(1)(c) (economic and other interests) of the Act .

[4]  The appellant appealed the university’s decision to deny him access to this record to the Information and Privacy Commissioner of Ontario (IPC), which assigned a mediator to assist the parties in resolving the issues in dispute. During mediation, the university sent a revised decision letter to the appellant stating that it was denying access to the record under the mandatory exemption in section 21(1) (personal privacy) of the Act . [2] This letter stated:

The record contains information which, if disclosed, could reveal the personal information of University of Waterloo students. The record itself does not contain personal information, but the University believes that if it were made public and combined with other publicly available information it would reveal the personal information of some of our students. In particular, engineering students' grades would be able to be "reverse engineered" by individuals, for example co-op employers, knowledgeable about where they attended high school, and to which programs they applied at the University of Waterloo. All of our engineering students list their high schools on their résumés, and are not required to provide their first year university grades. The University's minimum admission averages are public and the large majority fall into a very limited grade range. The University believes this is an unjustified invasion of our students' privacy and may expose them to bias from employers and others.

[5]  The appellant advised the mediator that he believes that there is a compelling public interest in disclosing the record. Consequently, the public interest override in section 23  of the Act  may be at issue in this appeal.

[6]  This appeal was not resolved during mediation and was moved to adjudication for an inquiry. Both the university and the appellant provided representations to me on the issues to be resolved in this appeal.

[7]  In this order, I find that the information in the record does not qualify as the “personal information” of any of the university’s engineering students. In addition, I do not accept the university’s argument that if the record were made public, the information in it could be combined with other publicly available information, which would reveal the personal information of its engineering students. Given that the personal privacy exemption in section 21(1) only applies to “personal information,” the information in the record cannot be withheld under that provision. I order the university to disclose the record to the appellant.

RECORD:

[8]  The record at issue in this appeal is a two-page chart entitled, “Adjustment Factors for 2016 Engineering Admission Cycle.” This chart includes the following categories: In Canada, Ontario Secondary Schools, and Outside Canada.

ISSUES:

1. Does the record contain “personal information” as defined in section 2(1) of the Act and, if so, to whom does it relate?
2. Does the mandatory exemption at section 21(1)  of the Act  apply to the information at issue?
3. Is there a compelling public interest in disclosure of the records that clearly outweighs the purpose of the section 21(1) exemption?

DISCUSSION:

PERSONAL INFORMATION

A. Does the record contain “personal information” as defined in section 2(1) of the Act and, if so, to whom does it relate?

[9]  The mandatory personal privacy exemption in section 21(1)  of the Act  states, in part:

A head shall refuse to disclose personal information to any person other than the individual to whom the information relates except,

(f)  if the disclosure does not constitute an unjustified invasion of personal privacy. 

[10]  The university claims that disclosing the list of Ontario high schools and the adjustment factor that it assigned to each school would constitute an unjustified invasion of its engineering students’ personal privacy and this information is therefore exempt under section 21(1). However, section 21(1) only applies to “personal information.” Consequently, it must be determined whether the information at issue is “personal information,” because if it is not, it cannot be exempt under section 21(1).

[11]  The definition of this term in section 2(1)  of the Act  states:

“personal information” means recorded information about an identifiable individual, including,

(a)  information relating to the race, national or ethnic origin, colour, religion, age, sex, sexual orientation or marital or family status of the individual,

(b)  information relating to the education or the medical, psychiatric, psychological, criminal or employment history of the individual or information relating to financial transactions in which the individual has been involved,

(c)  any identifying number, symbol or other particular assigned to the individual,

(d)  the address, telephone number, fingerprints or blood type of the individual,

(e)  the personal opinions or views of the individual except if they relate to another individual,

(f)  correspondence sent to an institution by the individual that is implicitly or explicitly of a private or confidential nature, and replies to that correspondence that would reveal the contents of the original correspondence,

(g)  the views or opinions of another individual about the individual, and

(h)  the individual’s name where it appears with other personal information relating to the individual or where the disclosure of the name would reveal other personal information about the individual;

[12]  The list of examples of personal information under section 2(1) is not exhaustive. Therefore, information that does not fall under paragraphs (a) to (h) may still qualify as personal information. [3]

[13]  An essential requirement of the definition of “personal information” in section 2(1) is found in its opening words, which limit it to recorded information “about an identifiable individual.” In Order P-230, former Commissioner Tom Wright set out the following test for whether information in a record is “about an identifiable individual”:

. . . If there is a reasonable expectation that the individual can be identified from the information, then such information qualifies under subsection 2(1) as personal information.

[14]  Subsequently, in Order PO-2240, former Assistant Commissioner Tom Mitchinson elaborated on this approach when applying it to information in records which do not specifically name or otherwise identify any individuals:

. . . The comments of former Commissioner Tom Wright in Order P-230 are the starting point for any discussion of "personal information" where no individual is named or otherwise specifically identified on the face of a record. In order to satisfy the definition of "personal information" in these circumstances, there must be a reasonable expectation that an individual can be identified from the information in the record.

[15]  The same general approach was upheld by the Ontario Court of Appeal in a judicial review of Order P-1880, in which Adjudicator Irena Pascoe followed the test set out in Order P-230. [4] Her decision had previously been upheld by the Ontario Divisional Court, which stated that in order to establish that information is identifiable, an institution must provide submissions establishing a nexus connecting the record, or any other information, with an individual. In the Court’s view, any connection between a record and an individual, in the absence of evidence, is “merely speculative.” It stated:

The test then for whether a record can give personal information asks if there is a reasonable expectation that, when the information in it is combined with information from sources otherwise available, the individual can be identified. A person is also identifiable from a record where he or she could be identified by those familiar with the particular circumstances or events contained in the records. [See Order P-316; and Order P-651]. [5]

[16]  Similarly, in Order PO-2518, Senior Adjudicator John Higgins stated that it is necessary to bear in mind that information that seemingly does not identify an individual may, in fact, not be anonymous where other information is combined with it to identify an individual.

Summary of the parties’ representations

University’s representations

[17]  The university acknowledges that the record, on its face, does not contain personal information but asserts that if the record were made public, it could be combined with other publicly available information which would reveal the personal information of its engineering students. It submits that the personal information that would be revealed is the students’ “education history,” as set out in paragraph (b) of the definition in section 2(1).

[18]  To explain how the personal information of its engineering students would be revealed, the university first provides a general explanation about the “adjustment factor” and sets out the information that is publicly available with respect to the grade averages of high school students admitted to its engineering program:

The "adjustment" document used by the Faculty of Engineering at the University to assess applicants contains identified high schools and an adjustment factor (which is a representation of the University's experience with past applicants from those schools as to how much students' grades will drop by the end of first year at Waterloo; it is used as one part of the University's decision making process as to whether to offer students admission to study at the University; see paragraph 23 for more detail).

The University's minimum engineering admission averages are public. [6] A clear preponderance of our engineering students' admission averages from high school range between 90 and 100%, with most being admitted with averages greater than 95%; some students are admitted with lower averages.

[19]  Elsewhere in its representations, the university also provides a detailed description of how it calculates the adjustment factor for each high school but asked that this part of its representations not be shared with the appellant. [7] Based on the IPC’s confidentiality criteria for sharing representations, I decided not to share this specific calculation method with the appellant, and I cannot, therefore, reveal its details in this order. However, generally speaking, the adjustment factor for each high school appears to be based not on a single student but on several years of grade data relating to different students admitted from that school to the university’s engineering program.

[20]  The university provides the following information about its co-op program and the information that engineering students provide to potential co-op employers:

The University's engineering students are all enrolled in the co-operative program (co-op) which means that over the course of their studies, they are required to obtain co-op employment for four or five terms. All engineering students must participate in the co-op program of study.

Engineering students complete either one term or two terms of study and then enter a co-op term, with engineering's architecture students starting co-op in their second year of study.

All engineering students list their high school on their résumés for co-op placements per University requirements. They are not required to provide their university grades. Students also self-identify their high schools in many cases, including through job applications, posting of their résumés through University job banks, social media platforms, and in general conversations and job interviews. This information could relate to past and current University students.

[21]  It then provides the following arguments to support its position that disclosing the list of high schools and the adjustment factor assigned to each school would reveal the “personal information” of its engineering students to both co-op and other employers:

If the University's adjustment factor document were public, any employer who has an application from an engineering student could therefore, and with a high degree of accuracy, know the very limited range of that student's high school grade average, and that student's likely grade average at the University. Further, students who seek employment following completion of their degree would suffer from the same issue.

For example, if a student attended a particular high school which has an adjustment factor of "25", and is enrolled in geological engineering at the University, a reviewer of that student's résumé would know that the student had a very high probability of graduating high school with a grade between 90% and 100% (with a 90% chance that the student's grade average was greater than 95%), and a very high probability that the student's grades after one year of study at the University would be in the 65% to 75% range (again, with a 90% chance that the grades are in the 75% range).

The University contends that even knowing the range of a student's grades comprises that student's educational history and therefore is that student's personal information. The University believes that all students should be in control of their own grade information and what is shared with potential employers.

Appellant’s representations

[22]  The appellant disputes that the list of high schools and the adjustment factor that the university assigned to each school would reveal the personal information of any of the university’s engineering students. He states, in part:

The University's assertion that the document at issue could be used to reveal personal information does not stand up to critical scrutiny.

An adjustment document of this kind would only be useful if it was drawn from a sufficiently large sample size. The largeness of the sample size acts to protect any given individual student's privacy.

Without knowing the numbers involved (the cutoff of students under which a high school wouldn't be included in the University's adjustment document) it is hard to comment further, but the dynamics of the situation tend not to lead to a small cell count issue.

University’s reply representations

[23]  The university claims that the appellant has missed the point that it is making. It states, in part:

. . . The high school attended is the key by which personal information about the University's engineering students is revealed: both their high school grades, and their university grades at the end of first year. Further, the University believes any employer may use the record, whether for a co-op position or not, to eliminate candidates for positions based upon factors not presented by the candidates.

Analysis and findings

[24]  For the reasons that follow, I find that the information in the record is not “about an identifiable individual” and does not, therefore, qualify as “personal information,” as that term is defined in section 2(1)  of the Act . In addition, I am not persuaded by the university’s argument that if the record were made public, the information in it could be combined with other publicly available information, which would reveal the personal information of its engineering students.

[25]  As noted above, the record at issue in this appeal is a two-page chart entitled, “Adjustment Factors for 2016 Engineering Admission Cycle.” This chart includes the following categories: In Canada, Ontario Secondary Schools, and Outside Canada.

[26]  The university’s representations focus on the “Ontario Secondary Schools” category in the record. No individual is named or otherwise specifically identified in this part of the record, which simply contains a list of various Ontario high schools and the adjustment factor that the university assigned to each school. The adjustment factor is not based solely on the grades of one particular student who attended a listed high school. It is a number that the university assigns to each high school that represents its experience with past applicants from each school as to how much the grades of those admitted will drop by the end of first year in the engineering program.

[27]  The university does not specifically refer to or apply the reasonable expectation test for determining whether information in a record is “about an identifiable individual,” and hence “personal information,” which has been established in previous IPC orders and upheld by the courts. On the contrary, it acknowledges that the record, on its face, does not contain “personal information.”

[28]  In my view, given that the adjustment factor for each high school in the record appears to be based on several years of admissions data relating to a number of students at each school, it is not reasonable to expect that any of these students may be identified from this information. However, as established in previous IPC jurisprudence, it is necessary to bear in mind that information that seemingly does not identify an individual may, in fact, not be anonymous where other information is combined with it to identify an individual.

[29]  The university does not directly assert that it is reasonable to expect that if the information in the record is combined with information from other sources, an individual may be identified. Instead, it submits that combining the list of high schools and the adjustment factor for each school with other information will reveal the personal information of its engineering students. In particular, if co-op or other employers combine the information in the record with other information, they will be able to identify both “the very limited range” of a student's high school grade average, and that student's “likely grade average” at university, which is information about an individual’s “education,” as found in paragraph (b) of the definition of “personal information” in section 2(1)  of the Act . In my view, this amounts to an argument that if the information in the record that seemingly does not identify an individual is combined with other information, it is reasonable to expect that an identifiable individual’s grade information will be revealed.

[30]  By way of example, the university claims that if a student attended a particular high school which has an adjustment factor of "25", and is enrolled in geological engineering, a potential employer who reviews that student's résumé would know that this specific student had a very high probability of graduating high school with a grade between 90% and 100% (with a 90% chance that the student's grade average was greater than 95%), and a very high probability that the student's grades after one year of study at the university would be in the 65% to 75% range (with a 90% chance that the grades are in the 75% range).

[31]  I do not find the university’s submissions on this point to be persuasive for a number of reasons. The university's minimum engineering admission averages are publically available on its website. The university provided me with a printout of this webpage, which contains a chart which lists the various programs in the engineering department, the various high school grade ranges of students admitted to each of those programs, and the corresponding probability of receiving an offer for each of these grade ranges. For example, it states that a high school student who applies to geological engineering would have the following probability of receiving an offer based on their high school grade range: 80-85% (5% probability), 85-90% (30% probability), 90-95% (65% probability), and 95%+ (90% probability).

[32]  A potential employer who reviews the résumé of a student enrolled in the geological engineering program could certainly use this publically available information to attempt to surmise that student’s grade average in high school. However, unless the student provided his or her actual grade average, the employer would be engaging in guesswork. The employer could deduce that there is a very high probability that the student had a high school grade average of 90 to 100% but it is also possible that he or she was part of the minority of students admitted to the program who had grade averages as low as 85-90% or in rare circumstances, 80-85%.

[33]  Given the public availability of this information, I fail to see how disclosing the list of high schools and the adjustment factor for each school would somehow enable co-op and other employers to know “the very limited range” of that student’s high school grade average, as claimed by the university. The adjustment factor represents the university’s experience with past applicants from each school as to how much students' grades will drop by the end of first year in the engineering program. It does not add anything to the publicly available information about the university's minimum engineering admission averages that would enable an employer to more accurately deduce the range of a particular student’s grade average in high school.

[34]  The university also claims that if the list of high schools and the adjustment factor for each is school is disclosed, employers will be able figure out an engineering student's “likely grade average” at university. In the example it provided above, the university claims that if a student attended a particular high school which has an adjustment factor of "25", and is enrolled in geological engineering, a potential employer who reviews that student's résumé would know that this specific student had a very high probability of graduating high school with a grade between 90% and 100%, and a very high probability that the student's grades after one year of study at the university would be in the 65% to 75% range (with a 90% chance that the grades are in the 75% range).

[35]  In my view, this would amount to an exercise in speculation and not one that would accurately reveal the “likely grade average” of any student in first-year university. If a student attended a particular high school which has an adjustment factor of "25", a co-op or other employer could attempt to figure out that student’s “likely grade average” in first-year university. However, as noted above, with respect to a student’s high school grade average, even if there is a very high probability that a student in geological engineering had a high school grade average of 90% to 100%, it is possible that he or she could fall within the minority of students who were admitted with grades below that range. As a result, the employer would be engaging in guesswork as to that student’s high school grade average or even the range when subtracting the adjustment factor of “25” from it.

[36]  In addition, it is important to note that the adjustment factor is a prediction or estimate based on past applicants from a particular high school as to how much students' grades will drop by the end of first year in the engineering program. Each student, however, is a unique individual and could presumably perform better or worse in first-year university than the “25” adjustment factor that the university assigned to his or her high school, which is based on the performance of a number of students from that same school who were admitted to the engineering program over several years.

[37]  In other words, rather than being able to accurately deduce an engineering student's “likely grade average” in first-year university, a co-op or other employer would, in my view, be engaging in guesswork. Although the university claims that an employer would know that there is a very high probability that a student's grades after one year of study at the university would be in the 65% to 75% range (with a 90% chance that the grades are in the 75% range), a student’s actual grade average could clearly fall outside this range, which means that an employer would be speculating about a student’s possible first-year grade average in geological engineering.

[38]  In these circumstances, I find that if the information in the record that seemingly does not identify an individual is combined with other information, it is not reasonable to expect that an identifiable individual’s grade information will be revealed, whether it be “the very limited range” of a student's high school grade average or their “likely grade average” in first-year university.

[39]  Finally, I note that in addition to the “Ontario Secondary Schools” category, the record also includes an “In Canada” category that shows the adjustment factor that the university assigned to other individual Canadian provinces and territories (e.g., the adjustment factor for Manitoba) and an “Outside Canada” category for various countries (e.g., the adjustment factor for India). The university’s representations do not address whether the information in these two latter categories is “personal information.” In the absence of such evidence and for the same reasons set out above with respect to the information in the “Ontario Secondary Schools” category, I find that it is not reasonable to expect that any student may be identified from this information, and it is not, therefore, “personal information,” as that term is defined in section 2(1)  of the Act .

[40]  In summary, I find that the information in the record is not “about an identifiable individual” and does not, therefore, qualify as “personal information,” as that term is defined in section 2(1)  of the Act . In addition, for the reasons set out above, I am not persuaded by the university’s argument that if the record were made public, the information in it could be combined with other publicly available information, which would reveal the personal information of its engineering students. Given that the personal privacy exemption in section 21(1) only applies to “personal information,” the information in the record cannot be exempt from disclosure under that provision. As the university is not relying on any other exemptions in the Act , [8] I will order it to disclose the record to the appellant.

[41]  Because the information in the record is not exempt from disclosure under section 21(1)  of the Act , it is not necessary to consider whether the public interest override in section 23 applies. However, I note that there are transparency and accountability interests at stake in this appeal. Ontario’s universities are covered by the Act  because they are publically funded and are expected to be transparent and accountable to the public, which include both high school students and their parents. To the maximum extent possible, information about the admissions process for any university program should be open, transparent and accessible.

[42]  It is evident from the university’s representations that the adjustment factor that it assigns to each high school is one factor that it considers in assessing students for admission to its highly competitive engineering program. In my view, disclosing this record will advance the principles of transparency and accountability that underlie the Act  by enabling the public to scrutinize the fairness of this particular component of the university’s admissions process.

ORDER:

I order the university to disclose the entire record to the appellant by July 23, 2018.

Original Signed by:		June 20, 2018
Colin Bhattacharjee
Adjudicator