Blake v. Ontario (Ministry of Correctional Services), 1984 CanLII 5043

Blake v. Ontario (Ministry of Correctional Services)

1984-09-10

Ontario Board of Inquiry

CHRR Doc. 84-065

Rosetta Blake Complainant

v.

The Ministry of Correctional Services and Mimico Correctional Institute Respondents

Date of Decision: September 10, 1984

Before: Ontario Board of Inquiry, Peter A. Cumming

Appearances by: D.A.J. D'Oliveira, Counsel for the Ontario Human Rights Commission and Rosetta Blake Janice A. Baker, Counsel for the Ministry of Correctional Services and Mimico Correctional Institute

RACE, COLOUR AND PLACE OF ORIGIN — employment denied — SEX DISCRIMINATION — position as correctional officer denied — SYSTEMIC DISCRIMINATION — pattern of conduct discriminatory on the basis of age, gender, race

EVIDENCE — availability and admissibility of similar fact evidence, statistical evidence — DISCRIMINATION — intention to discriminate

Summary: The Board of Inquiry dismisses a complaint alleging discrimination on the basis of sex, race, and age by Rosetta Blake. Ms. Blake alleged that she was refused employment as Correctional Officer at Mimico Correctional Institute because she is black and a woman and was 50 years old at the time of her application.

On Ms. Blake's behalf, the Ontario Human Rights Commission presented considerable statistical evidence which the Board of Inquiry accepted as establishing a prima facie case of discrimination on the basis of sex and race. The statistical evidence demonstrated that women and non-Europeans had a significantly lower chance of receiving an interview for a job despite their educational qualifications being the same.

Despite the statistical evidence, however, the Board of Inquiry rejects the complainant's allegation of discrimination finding that in her individual case Ms. Blake was treated in a fair and objective manner and the refusal to employ her for the position of Correctional Officer was a reasonable one in the circumstances.

1The Complainant, Ms. Rosetta Blake, alleges discrimination on the basis of her race, colour, age and/or sex by the Respondent, Her Majesty the Queen in Right of Ontario in respect of the Ministry of Correctional Services, through the Mimico Correctional Institute, (hereafter "Mimico") in contravention of paragraphs 4(1)(a) and/or (e) of the Ontario Human Rights Code, R.S.O. 1980, c. 340 (hereafter, the "Code"), which read as follows:

4(1) No person shall,

(a) refuse ... to recruit any person for employment; (or)

(e) establish or maintain any employment classification or category that by its description or operation excludes any person from employment ... ;

because of race ... colour, age (or) sex ... of such person ...

2Ms. Blake, 50, is a black female, born and raised in Jamaica. She completed her high school there, then graduated after a three year course in "mental nursing" from the Mental Hospital in Jamaica, and worked in that hospital until 1964. Her nursing duties included custodial care, the bathing of, assisting in the medication and therapy of, and movement of, patients. The 2000 mental patients of the hospital had a full range of emotional disturbances.

3Ms. Blake immigrated to Canada in 1964, where she worked at the Goderich, Ontario, Hospital from October, 1964 to June, 1966. She then requested, and received, a transfer to the Lakeshore Psychiatric Hospital where she worked until February, 1972. As well, while working at that hospital, she also took the three year nursing program at Ryerson, although she was unsuccessful (Exhibit #6) in the nursing course part of the program.

4In 1972, she commenced a two year program at Centennial College, successfully receiving her diploma (Exhibit #7) as a "Community Service Research Technician" in 1975. She then worked for a Local Initiatives Program, counselling mainly immigrant citizens, and also successfully completed a B.A. degree (Exhibit #8) program at York University in October, 1977, majoring in psychology.

5In her concluding months at York, she sought employment with various agencies, including the respondent, Mimico. Ms. Blake is a gentle, well-spoken person, who is obviously industrious, and it would seem successful in her past employment record.

6Ms. Blake sought employment as a correctional officer with Mimico in 1977. Her Complaint (Exhibit #3) states:

On June 9, 1977, I sent by mail an application for a job as Correctional Officer to the above mentioned Correctional Institute. I received a letter of rejection dated June 10, 1977 advising I had been rejected on the basis of "a combination of related experience, educational qualifications and personal suitability." I had, of course, never been interviewed, seen or spoken with anybody at the Institute.

On advice from the above Ministry, I telephoned the Mimico Correctional Centre on July 28, 1977 to ask about vacancies as Correctional officers. I was told by a male person that this was a male institution and "what do you think you can do here." After I said that my request was prompted by the Ministry he asked me my age and said that I sounded "like a little girl." Then he said I should phone later and speak to a Mr. Burns.

I did, and I was invited to fill in an application form, after which I would be interviewed. Two weeks after completing the form I telephoned to ask when I would be interviewed. I was given no satisfaction at that time, or since in many attempts to clarify the status of my application.

7Although her Complaint does not refer to it, Ms. Blake filled out a third application by letter dated November 25, 1977, and was interviewed shortly thereafter by Superintendent De Grandis who decided she should not be hired.

8At the time, Mimico was a minimum security institution. Mr. Carl De Grandis, 38, is the Superintendent of the Mimico Correctional Institute, with an extensive background in correctional services work. His spouse is a correctional officer. He became Superintendent at Mimico on May 23, 1977, and, actually started May 30, 1977. Perhaps because Mimico had a very uncertain future at the time, he found it to be understaffed and the staff to be quite unmotivated, (Transcript, Vol. VI, pages 140–142).

9Mr. De Grandis is, quite obviously, a dynamic, dedicated, thoughtful and competent Superintendent with enthusiasm for his job and a broad and enlightened perspective as to the objectives of a correctional institution. When he first arrived, he found inmates were largely just "laying around purposelessly," but Mr. De Grandis put people to work renovating the facilities (Transcript, Vol. VII, pages 60–62). He initiated programs seeking to facilitate the return of inmates to the community. Under his leadership, Mimico developed the largest "half-way house" program, the largest community services program, and the largest daily release for employment purposes programs in the country. (Transcript, Vol. VII, page 65). In a nutshell, Mr. De Grandis has instilled a sense of pride and purposefulness in his staff and in many of the inmates under his charge. Mr. De Grandis seems, having heard his testimony, and that of other witnesses, to be an excellent Superintendent.

10At that time, Mr. Thomas Burns was his deputy superintendent, Mr. John Stone, the senior assistant superintendent, and Mr. William Aird the assistant superintendent. Mr. Van Horne, who looked after several correction institutes as regional personal administrator, came to Mimico about one-half day a week. As well, Mr. Ross Kennedy, at one time a Sergeant and later a Lieutenant, worked at Mimico.

THE EVIDENCE

11Ms. Blake first applied, by letter, to the Mimico Correctional Institute for a position as a correctional officer on June 9, 1977 and was rejected by letter dated June 10, 1977, the day her letter was received (Transcript, Vol. II, pages 30–32, Exhibit #9; Exhibit #10). However, this application was submitted in response to a newspaper advertisement which set a deadline for applications of May 25, 1977. (Exhibit #16, Vol. 2, copy attached to application of Mr. William Daniels).

12Her letter of rejection was signed by J.R. Stone, who was at that time the senior assistant superintendent. Mr. De Grandis gave evidence that he recalled "finding out that the interviews were being done in those early days by Mr. Burns and Mr. Stone." (Transcript, August 16th, Vol. VI, pages 145–147). Neither Mr. Burns nor Mr. Stone were called as witnesses by the respondent, without explanation, but neither was Mr. Stone apparently interviewed during the investigation of the Complaint.

13Numerous applications were rejected prior to June 10, 1977. (Exhibit #16; eg. Vol. 4, letter of rejection to Mr. T. Ireland, dated May 26, 1977). However, other applications were considered after that of Ms. Blake. Mr. Darcy Goodwin submitted an application a few days after Ms. Blake and was interviewed on June 17, 1977. (Exhibit #16, Vol. 4, pages 784–89). Mr. Goodwin was therefore interviewed about a week after Ms. Blake was summarily rejected. Messrs. Douglas Holmes and Robert More both applied on June 17, 1977 and make references to a newspaper as having brought the vacancy to their attention. (Exhibit #16, Vol. 4, pages 878–80; Exhibit #16, Vol. 6, pages 1372–77).

14Others, including Richard Panetta and Joseph Mattucci, who submitted applications prior to the deadline were only rejected approximately a month after submitting their applications. (Exhibit #16, Vol. 7, pages 1478–80 (Panetta); Exhibit #16, Vol. 6, pages 1256–58 (Mattucci)).

15There was some testimony vaguely suggesting, by way of hearsay and about uncertain incidents, that Mr. Stone might hold racist views, but it was not nearly sufficient to establish a prima facie case of discrimination. (Transcript, Vol. II, pages 3–95, 98, 99; Vol. III, pages 1718; Vol. III, pages 96, 97, 101–102, 105–106, 160, 184–185). Mere suspicion of racist proclivities is not sufficient. (See Fuller v. Candur (1981), 1981 CanLII 4318 (ON HRT), 2 C.H.R.R. D/419, para 3740; Mitchell v. Nobillium (1982), 1981 CanLII 4319 (ON HRT), 3 C.H.R.R. D/641, para. 5762; and Singh v. Domglas (1981), 1980 CanLII 3929 (ON HRT), 2 C.H.R.R. D/285, para. 2509). Similarly, there was only vague evidence, at most, suggesting Mr. Starnes, the Superintendent prior to Mr. De Grandis, did not want to hire females. (Transcript, Vol. III, page 32; Vol. II, page 113).

16On or about July 28, Ms. Blake telephoned Mimico to make inquiries, and spoke to Mr. Kennedy, (Transcript, Vol. V, page 12). The Respondents conceded that Mr. Kennedy's conduct to Ms. Blake was not appropriate, and, in fact, when his conduct was brought to the attention of his Superintendent, he was reprimanded for it. (Transcript, Vol. II, page 33; Vol. VI, pages 164–165; Vol. IX, page 94). Mr. Kennedy had told Ms. Blake that Mimico was a male institution and asked "What do you think you can do here." (Complaint Exhibit #3).

17Whatever Mr. Kennedy's proclivities or state of mind, (and he was the one individual within Mimico's management in respect of whom there was considerable evidence from witnesses that he held racist views) his statements did not discourage Ms. Blake from putting in an application, nor did they have any bearing on the Respondents.

18Ms. Blake filled out a second application form dated July 28, 1977 in response to another advertisement of July 27, 1977. She was not called for an interview and in spite of attempts to ascertain the status of her application, was unable to elicit any information. (Transcript, Vol. II, pages 35–38; Exhibit #11).

19Mr. Kennedy had no part of the recruiting process, and there is no evidence that he in any way ultimately influenced anyone's decision with respect to Ms. Blake. (Transcript, Vol. VI, page 151). Even where an inference is drawn that an individual was biased, where there is no evidence before a tribunal that the individual influenced the decision in question, it must be concluded that the individual's bias was not an "operative element" in the decision. (See Fuller v. Candur, supra, para. 3739). Ms. Blake never received a letter of rejection in respect of this second application. The application was handed to an unidentified female person, who worked at the switchboard. There is no evidence as to what happened to the application subsequently, particularly whether it came to the attention of Mr. Burns, or anyone else in management or personnel. The form was not date stamped, which suggests it may have been mislaid or misfiled. In the ordinary course of things, the application would be directed to the personnel office, not to Mr. Burns. (Transcript, Vol. VI, page 158).

20Whatever happened to the form, there is no evidence to link it to Mr. Burns in particular. While Mr. Burns had had some involvement in hiring at one point, the evidence shows that Mr. De Grandis made a deliberate decision to remove him from that process because he had no operational responsibility for correctional officers. The point in time at which Mr. Burns completely ceased to have any responsibility for hiring is unknown, but by late July, his responsibilities were little or nil in that area. (Transcript, Vol. VI, pages 150–152, 156).

21The fact that Mr. Kennedy referred Ms. Blake's call to Mr. Burns is not probative of Mr. Burns' involvement in hiring. Mr. De Grandis explained that it would be quite normal to refer a call of that nature to anyone in management other than the superintendent, because of the natural reluctance to bother the superintendent. (Transcript, Vol. VI, page 156; Vol. IX, page 53).

22After putting in her application, Ms. Blake attempted to obtain an immediate interview and began a series of phone calls to Mr. Burns about her application, but was never able to reach him. Given all the circumstances, that is not so surprising. Mr. Burns, as part of his job, was expected to be out of his office the greatest portion of his time. When in his office, he would not always be available. Ms. Blake apparently spoke to a receptionist and never left a message asking Mr. Burns to call her, even though she was available at home much of the time. (Transcript, Vol. II, page 67; Vol. VI, page 158).

23One cannot draw from that evidence the deduction that Mr. Burns deliberately refused to take Ms. Blake's calls because of her colour, age, sex, race or origin. From his point of view, it is as reasonable to conclude (assuming the receptionist forwarded all the calls to him), that he was on the receiving end of a series of phone calls from an unknown female who would not leave a message or means of returning the call, about an application he had no apparent knowledge of. As a peripheral participant in the hiring process of correctional officers, his incentive to have an interest in or to give priority to those calls would be low.

24Evidence was led, presumably as "similar fact," to show that Mr. Burns had ignored another applicant, Mr. Lionel Grant, a non-white male, in a similar situation. Lionel Grant, born in Jamaica, and a former policeman in that country, came to Canada in 1967 and sought a position with Mimico as a correctional officer in 1974. He testified he had a great deal of difficulty in seeing Mr. Burns, in charge of personnel, for an interview, and that when one was scheduled eventually through the initiative of the Regional Director, he waited two to three hours for Mr. Burns, but was ultimately interviewed by someone else who hired him. Mr. Grant said he later learned who Mr. Burns was, and that in fact Mr. Burns had passed him while Mr. Grant had been waiting for his interview. (Transcript, Vol. III, pages 52–55).

25The suggested inference is that Mr. Burns did not wish to hire non-whites and therefore ignored them. However, Ms. Blake's second application does not make any reference to her race or to her origin or to any place of education or employment outside Canada. Unlike her first and third applications, the fact that Merle Grove High School is in Jamaica is not mentioned. It would be difficult to suggest that Mr. Burns had any idea of Ms. Blake's colour, race or origin. (Exhibit #11). Mr. De Grandis testified that in July, 1977, Mr. Burns was not involved in hiring. (Transcript, Vol. VI, page 150). As well, Mr. Grant's evidence regarding Mr. Burns was premised on hearsay. Mr. Grant was told by someone, not at Mimico, of an appointment for Mr. Grant with Mr. Burns but when Mr. Grant arrived at Mimico, he was dealt with by a receptionist. There is no first-hand evidence that Mr. Burns had any idea who Mr. Grant was or why he was there although one can have suspicions about the situation, and given the very frustrating circumstances related by Mr. Grant it seems that Mr. Burns was very indifferent and discourteous toward him in keeping him waiting for an interview. (Transcript, Vol. III, pages 51–52).

26In any event, Mr. Burns' alleged ignoring of Mr. Grant is hardly probative of bias against females. There was no evidence that Mr. Burns displayed discriminatory attitudes towards females as correctional officers or that he had behaved in a discriminatory manner towards them. In fact, Ms. CasaDiBari testified that he was partially responsible for hiring her, first on a part-time, and later on a full-time, basis. (Transcript, Vol. III, pages 29, 32).

27From the facts that first, Ms. Blake made an application on July 29 (Exhibit #11) which gave no hint of her colour or race, second, that she telephoned Mr. Burns several times, was unable to reach him and left no message to call her back, and third, that a black male had difficulty one day in seeing Mr. Burns, this tribunal, was, in effect, asked to conclude that Mr. Burns received and examined Ms. Blake's application, decided not to interview her because of her colour, race, sex and/or age; and that Mr. Burns received a series of messages, connected those messages to the application, and decided not to speak to Ms. Blake because of her colour, race, sex, age and/or origin. However, there was no evidence that the application form ever came to his attention or that he received the messages or that he was available when she called or that he (or anyone in management) knew that she was non-white or that he was in the habit of or had on other occasions ignored calls from females or people over forty, or indeed, that he had any significant responsibility for hiring at that time. It is one thing to draw conclusions about a person's state of mind or motive or intention from a series of surrounding facts or evidence of habitual conduct, it would be quite another to draw conclusions of fact simply from the Complainant's sense of annoyance or from her perception of things. Perhaps Ms. Blake's second application was never actually rejected. Mr. De Grandis was attempting to rationalize the recruiting and interviewing system, admittedly with only partial success. The recruiting system had been, at the least, very haphazard and disorganized. One of Mr. De Grandis' objectives was to keep applications on file for a period of time so that they could be referred to as the need arose. (Transcript, Vol. VI, pages 165–166). The fact that Ms. Blake's second application was never located in the subsequent investigation suggests that it may have been misplaced or discarded in the systematic recruiting system then in place.

28Applicants were not necessarily immediately rejected when vacancies had been filled, many applications in response to the July ad being held for some time with the applicants not being given an interview until late October or November, or then being rejected. As Ms. Blake never received a rejection letter regarding her July application, Ms. Blake's application, like many others, may well have been held for further consideration but later misplaced or discarded. (Exhibit #16; Transcript, Vol. IV, pages 170, 175).

29Ms. Blake filed a Complaint dated August 23, 1977 (Exhibit #4).

30In November, 1977, she saw a newspaper advertisement for correctional officers and upon telephoning the Ministry of Correctional Services was told that employment was being done at Mimico Correctional Centre. She called Mimico and was told to address her application to the personnel officer, Mr. Van Horne, which she did, her third application being dated November 24, 1977, (Exhibit #12). At that point, Mr. Van Horne apparently knew of Ms. Blake's complaint. (Transcript, Vol. II, page 69). Mr. De Grandis had learned of the Complaint in early September, 1977, through a telephone call from The Human Rights Commission, and had requested a memo from Sergeant Kennedy which was done September 8, 1977 (Exhibit #45). As of November, 1977, it was either Mr. Aird or Mr. De Grandis, but mainly Mr. Aird, who was interviewing people, however, they interviewed a good many people together about the end of November, 1977. (Transcript, Vol. VI, pages 67; 169–172). Ms. Blake was called in for an interview within a few days with Mr. De Grandis. There was no essential difference in the testimony of Ms. Blake and Mr. De Grandis as to what took place at the interview (Transcript, Vol. II, pages 42, 43, 71). Ms. Blake was interviewed at 5:20 p.m., November 29, 1977, in the regular course of Mr. De Grandis' interviewing. As Mr. De Grandis knew at that time of Ms. Blake's complaint, it might be suggested that he would be biased against her so as to justify Mimico's rejection of her earlier applications. However, Mr. De Grandis impressed me as someone who would impartially and objectively interview someone in Ms. Blake's position, and who would admit an error on the part of Mimico if such there was.

31Mr. De Grandis was concerned that inmates would have the perception of Ms. Blake that she could be manipulated (Transcript, Vol. VI, page 119). He asked her the questions normally asked of an applicant, but did not find her answers satisfactory. As his interview of her was critical to the issues in this Inquiry, an extensive quotation is appropriate:

Q. And, tell us about the interview, and what you decided, after interviewing Ms. Blake?

A. Well, I interviewed Ms. Blake, and as far as I can remember, all the questions were in the same topic area, with the same intent as anybody else, any of the other people that I interviewed. I remember asking her, as I indicated earlier, asking her questions that would give me some idea of the common sense of the general approach to typical situations that might occur in the institution, and tried to get an idea of her appreciation for what a correctional centre might be like, what appropriate response, vis-a-vis, large groups of people might be, and, also, try to get an appreciation whether she was able to determine that a response that might be appropriate in many situations, and many work environments, might, at the same time, be inappropriate in a correctional institution.

That's the kind of questioning, as I have indicated, that I looked for, in interviewing any candidate.

I recollect asking Ms. Blake about racial slurs, directed to her, by inmates, and how she might handle that sort of thing.

My recollection is that she indicated that she had had to deal with that unfortunate situation many times. She had learned to ignore it, and would do the same.

It's not the answer that I was looking for, and it's not the general train of thought that I was looking for, simply because I don't expect my officers to ignore inappropriate conduct of any kind, whether they themselves can put up with that conduct or not is not important.

That kind of feeling, that kind of permissiveness if you want, were allowed to invade a group of 30, 40, 50, 60, 70 people, under one officer's control, and, perhaps, ... not perhaps, but most likely, invade the population, can draw from it, more testing ... perhaps more serious testing, of the officer's reaction, and how much they would be allowed to get away with, until such a point where that testing can turn into some very, very serious consequences.

I was looking for someone that would say, "I don't know your specific regulations. I don't know your specific approach to these things, but it's obviously something that cannot be tolerated, and would have to be dealt with. I would hope to get training in the appropriate way of dealing with it, but I would not ignore it."

That's, generally, the kind of answer that I would look for, in that regard.

Q. You were here when Ms. Blake gave her testimony. She indicated you asked a question about coming upon a group of naked men in the shower.

What was the purpose of that question?

A. One of the things that occurs often in a correctional institution is that, for any number of given reasons, the Lieutenant will have to call, at the spur of the moment, for a count of the institution.

And all activities, at that point, become secondary to officers counting the population, whether it's in recreation, whether it's in work gangs, whether people are showering at that time, and what have you.

And the officer, male or female, has to go into areas to count people, because, normally the reasons for such counts being taken is that we suspect someone has run away from the institution, we have got a tip that a neighbour has seen someone climbing the fence, or that sort of thing, and, in order for us to raise the proper alarm, in order for us to give the police an immediate description ... we have a tie-in with the police, a direct line, a no-dial line, that will allow us to instantaneous contact ... in order for us to do that, we have to determine: a) is anybody missing; b) who is missing; and c) draw that file, and get that description out on the air, so that people will know who, and what, they are looking for.

Well, if an officer is in charge of a dormitory, and a bunch of people are showering, we can't wait for them to be counted. We have to get in and take a count.

And what I was looking for, since that it is a likely occurrence, that could occur at any time, many times a week, what I was looking for is some indication of an appreciation that that kind of entrance into a shower can be embarrassing, not only for the officer, male or female, but can also be embarrassing for the inmate; and therefore, I was looking for some sort of indication that there was an appreciation for that embarrassment, and that there was some thought as to how it might be done.

I have had many people answer, to that type of question, and that would be asked of male and female applicants, many people would say, "I will give a big yell, telling them I am coming, and cover up," or something to that effect, which gives me an idea, "Hey, there's an appreciation that that kind of thing, you know, is not a run of the mill thing, and can have some backlash."

I didn't get that kind of answer from Ms. Blake ... certainly, not in the appreciation of what those things can mean.

Q. And, you heard her testimony about that question regarding the appearance of an intoxicated, or drunken, colleague, at work.

What is the purpose of that sort of question?

A. Well, again, that's a question that I ask, and most people would ask, of an applicant, because, obviously, your best friend ... in fact, in most situations, the only people you can count on, are the officers beside you, and their ability to react, their ability to survey the situation, and act accordingly, and, in some situations, he's the only, or she's the only help that you will get for a matter of minutes, in those crisis situations. The first few minutes are the most important.

And so, my reason for asking that question, and I usually phrase it something ... you have been on shift, you are walking along the corridor, and you, in your conversation, or smell, you sense that this fellow is intoxicated, or this person is intoxicated.

How would you handle that?

At the very least, I would expect the individual to say, "I would report him immediately, to the Lieutenant ... not to allow him to take his duties," because a person so intoxicated, or so under the influence of alcohol, quite likely will have impaired judgment, quite likely will not have the instantaneous reaction that is required.

I didn't get that kind of answer from Ms. Blake. I got the kind of answer that she would attempt, somehow, to counsel, or talk to this fellow.

Well, that, in fact, may be appropriate, at some other time, but certainly not at the point of going on duty.

Q. Ms. Blake indicated that she had some discussion with you as to the relevancy or the background that she had in psychiatric nursing, and whether that was helpful, or similar, to what she was dealing with here.

What's your comment on that?

A. Well, I don't believe it is relevant in any way that I have been able to determine, through my experience.

Q. Why do you say that?

A. Because I just don't see the two roles, the psychiatric role that places, I believe, and I am not an expert in the field, but my appreciation of that, is places a great deal of emphasis on counselling, on trying to work through problems on, generally, a one-to-one basis ... certainly for smaller groups than what we are used to

... Has a greater emphasis in terms of pathology. Has a greater emphasis in terms of "why the individual is at the stage that they are at," and that sort of thing, and that's simply not germane to the job of Correctional Officer.

There is also one other question that I do recall, and I ask that of most people, and that has to do with what you do if you come across somebody beating up an inmate.

And it's not asked because there are occasions of that kind, in the Mimico Correctional Centre, because, in the five years that I have been there, there has been only two such allegations, both of which were investigated by the police.

The reason that I ask that question is that it places obvious responsibility on the officer who may see inappropriateness, of whatever kind, to report that fellow officer to the administration ... to the Superintendent.

Some people fail to report inappropriateness, for whatever reasons they may have, and it's my determination that I want the kind of individuals who appreciate the need to report that, regardless to what it may do to them, as an individual.

I usually get two kinds of answers. I get the answer of "I would try to break it up, and then later have a word with this fellow, in private, that I didn't appreciate that, and I wouldn't put up with it any more," which leads me to the line of questioning of the individual that, "How did they get away from their responsibility to report inappropriateness?"

Or I get the second kind of answer that says, "I would report that individual to the powers to be," which then would lead me to a line of questioning of "How are you going to, then, face the consequences of your fellow officers, and what they may feel about you as someone who would report them, and may have some fear that you are looking at them in order to report them to the Superintendent. How would you handle that kind of tension between yourself and fellow officers, in a staff room, in a locker room, that requires inter-action?"

And it gives me a nice insight, I think, an appropriate insight, into the train of thought, the appreciation for some of these problems of the individual.

And I think that's the kind of individual that we need in that kind of work.

Q. And, do you recollect Ms. Blake's response?

A. I don't recollect it word for word, but I believe she opted for the counselling aspect later on ... the talking to them.

I have had one individual, over the years, who responded he would join in, so ... he didn't get hired.

Q. I am glad to hear that.

A. M'hm.

Q. Now, at the conclusion of the interview, Mr. De Grandis, or at the point at which you made your decision, what was your judgment of Ms. Blake's ability to function in the future, as a CO?

A. I thought she was a nice lady. Nothing that has happened in the past week or so has changed my mind on that. I have had several general discussions on whether, and so on, and I thought ... and the kind of work she is doing now, I thought she was a nice lady.

I just came to the conclusion that she simply didn't have the characteristics, the ability, to monitor, to direct, to lead, to order, to ensure that certain difficult tasks were done by inmates, in the large numbers that would come under her sole direction, at the Mimico Correctional Centre.

Q. And what made you come to that conclusion?

A. I think her voice, was one. I have heard the term in here used, "soft-spoken." I don't believe I ever used the term "soft-spoken," because I have many soft-spoken people, and if the term "soft-spoken" is used as an opposite of yelling and screaming, I certainly don't want any yellers or screamers.

But there is a tone ... a quality to the voice, that says, "I am in charge here. I have certain things that have to be accomplished, and you people are going to have to do them, and I don't have a particular choice in what needs doing, and neither do you."

And that was missing, in my best judgment. In my best judgment, I felt that she was the kind of individual, whether, in fact, she would turn out to be or not, she was the kind of individual that every new inmate would attempt to test ... would attempt to find out just how far they could go, and what they could get away with.

And that kind of reaction on the part of new inmates, has a way, again, of filtering through a population, and leading to events that should not be led to, and should be avoided.

As I say, it's not a matter of yelling and screaming, it's just a quality, a tone, in the voice. It's certainly one of the reasons why.

As I said, the questions led me to believe a lack of appreciation of possible consequences, in that kind of environment.

It was just a general feeling on the interview, a general feeling that, in my best judgment, it would be unfair to hire her ... unfair to her, because I thought she would have been involved in things that would have been, perhaps, of her own doing, or partly, that she need not be involved in.

Certainly unfair to the other staff, who would have to deal with possible repercussions of her dormitories coming out of order, and unfair to the inmate, because there could be a mood, in groups under her direction that, people who might not want to get involved in events that started, and in that environment, it's very much like a snowball ... it starts with one or two, and all of a sudden, 35 and 40 are involved.

Unfair to the inmates, that I would allow the possibility of that situation occurring, people who might not otherwise have got involved ... there's no inmate in the world who is going to sit in a dormitory of eight, or nine, or ten, people being uproarious, and say "No, I don't want to."

He does so of his own peril, because the other people, then, may, in fact, turn on him. so you have to keep, to whatever extent you can, and I submit that we have done so at Mimico for five years, when one looks at what happened, before, there.

You have to keep the possibility of those events down to a very, very bare minimum, and you can't, if, in your best judgment, the individual may contribute to that, you can't allow that to happen. (Transcript, Vol. VII, pages 81–92).

About a week later Ms. Blake received a letter dated December 6, 1977, from Mr. Van Horne (Exhibit #13) saying she had not been accepted. She had already telephoned Mr. Van Horne who told her Mr. De Grandis "said my voice was too soft and it would not command authority" (Transcript, Vol. II, page 48).

32Ms. Blake then went to see the personnel officer for the Ministry who told her to write to Mr. De Grandis, asking him if he might reconsider, and Mr. De Grandis replied December 12, 1977 (Exhibit #14) to the effect that he could not.

33Vincent Ferreira, born in Dominica, West Indies, testified. He came to Canada in 1978, and applied to be a correctional officer with Mimico (Exhibit #16; Transcript, Vol. III, page 631; also marked as Exhibit #17; Transcript, Vol. II, pages 85, 86), receiving an interview, apparently with Mr. Van Horne. Mr. Ferreira was hired, commencing work July 31, 1978.

34Mr. Ferreira testified that he received racial insults from a Mr. Quinn, a head shift officer at Mimico (Transcript, Vol. II, pages 95, 98, 99). It is obvious that Mr. Ferreira complained to Mr. De Grandis, and Mr. De Grandis reprimanded Mr. Quinn and Mr. Ferreira was quite satisfied with the way Mr. De Grandis handled the matter. Mr. Ferreira testified that Mr. Quinn has been quite respectful of him since. (Transcript, Vol. II, pages 100, 104–106).

35Mel Sukhu testified. He was born in British Guiana, coming to Canada in 1968, and then completing a university degree. He obtained a job at Mimico as a correctional officer in 1974 under Superintendent John Main, who was followed by Superintendent Ian Starkie. Mr. Sukhu testified that his impression was that more women and non-whites were hired when Mr. De Grandis later became Superintendent. (Transcript, Vol. II, pages 115, 117–121).

36Ronald Brett became office manager at Mimico in April, 1978. He testified that the system which pre-dated his arrival at Mimico for handling employment application forms was that they were date stamped upon receipt and would be put into two stacks – one for acceptable applications, and one for non-acceptable applications (due to less than Grade 12 education or criminal records) in respect of which a letter of rejection was sent immediately (Transcript, Vol. II, pages 124–126). Under that approach, Ms. Blake would, of course, qualify for an interview in respect of her first two applications. Mr. Brett prepared, at the request of the Commission, a list of those who left the employment of Mimico as correctional officers, over 1977 (Exhibit #18).

37David McClary started to work as a correctional officer at Mimico, commencing March 1, 1976. He testified that he overheard a conversation between Assistant Superintendent Stone and a Sergeant Doherty at Mimico in 1977 when one of them (he was not sure which one – but thought it was Mr. Stone – Transcript, Vol. III, pages 17, 18) made a racially derogatory remark. His evidence did not add anything.

38Mr. Lionel Grant testified that there were racial "slur remarks and puns" at Mimico by management, including Messrs. Kennedy and Quinn. (Transcript, Vol. III, pages 83,84).

39Thomas Allan Tangie, a non-status Canadian Indian, testified. He commenced work at Mimico October 6, 1975. He said he could not recall much in the way of specifics about racial slurs but felt the non-white correctional officers were given the least pleasant tasks. (Transcript, Vol. III, pages 96–100).

40Dianna CasaDiBari was hired as a correctional officer at Mimico in April, 1975. A cousin, Rosemary Neilson, a Linda Ashley and a Mrs. Evans were correctional officers there when Ms. CasaDiBari started. (Transcript, Vol. III, pages 34–43).

41Natalie Dudnik worked at Mimico as a correctional officer from November, 1977 to December, 1980. (See Exhibit #16, Vol. III, pages 537–539 for her application form; also entered separately as Exhibit #19). She was interviewed by both Mr. Aird and Mr. De Grandis before being hired. Shortly after she started, another female correctional officer, Maureen Ryan, was hired. (Transcript, Vol. III, page 115).

42Norman Malcolm testified. He was born in Jamaica, but came to Canada as a baby, graduating in engineering from the University of Buffalo, then returning to Canada where he sought work as a correctional officer January 31, 1978. He was later interviewed and hired by Mr. Aird at Mimico, where he is still employed (Transcript, Vol. III, pages 126,127). Mr. Malcolm was a reticent witness who did not want to recall much, but did state that Sergeant Kennedy had used racial slurs and names against him. He testified that Sergeant Kennedy was the only member of the management of Mimico to use racially abusive language toward him. (Transcript, Vol. III, pages 136–147). Mr. Malcolm testified that Mr. De Grandis has always treated him fairly and that he has full confidence in Mr. De Grandis. (Transcript, Vol. III, pages 150–151).

43Kingsley Lyn, 40, testified. He was born in Jamaica and came to Canada in 1966. In 1972 he became a correctional officer at the Toronto Jail, and became a casual worker at Mimico in 1976. He expressed his view that there were some problems with former management in respect of hiring non-whites. He said that racial remarks were made by Sergeant Kennedy, but Mr. Lyn was also of the view that Superintendent De Grandis deals with people fairly and honestly, is considerate of others and does not discriminate unlawfully, and had disciplined an employee who had made a racial slur. (Transcript, Vol. III, pages 162–164; 168–171; 183; 196–197).

44Kermit Lyn, 49, came to Canada from Jamaica in 1957, and became a correctional officer at Mimico in 1973. He said there were problems with some of the management staff toward blacks, and that he was falsely accused by Mr. Quinn in respect of an altercation that Mr. Lyn had with a white guard. (Transcript, Vol. III, pages 203–209). He felt that there was prejudice on the part of Messrs. Starkie, Burns and Kennedy. (Transcript, Vol. III, pages 214, 215). However, he also testified that Superintendent De Grandis had put a stop to racial slurs. (Transcript, Vol. III, page 213).

45Mr. Pierce Emanuel Edesse Cabon, a registered nurse, born in Mariutius, has worked at Mimico since 1975. His testimony as to Superintendent De Grandis' dealings with Mrs. Jean Barnett, the head nurse at Mimico, was not relevant to the issues of the inquiry. (Transcript, Vol. IV, pages 8, 9; 18–23). Mrs. Jean Barnett, 67, the retired head nurse at Mimico, testified. While she and Mr. De Grandis did not see eye to eye on everything, her evidence was not helpful to resolving the issues in this Inquiry. He testified that Sergeant Kennedy would make racial slurs and engage in racial name-calling. (Transcript, Vol. III, pages 36–38, 59, 60, 57).

46Victor Bedeau, a former policeman from Trinidad, has worked in the Don Jail since 1975. He worked at Mimico on a casual basis for three weeks in 1976, and expressed an interest in working full time, but said management (Superintendent Starkie and Mr. Burns) said there were no vacancies when in fact there were and others were being hired to fill them. He accused Lieutenant Ross Kennedy of making racial slurs while he was at Mimico. Mr. Bedeau then went with the Metro Police Department for a while, later returning to Mimico on a casual basis, unsuccessfully seeking permanent work, although others were being hired. Mr. Bedeau said he was hired to work on a permanent basis at the East Detention Centre, starting in March, 1977. (Transcript, Vol. III, pages 85–92).

47Ms. Fern Gaspar testified. She is an experienced, competent Human Rights Officer, being stationed in Toronto since 1975. She undertook a very difficult task in reviewing the mass of data with respect to application for the position of correctional officer at Mimico. The Respondent readily gave her access to Mimico's files.

48Ms. Gaspar took charge of the matter for the Commission in September, 1979. At that time, she knew from the previous investigator's work that few women were being hired as correctional officers at Mimico, specifically only 2 of the 46 persons hired for the period that had been reviewed, although 45 percent of the rejected applications for the period were from women.

49Deciding that more information was necessary, Ms. Gaspar attended at Mimico to review its application forms for the period May, 1977, to May, 1978. She perused the application forms in Mimico's eleven folders (Exhibit #25), and had some of them photocopied, making notes of the rest. For example, in respect of folder #2, Ms. Gaspar determined there were 124 applications, of which she photocopied 55 and had made notes (Exhibit #27) of the rest when she first went to Mimico's offices. (Transcript, Vol. IV, pages 106, 107, 112). The materials in the folder generally indicated the gender of each applicant. An "Index of Applicants" (Exhibit #26) was prepared.

50The photocopies of the applications that were photocopied by Ms. Gaspar were bound in nine volumes (Exhibit #16). The list of names of applicants was entered separately as Exhibit #31.

51In addition to the 46 persons accepted for employment, as first determined by Ms. Gaspar, the Respondent later advised that a further five had been hired, (Exhibit #29), making a total of 51 hired for the period May, 1977 to May, 1978. (Transcript, Vol. IV, pages 121–123).

52Mrs. Gaspar was trying to understand why some persons would be interviewed and others not, and why after the interviews some would be hired and others not (Transcript, Vol. IV, pages 131, 132). From her initial impression, Ms. Gaspar thought that some women were rejected who should have been accepted, (Transcript, Vol. IV, pages 134–138; 140–142; 165–176), and thought that some from third world countries had experience that suggested interviews should be granted, but they were not interviewed. It seemed that white males, who were large in physical stature, had the best chance of being interviewed. (Transcript, Vol. V, page 10).

53A list of those who applied between May and October, 1977, was prepared (Exhibit #32). For this period, of the 143 male applicants, 19 were interviewed, (and about 15 hired) while for the 63 female applicants none were interviewed during that period of time (Transcript, Vol. IV, page 146), although a few may have been interviewed later (Transcript, Vol. IV, page 176).

54A list of applicants sent rejection letters January 9, 1978, was filed as Exhibit #33.

55Ms. Gaspar prepared a chart with details on the backgrounds of 46 hired applicants (Exhibit #34). They included at least one black, Mr. Malcolm, who was a witness in the hearing (Transcript, Vol. V, pages 9, 10).

56The criteria for hiring included a completed Grade 12 and a consideration of previous experience in related fields. Since 1978, the height restrictions have been removed (Exhibit #35; Transcript, Vol. V, page 15).

57Hugh McLennan, Assistant Regional Personnel Administrator at the Ministry of Correctional Service for the Central Region (including Mimico) in Toronto, testified. His duties include the recruitment and staffing of correctional officers. Before November, 1980, most institutions, including Mimico, did their own recruiting. The criteria (Exhibit #35) are Grade 12, relevant work experience, such as in the related fields of the military, police or the security field, and working with people (such as teaching, with juveniles, or in nursing), maturity and physical size and strength. (Transcript, Vol. V, pages 120).

58Since November, 1981, there is a central recruitment process whereby applications are received by the government and all applicants with a completed Grade 12 are invited to an information centre, given an overview of the role of a correctional officer (through a slide show and question and answer session) and tested (with an aptitude test, and a 100 word essay written).

59The applicants remaining, after low scores are screened out, are invited for an interview by a panel of three, chaired by Mr. McLennan. There are no minimum standards in terms of height and weight. (Transcript, Vol. V, pages 131–138). Moreover, while there is certainly not any formal affirmative action policy in favour of women, the present Ministry hiring practice seems to favour females to the extent that departing female correctional officers are generally replaced by females, and the general intent is "towards putting more females in positions throughout the Ministry." (Transcript, Vol. V. pages 129. 130). It is clear that the central recruitment process in place since November, 1981, is vastly more organized, systematic and professional, than the recruitment process in place at Mimico before then, and particularly with respect to the process before Mr. De Grandis brought some changes in mid 1977.

GENERAL LAW APPLICABLE

60The onus of proof to establish on the balance of probabilities a breach of the Code lies with the Complainant and the Commission. Once a prima facie case of discrimination has been established, the onus shifts to the Respondent to rebut the Complainant's prima facie case or otherwise establish a defence. (The Ontario Human Rights Commission v. The Borough of Etobicoke (1982) 1982 CanLII 15 (SCC), 3 C.H.R.R. D/781 (S.C.C.) at D/783); Bhinder v. Canadian National Railways (1981) 1981 CanLII 4297 (CHRT), 2 C.H.R.R. D/546 at D/557). Should the Respondent lead evidence of a non-discriminatory reason for refusing to employ the Complainant, the Complainant and Commission can still establish that the reason advanced for non-employment is in fact a pretext, and that discrimination on an unlawful ground was one of the operative reasons for the Respondent's actions. (McDonnell-Douglas Corporation v. Green (1973) 411 U.S. 792, at 804).

61Discrimination can be established by direct evidence or by circumstantial evidence (Sweeney v. Board of Trustees of Keene State College (1978) 569 F.2d 169, 177 (U.S.C.A.); Retail Wholesale and Department Store Union and National Automatic Vending Ltd. (1963) 63 C.L.L.C. 1161, 1162).

ESTABLISHING A PRIMA FACIE CASE OF DISCRIMINATION

62In the case at hand, a prima facie case of discrimination under the Code could be established by the complainant and Commission showing either simply through witnesses that there was discrimination, (as is the usual mode of adducing evidence in human rights cases), or alternatively, showing through statistical evidence (a topic to be discussed at length infra) that there was discrimination, or through adducing both oral testimony and statistical evidence together to establish a prima facie case of discrimination.

63Ms. Blake, being a black female, belongs to a racial minority in Canada, and is of a gender that is often discriminated against in employment. Moreover, it is clear from the evidence that while on paper, at least, she appeared to have the formal qualifications to be considered for an interview further, after applying for the position of correctional officer at Mimico, she was rejected, (with respect to her first two applications, without an interview, and in respect of her third application – after making a Complaint – after an interview) and after her rejection each time the Respondent continued to seek applicants for the employment position.

THE LAW IN RESPECT OF THE INTRODUCTION OF STATISTICAL EVIDENCE IN HUMAN RIGHTS CASES

64Often discrimination is not overt. Rarely does an employer expressly state that it refused to hire a qualified applicant because she was a woman. Acts of discrimination and intent to discriminate are often proved by circumstantial evidence (Re: Windsor Board of Education and Federation of Women Teachers' Associations of Ontario (1982), 1982 CanLII 5080 (ON LA), 3 L.A.C. (3d) 426, at 430). "Statistical evidence is an important tool for placing seemingly inoffensive employment practices in their proper perspective" (Senter v. General Motors Corp., 532 F.2d 511 (1976) at 527, cert. denied 429 U.S. 870 (1976)). Statistical evidence is commonly used in American discrimination cases but has only recently been adduced at Canadian human rights hearings. In the case at hand, statistical evidence was sought to be used as the significant component of the evidence in seeking to prove discrimination, and in this respect the case would seem unique to Canada. For this reason, it is pertinent to consider the extensive American case law with respect to the question of statistical evidence in human rights cases.

a. CIRCUMSTANTIAL EVIDENCE

65It was suggested at the hearing that statistical evidence was a form of similar fact evidence and should be subject to the strict rules of admissibility applied when similar fact evidence is adduced. A similar argument was rejected in Re: Windsor Board of Education and Federation of Women Teachers' Association of Ontario (1982), 1982 CanLII 5080 (ON LA), 3 L.A.C. (3d) 426:

It must be borne in mind that similar fact evidence is concerned with "facts or transactions similar to the fact or transaction directly in issue." ... It appears to me that statistics, by their composite and general nature, cannot be concerned with specific "facts or transactions" and, therefore, they could not be admitted as similar fact evidence (Supra, at 430).

66I agree. Statistics show patterns of conduct rather than. specific occurrences. Statistics represent a form of circumstantial evidence from which inferences of discriminatory conduct may be drawn (Davis v. Califano, 613 F.2d 957 (1979) at 962). It is within the rubric of "circumstantial evidence" that statistical evidence in human rights cases should be considered. Like all circumstantial evidence, statistics are to be considered along with all surrounding facts and circumstances (International Brotherhood of Teamsters v. U.S., 97 S.Ct. 1843 (1977), at 1857).

67Statistical evidence may be used in a number of ways to buttress both complainants' and respondents' cases. Statistics may show racial or sexual disparities in decisions to hire, promote (Teamsters, supra; Croker v. Boeing Co. (Vertol Div.), 437 F. Supp. 1138 (1977); Rich v. Martin Marietta Corp., 467 F. Supp. 587 (1979)) or dismiss (Ingram v. Natural Footwear Ltd. (1980), 1980 CanLII 3904 (ON HRT), 1 C.H.R.R. D/59) employees. They may show disparities between the number of women employed in a particular job and the number of qualified women in the labour market (Offierski v. Peterborough Board of Education (1980), 1980 CanLII 3898 (ON HRT), 1 C.H.R.R. D/33; Windsor, supra). They may show that subjective and discretionary decisions by employers are being made in a discriminatory manner (Senter, supra, at 529; Phillips v. Joint Legislative Committee etc., 637 F.2d 1014 (1981), at 1026; Brown v. Gaston County Dyeing Machine Company, 457 F.2d 1377 (1972) at 1383; Lilly v. Harris-Teeter Supermarket, 545 F. Supp. 686 (1982) at 708; E.E.O.C. v. Akron National Bank, 497 F. Supp. 733 (1980), at 745; (Segar v. Civiletti, 26 E.P.D. 21,415 (1981), at 21,434; and (Payne v. Travenol Laboratories, Inc., 28 E.P.D. 24,862, at 24,877). They may demonstrate that tests and requirements imposed by an employer have a discriminatory impact (Marcotte v. Rio Algom Ltd. (1982), 1982 CanLII 4856 (CHRT), 3 C.H.R.R. D/988; Riverd v. City of Wichita Falls, 28 E.P.D. 23,776 (1982); U.S. v. Georgia Power Co., 474 F.2d 906 (1973); Griggs v. Duke Power Co., 401 U.S. 424 (1971)). They may be used to show that an employer's non-discriminatory reason for rejecting an applicant is a mere cover-up for a discriminatory reason (McDonnell-Douglas Corp. v. Green, 411 U.S. 792 (1973) at 805; Pelmer v. Parsons-Gilbane, 713 F.2d 1127 (1983), at 1137).

68It must be kept in mind that statistical evidence can be used on behalf of both complainants and respondents. A respondent employer may use statistics in its defence to show that it did not discriminate, that a discriminatory requirement is a bona fide occupational requirement, (Carson et al. v. Air Canada (1982), 1982 CanLII 4854 (CHRT), 3 C.H.R.R. D/818, at D/832) that requirements and tests are job-related, (Vuyanich v. Republic National Bank of Dallas, 505 F. Supp. 224 (1980), at 369) or in seeking to establish that the discrimination is justified (Carson, supra).

69Statistics are best used in conjunction with testimony describing the specific instances of discrimination. When individuals testify about their personal experiences with the employer, they bring the "cold numbers convincingly to life" (Teamsters, supra, at 1856). While statistics show the objective results of discriminatory behaviour, anecdotal evidence describes the behaviour and brings to light the causes of the statistically proven discriminatory results (Vuyanich, supra, at 260). Proof by oral testimony is particularly important in cases such as those involving highly trained applicants for high-level positions or those involving small companies with few employees where the labour pool of qualified persons or the number of persons hired is too small for adequate statistical study (Otero v. Mesa County Valley School District No. 51, 470 F. Supp. 326 (1979), at 332; E.EO.C. v. H.S. Camp & Sons Inc., 542 F. Supp. 411 (1982), at 443–44).

b. PRIMA FACIE CASE OF INTENT TO DISCRIMINATE

70In some cases, statistical evidence alone may be sufficient to establish a prima facie case of discrimination and shift the onus of proof to the respondent (Windsor, supra, at 431 and 438; Davis v. Califano, supra, at 862; Croker, supra; Hazelwood School District v. U.S. 433 U.S.S. 299 (1977) at 307–8). In such cases, the statistical evidence must effectively show such gross disparities in the treatment of persons of different races or sex that the disparities are unlikely to be the result of random selection (Windsor, supra, at 433; Vuyanich, supra). In the Windsor case the grievor was also required to etablish that she was qualified for the job sought and that the successful applicant was no more qualified than she was, in order to etablish a prima facie case.

71In the United States, actions are often brought against an employer who imposes a facially neutral requirement or follows a facially neutral practice which has a disparate impact on one class of persons. Intent to discriminate need not be proven. Many American cases in which statistics have been used extensively are of this type (Teamsters, supra; Coe v. Yellow Freight Systems, Inc., 646 F.2d. 444 (1981), at 450; E.E.O.C. v. American National Bank, 26 E.P.D. 21,083 (1981), at 21,089; Rowe v. Cleveland Pneumatic Company, 30 E.P.D. 27,382 (1982)). However, the American courts require proof of intent to discriminate where an individual alleges disparate treatment by an employer. These cases (Payne, supra, at 24,872; Coe, supra, at 448; Wheeler v. City of Columbus, Mississippi, 30 E.P.D. 27,444 (1982); Rowe, supra) have held that statistics alone may constitute prima facie proof of intent to discriminate but that they must be even more persuasive than statistics used to prove disparate impact (Gay v. Waiters' and Dairy Lunchmen's Union, 694 F.2d 531 (1982), at 552).

72In Ontario, as the law stands at present, intent to discriminate must be proven in all cases under the old Code (O.H.R.C. and O'Malley v. Simpsons-Sears Ltd. (1982), 1982 CanLII 4911 (ON CA), 3 C.H.R.R. D/1071 (Ont. C.A.), appeal pending before the Supreme Court of Canada). If the Complainant in the instant situation was qualified for the position of correctional officer, at least to the point of the interview stage, and the candidates who were hired at the time she was rejected were no better qualified than she was, strong statistical evidence of a pattern of discriminatory treatment of female applicants by the Mimico Correctional Institute, though circumstantial, might amount to a prima facie case of intent to discriminate against the complainant and shift the burden of proof to the Respondent. Because discrimination is often covert and intent to discriminate difficult to prove, a trier of fact must be sensitive to obvious inferences that may be gleaned from statistics.

c. REBUTTAL OF STATISTICAL EVIDENCE

73In Ingram v. Natural Footwear Ltd., Chairman McCamus cautioned that, though statistics may give rise to a prima facie case of a general intent to discriminate against women, they must be linked to the specific incident complained of (Ingram, supra, at D/68). It may be that though an employer often discriminated against women, it did not do so on the occasion in dispute. One way that an employer may rebut a prima facie case raised by statistics is by evidence that the complainant personally was not discriminated against.

74Another defence to statistical evidence of discrimination is to prove that sex is a bona fide occupational qualification. Often statistical and scientific evidence is essential to proof of this defence. (Carson, supra).

75As well as rebutting a prima facie case of discrimination against a complainant established through statistical evidence, by a respondent adducing its own evidence in its defence to show there was not discrimination even though the statistics are valid, the respondent can attack the statistical evidence as fallacious so as to destroy the complainant's prima facie case. There are several ways that a complainant's statistics may be directly discredited (E.E.O.C. v. Federal Reserve Bank of Richmond, 698 F.2d 633 (1983), at 646; Croker, supra, at 1182; Payne, supra, at 24,873; Segar, supra, at 21,432). As Mr. Justice Stewart of the American Supreme Court stated in the Teamsters case:

We caution only that statistics are not irrefutable; only come in infinite variety and, like any other kind of evidence, they may be rebutted. In short, their usefulness depends on all of the surrounding facts and circumstances (Teamsters, supra, at 1856–57).

76Several factors are pertinent in challenging statistical evidence. First, the qualifications of the statisticians and the methods used by them in compiling and analyzing the data may be challenged. Second, the disparity in the repondent's hiring practices demonstrated by the statistics may be shown to be insignificant and the result simply of random selection. Third, disparity in the number of female employees in high-level positions may be shown to be entirely the result of discrimination that occurred prior to the enactment of the pertinent human rights legislation and not the result of post-enactment discrimination (E.E.O.C. v. American National Bank, supra, at 21,090; Hendry v. L.C.B.O. (1980), 1980 CanLII 3901 (ON HRT), 1 C.H.R.R. D/160, at D/164–65; Croker, supra, at 1183; Detroit Police Officers' Association v. Young, 608 F.2d 671 (1979) at 689; Williams v. City of New Orleans, 543 F. Supp. 662 (1982) at 673). Finally, a respondent may present statistical evidence prepared by its own experts which paint a picture different from that shown by the complainant's statistics (Croker, supra, at 1190).

77Statistical evidence may not be rebutted by mere hypothesis or conjecture that the exclusion or inclusion of certain variables improperly skews the results rendering the statistical unreliable. The respondent must prove that these variables have a relevant effect on the statistical result either by reworking the complainant's data or presenting its own statistical evidence incorporating the relevant variables (Segar, supra, at 21,432–33; Trout v. Hidalgo, 25 E.P.D. 20,286 (1981), at 20,289).

d. ADMISSIBILITY

78The admissibility of statistics in civil rights cases in the United States is not disputed. As well, statistics have been entertained by a number of Canadian Boards of Inquiry (Offierski, supra; Ingram, supra; Hendry, supra; Marcotte, supra; Carson, supra). In O.H.R.C. v. Borough of Etobicoke, Mr. Justice McIntyre of the Supreme Court of Canada encouraged the use of statistics in support of testimonial evidence in human rights cases:

It seems to me, however, that in cases such as this statistical and medical evidence based upon observation and research on the question of aging, if not in tall cases absolutely necessary, will certainly be more persuasive than the testimony of persons, albeit with great experience in firefighting, to the effect that firefighting is a "young man's game". (O.H.R.C. v. Borough of Etobicoke (1982), 1982 CanLII 15 (SCC), 3 C.H.R.R. D/781 (S.C.C.), at D/784).

In the Windsor case, the admissibility of statistical evidence was challenged at the outset of a labour arbitration hearing, The board had this to say:

My initial concern in hearing the statistical evidence was that it was, according to counsel, going to be of a very sensitive nature and would likely take up many days of hearing and argument. If the evidence was clearly irrelevant it might be properly excluded. If it could have no cogency, it might similarly be excluded from this board. It is unnecessary to prolong this hearing by admitting evidence which could not be acted upon, because it could be neither cogent nor relevant. (Windsor, supra, at 428–29).

The board admitted the statistical evidence presented by the grievor as it did not "obviously lack relevancy." The board cautioned that statistical evidence should not be rejected for being singularly insufficient to create a prima facie case as this has no bearing on its admissibility (Id, at 434). Insufficient statistics when combined with testimonial and other evidence may together give rise to a prima facie case.

79Raw statistical data may also be introduced through the compiler. Statistical analysis may be introduced through the expert who conducted the analysis. Any weaknesses in the qualifications or methods of these witnesses goes to the weight to be given the statistical evidence and not necessarily to the admissibility.

80Counsel for the Respondent submitted that the person gathering the data should testify as an expert and should have no personal interest in the outcome or bias to any party. In support she cited a number of patent, trademark and obscenity cases where public opinion polls had been introduced as evidence (Zippo v. Rogers, 216 F. Supp. 670; Miles v. Frolich, 195 F. Supp. 256 (1961); Aluminum Goods v. Registrar (1954), 1954 CanLII 720 (CA EXC), 19 C.P.R. 93; People v. Franklin National Bank, 105 N.Y.S. 2d 81; R. v. Times Square, (1971) 1971 CanLII 518 (ON CA), 3 O.R. 688 (C.A.); R. v. Prairie Schooner News (1970), 1970 CanLII 1059 (MB CA), 75 W.W.R. 585 (Man. C.A.); R. v. Pipeline News (1971), 1970 CanLII 764 (BC CA), 1 W.W.R. 241). The results of public opinion surveys may easily be influenced by the biases of the surveyor and his or her lack of expertise. The surveyor makes up the questions and selects the interviewees. Here we are dealing with a different type of information collection altogether. The field of statistics involves the collection, compilation and analysis of hard data. One needs no expertise to gather and compile existing unalterable facts. Common sense can suffice.

81Any lack of understanding of the subject matter on the part of the compiler is only relevant if there is any discretionary selection of data at this stage. If there is, the weight to be given the statistical evidence may be affected but the evidence need not be rejected in its entirety merely because the compiler lacks expertise in statistics or was employed by the Commission. Public opinion polls may be rejected on the slightest hint of bias or of some other factor that might skew the results. As raw statistical data is not so easily influenced as interviewee's answers, rejection of the evidence is not necessary. Moreover, the weight to be given statistical data will not be affected by the mere assertion of challenging counsel that the compiler is biased or lacked expertise. The challenger must show, either through cross-examination or presentation of statistical evidence unaffected by bias, that the compiler's biases affected the data (Vuyanich, supra, at 256).

82In contrast to the mere compilation of data, the analysis of statistical data does require statistical expertise. Any lack of expertise on the part of the analyst affects the weight to be given his or her analysis but does not require its outright rejection.

e. ADEQUACY AND ACCURACY OF STATISTICAL EVIDENCE

83Statistics are not perfect mathematical tools for the precise proof or disproof of the existence of a discriminatory practice. All statistical measures tend to be imperfect in various ways and to varying degrees. As U.S. Circuit Judge Wisdom said in Phillips v. Joint Legislative Committee:

The best the court can do is to accept what figures are available; allow for imperfections, skewing factors, and margins of error; and then take the figures for what they are worth. Sometimes this is much, sometimes little (Phillips, supra, at 1025).

Circuit Judge Friendly in Vulcan Society of New York City Fire Dept. Inc. v. Civil Service Commission said:

It may well be that the cited figures and other more peripheral data relied on by the district judge did not prove a racially disproportionate impact with complete mathematical certainty. But there is no requirement that they should. "Certainty generally is illusion and repose is not the destiny of man" (Vulcan, 490 F.2d 387 (1973), at 393; See also Vuyanich, supra, at 255; Capaci v. Katz S. Besthoff, Inc., 711 F.2d 647 (1983) at 653).

Requiring the statistician to take into account every conceivably relevant variable and to ensure that every irrelevant variable that may possibly affect the statistical result is precluded from doing so, is to demand perfection from an imperfect science (Trout, supra, at 20,289–90). To require that the Human Rights Commission hire experts, not only to analyze the data, but also to collect and compile it imposes excessive and, in my view, unnecessary expense on the Commission (Perera v. Civil Service Commission and Another (No. 2), (1982) 1 C.R. 350 (E.A.T.), at 359). Statistical evidence may not be rebutted by mere assertions that there are deficiencies in the data base, that the data was not randomly selected, that the analyst failed to consider the effect of a relevant variable, or that an imprecise statistical method was applied. The respondent must show that these errors, omissions and weaknesses affected the statistical results in a systematic way (Vuyanich, supra, at 255–56 and 315; Capaci, supra, at 654; Trout, supra, at 20,289–90). It must persuade the trier of fact that these deficiencies rather than discrimination account for the racial or sexual imbalance indicated by the statistics. As Circuit Judge Friendly said:

While rather crude procedures of physical observation used in the surveys doubtless led to error in some cases, it is hard to believe that survey errors could have accounted for the striking racial imbalance that the results indicated (Vulcan, supra, at 392; See also Detroit Police, supra, at 687).

84The Respondent argued at length that the statistics should be rejected because the data was not randomly selected; however, it made no attempt to prove that this non-random selection affected the statistical results, but rather requested this Board of Inquiry to assume that it did. The statistical evidence cannot be rejected on mere conjecture. "(I)n most cases, conditions are far from ideal, with incomplete qualification data and non-random samples being the rule rather than the exception" (D. Baldus & W. Cole, Statistical Proof of Discrimination (1980), at 26–27).

85The Respondent also complained that some applicants applied for a position more than once and would have been counted twice. Again it offered no proof that this double counting affected the statistical results in any way (Reynolds v. Sheet Metal Workers Local 102, 498 F. Supp. 952 (1980), at 965; Payne, supra, at 24,875–76). It is unlikely that it did. White males were as likely to apply twice as were black females. Random imperfections in the data do not affect the accuracy of the statistical results (Vuyanich, supra, at 306).

86For statistical evidence to be given any weight, the raw data used as a basis must be reliable. The data must be collected competently, even if done by a lay compiler. The data must then be evaluated in accordance with recognized scientific principles and introduced in evidence through an expert witness.

87In the instant situation, the data arose from the investigation of Ms. Gaspar, the Human Rights Officer in charge of the investigation of the Complaint for the Human Rights Commission.

88She was seeking to determine and evaluate the reasons for rejection and acceptance of applicants for the position of correctional officer with Mimico over the period of June, 1977, to May, 1978, inclusive. Transcript, Vol. IV, pages 99, 100). The applications of unsuccessful candidates had been stored in boxes by Mimico on its premises (Transcript, Vol. IV, page 99). The Respondent challenged the independence of Ms. Gaspar, but I have no doubt in finding that she was an independent investigator trying her very best to determine objectively the evidence.

89Unfortunately, the information in respect of candidates was not uniform in its content or form, because there was not a standard application form always used, and because the Respondent was disorganized and unsystematic in its recruitment process and in its retention of applications and information.

90Moreover, the original applications were destroyed, in accordance with normal Ministry policy after the usual retention period had expired. This was an honest bureaucratic mistake, although an unfortunate blunder, given that Ministry officials knew there had been an investigation of its file and that there was an unresolved Complaint which might proceed to a board of inquiry at which the evidence would be needed. (Vol. VI, page 121; Vol. II, pages 152, 153; Vol. IX, pages 47, 48). All parties to a human rights proceeding are under a duty to preserve evidence which is known to be relevant to a Complaint. Moreover, if the evidence is destroyed, the person doing so cannot argue that its act of doing so is prejudicial to itself such that, therefore, the other party cannot succeed. (See Barnard v. Gulf Oil Co. (1979) 20 E.P.O. page 30,001 at page 11009).

91Therefore, Mimico cannot reply on its own actions in destroying the original applications so as to suggest, for this reason alone, that the statistical data is defective. The Respondent was on notice from the day the Commission's investigation began of the critical importance of the protection of the applications from destruction (E.E.O.C. v. American National Bank, supra, at 21,095). The Respondent must bear the risk of unsatisfactory proof created by the destruction (Rich, supra, at 609; Camp, supra, at 444).

92However, in this case the original copies of the applications would not have added much in any event. Ms. Gaspar made photocopies of most of the applications and notes of all the relevant particulars in respect of the rest, at least so far as she could honestly determine. As well, quite clearly, such additional information as might have been obtained from the original applications was not excluded for the purpose of skewing the data, with a large enough sample still being present to provide statistically reliable evidence. As well, the data omitted by Ms. Gaspar in her compilation, and then destroyed by the Respondents, would have been irrelevant as its irrelevancy in the opinion of Ms. Gaspar was the very reason she ignored it. I do not doubt the honesty of Ms. Gaspar in trying to make as thorough and objective an investigation as possible.

93Ms. Gaspar determined there were 762 applicants but there are only 595 about whom specific individual information is available, with the information on the remaining applicants never having been made available to Ms. Gaspar (not because the Respondent wanted to preclude their use but because the applications were simply lost through the disorganized record keeping system of the Respondent). An index of these 595 was marked as Exhibit #26, and the information gathered in respect of them was included in Exhibits #16, #27, #29 and #31.

94In gathering data through utilizing the 595 applications, as mentioned, the information was not uniform or complete, and the investigator, Ms. Gaspar made notes and some photocopies. Thus (not realizing the original applications of the 595 would be destroyed later), Ms. Gaspar's notes were incomplete but only in respect of data she considered irrelevant. Moreover, she had to exercise some interpretive judgment in forming an opinion as to some applicants' colour by considering the other data given. (Transcript, Vol. VI, pages 11, 60–71). Respondents do not assert that Ms. Gaspar deliberately skewed the data, but argue that her approach was incomplete (in a literal sense – she did not simply make photocopies of all the applications) and, therefore, introduced an unintentional skewing and bias.

95Respondents asserted that Ms. Gaspar's sample was not made in a random way so as to exclude human judgment. Ms. Gaspar's collection of data was, indeed, not "random" in the sense of being willy nilly, but was collected objectively and impartially, with her judgment in collecting and interpreting data done with a view to completeness. Where interpretive judgment was exercised, it was in respect of simple data that anyone with common sense could interpret. Expertise was not required. Thus, Ms. Gaspar would interpret an applicant's colour or racial origin by considering such data as name and place of education. (Transcript, Vol. VI, pages 27–33). Any statistics expert could not have done this any better than Ms. Gaspar: indeed, her expertise as a human rights officer probably was the best experience one could have for the task at hand. Certainly, there was room for minor error through mis-interpretation but such errors would be honestly made, and given the large number of the sample, should not have skewed the statistical results. In my opinion, the data was properly gathered and compiled in the first instance by Ms. Gaspar.

f. STATISTICAL METHODS AND JUDICIAL COMPETENCE

96There are many methods of statistical analysis (Camp, supra, at 442–43; Reynolds, supra, at 963), some more appropriate in discrimination cases than others, but none which are flawless. The best method in cases of discrimination in hiring is the applicant flow analysis (Phillips, supra, at 1025; Camp, supra, at 442; Carter v. Newsday, Inc., 528 F. Supp. 1186 (1981) at 1193; Offierski, supra, at D/38; Kilgo v. Bowman Transportation, Inc., 570 F. Supp. 1509 (1983) at 1526). This is the method used by the Commission in the case at hand. This involves proof of a disparity between the percentage of, for example, women among those applying for a position and the percentage of women among those hired for the position. This approach is not always appropriate. One problem with this method can arise when an employer's discriminatory practices are so well known that no women bother even applying for the position (Teamsters, supra, at 1870).

97Another method is the static analysis method (Camp, supra, at 443). This compares the percentage of women among those hired for a position with the percentage of women among qualified persons within the general population from which the employer would be expected to draw its employees. Choice of the appropriate labour pool (De Medina v. Reinhardt, 30 E.P.D. 26,866 (1982 at 26,869; Detroit Police, supra, at 688; Trout, supra, at 20,291; Reynolds, supra, at 967–69; Vuyanich, supra, at 308, 357 and 375; Coe, supra, at 452; Rivera, supra, at 23,780–83; Dorsaneo, "Statistical Evidence in Employment Discrimination Litigation: Selection of the Available Population, Problems and Proposals," 29 S.W.L.J 859 (1975) at 866–71; Hazelwood, supra; Rich, supra, at 610; Camp, supra, at 443–444; Paxton v. Union National Bank, 688 F.2d 552 (1982) at 564; Moore v. Hughes Helicopters, Inc., 708 F.2d 475 (1983) at 482) and choice of the relevant variables is difficult. Poor choices may easily render the statistical analysis inaccurate or unreliable (Camp, supra). Complex multiple regression analyses, which provide the ability to determine how much influence several factors such as sex, education and experience have on a variable such as promotion or salary, are typically applied in these cases (Vuyanich, supra; Trout, supra, at 20,290; Sobel v. Yeslriva University, 566 F. Supp. 1166 (1983), at 1173.

98In some cases, the statistical evidence may be so complex that there may be some concern as to the competence of the trier of fact, a layperson in statistical matters. Chief Judge Winner of the U. S. District Court (Colorado) said:

(A)ll a trial judge whose statistics course dates back 45 years can do is try to use his limited knowledge of this quasi-mathematical approach to a problem and then temper the argued-for results with a pinch of common sense. That's why I wish the appellate courts would talk about "percentages" instead of "statistics," because most trial judges learned percentages in the sixth grade (Otero, supra, at 335).

There is a significant risk of error when a trier of fact attempts to apply his or her own multiple regression analyses to raw data when he is not satisfied with the statistician's analyses. A little knowledge can be a dangerous thing. However, a trier of fact need not accept a statistician's analysis as presented, but may challenge it and give it what weight is deemed appropriate. Where a simple statistical method is properly explained, there is no harm in the trier of fact performing a few simple calculations to fill in gaps left by the statistician (Vuyanich, supra, at 258–59). Primarily, each trier of fact must keep in mind his or her own limitations when dealing with statistical evidence.

g. STATISTICAL DISPARITIES

99The telling result of statistical analysis in a discrimination case is the size of the disparity between, for example, the percentage of women among qualified applicants for a position and the percentage of women among those hired. The disparity when a neutral employer randomly selects employees from among the applicants is generally less than 2 to 3 standard deviations. A greater disparity gives rise to an inference of discriminatory intent and, if sufficiently large, shifts the burden of proof to the employer to show that discriminatory intent is not the cause of the disparity, or that the statistics inaccurately show the disparity to be greater than it in fact is.

100There is no hard and fast rule that a disparity of greater than 2 percentage points proves discriminatory intent while a disparity of less than 2 proves random selection. A disparity of greater than 2 may be caused by factors other than sexual discrimination. On the other hand, a disparity of less than 2 may be some evidence of discriminatory intent where there is other evidence of discrimination. The importance of the size of the disparity depends on the cogency and persuasiveness of other evidence in proving or disproving discrimination. The less there is of other evidence of discrimination the greater the statistical disparity must be in order to make out a prima facie case of intent to discriminate. Circuit Judge Wallace of the U.S. Court of Appeals (Ninth Circuit) said:

It would be improper to post a quantatitive threshold above which statistical evidence of disparate racial impact is sufficient as a matter of law to infer discriminatory intent, and below which it is insufficient as a matter of law. Indeed, if any rule of law can be isolated from the Supreme Court's decision in Castaneda v. Partiela ... and Hazelwood ... it is that for sufficiently large samples, a variance between expected and observed results of greater than 2 or 3 standard deviations is sufficient to make the data "suspect" for a social scientist. However, the statistical differences which the Court in these cases held were sufficient to support an inference of intentional discrimination, respectively 12 and 5 to 6 standard deviations, are far greater than those which would be "suspect" to the social scientist ... (Gay, supra, at 551; See also Sobel, supra, at 1173–74; Federal Reserve Bank of Richmond, supra, at 647–48; American National Bank, supra, at 21,091–93; Dorsaneo, supra, at 871–73).

It appears from reading the American cases that for statistical evidence standing alone to create a prima facie case of intent to discriminate, a deviation of greater than 5 points is considered an appropriate standard.¹

APPLICATION OF THE LAW TO THE EVIDENCE

101The law has been reviewed as to the use of statistical evidence in discrimination cases.

102Ms. Gaspar transmitted her collected data to commission counsel, not directly to a statistician, and Commission counsel, together with his staff, correlated the data and coded them on computer cards. (Transcript, Vol. V, pages 101–103; Vol. VI, pages 9–13; Vol. VI, page 24). The statistician, Mr. Duncan Alexander McKie, received only the computer cards, already coded with information, but did formulate the framework and categories into which the data was to be placed by Commission counsel and his staff. Respondents argue that there is a fatal gap in the evidence, in that those who transcribed the data collected by Ms. Gaspar made judgments (for example, as to racial grouping for applicants) in coding the data, and they were not called as witnesses concerning the factual basis of the data as used by them. (Transcript, Vol. V, pages 102–107). Persons, in particular Commission counsel, made interpretative judgments and decisions about the data in putting the data into a form for statistical compilation, yet none of those persons were available as witnesses. Ms. Gaspar, who was a witness, was unable to say whether the data she had compiled was used by the statistician, Mr. McKie, in the state she had prepared it, and decisions as to putting data into specific categories was made by Commission counsel, before giving the coded data to the statistician. (Transcript, Vol. VI, pages 12–14, 17). (Transcript, Vol. VI, pages 12–24; Vol. VIII, P. 6). Respondent's counsel argued that this approach did result in some minor discrepancies between, Ms. Gaspar's notes and Commission counsel's categorization of data. These asserted discrepancies were very minor, and of no real significance.

103These discrepancies are relatively minor, and some favour the Respondent's position. There was no intent on the part of the Commission counsel and his staff of 5 or 6 to skew the data in transcribing it to computer cards, nor was this allegation asserted by the Respondents. However, it is essential that caution be exercised in the use of data as statistical evidence, and when data is evaluated with interpretive judgments expressed before being utilized in a statistical way, the person or persons doing such should certainly be available as witnesses. Every step in physically transmitting the data should be addressable by those involved, as witnesses. Of even greater importance, every step in evaluating the data through making interpretive judgments and through categorizing the data, should be addressed by those involved, as witnesses. Data can be misunderstood and misconstrued, but cannot, of course, be cross-examined. The reliability of the statistical evidence at the end of the process depends upon the sanctity of the underpinning, original raw data being preserved at every step and stage thereof.

104In the instant case, the only area in which there was any significant question of categorization, using the raw data, was whether someone should be put in the categories of "white or non-white" and in respect of identifying of "national origin" (Transcript, Vol. VI, pages 17–21).

105Mr. Duncan Alexander McKie is the Research Director of Environics Research Group, a social survey and marketing research firm in Toronto, has a Masters Degree in Sociology, has completed his course work for a Ph.D. in Sociology, and in his work conducted many studies from the standpoint of doing the statistical analysis, as set forth in his curriculum vitae. (Exhibit #36). His credentials as an expert in statistical analysis were not challenged.

106There were 595 applications by named individuals for which "fundamental information on the nature of the application" was available (Transcript, Vol. V, page 54). Any data missing in respect of the 595 was not systematically excluded, but rather, occurred in a random way. Mr. McKie's expert opinion was that the sample was easily sufficient in number, and the data sufficient in detail, to provide statistically reliable evidence (Transcript, Vol. V, pages 54–58).

107Mr. McKie's first analysis done was a cross-tabulation of a number of demographic factors (Exhibit #38) with "national origins" failing into two groupings – "North Europe" and "non-Europe" (Transcript, Vol. V, pages 59–65). With respect to national origin, Mr. McKie found that the statistics showed that "the probability of getting an interview significantly declines if one is a member of the second group" (Transcript, Vol. V, pages 74, 91). Mr. McKie's analysis included a "multiple regression" analysis to ensure that the two groups were being measured simply on a basis of "national origin" where other factors, such as education, were constant. (Transcript, Vol. V, page 75).

108Mr. McKie found that the statistical evidence suggested "a certain amount of non-random distribution ... whether or not one gets interviewed, on the basis of their age." (Transcript, Vol. V, page 67). 34 percent of persons under 40 were interviewed, whereas only 23 percent of those over 40 were interviewed. Mr. McKie said that the 11 percent difference was "minimal," if suspicious, in the circumstances. (Transcript, Vol. V, pages 67, 79).

109There was a "definite, strong significant relationship between sex, and whether one was interviewed or not." (Transcript, Vol. V, page 68). 31.6 percent of males were interviewed, whereas only 18.5 percent of female applicants were interviewed in this period. (Transcript, Vol. V, pages 68, 69, 80).

110Mr. McKie applied the "chi square" test to the statistics which

tests it to see whether there is a signficant difference between a random distribution of those numbers, and the distribution that obtains in the table, and what it finds is that there is a certain probability of something occurring by chance, and, in this case, the probability of this occurring by chance is less than 1/100th of a percent. (Transcript, Vol. V, pages 68, 69).

111Put simply, his conclusion was that the statistical evidence shows "conclusively" that there is a significant bias in favour of male applicants being interviewed simply because they are males. (Transcript, Vol. V, pages 71–73).

112A cross-tabulation of "Interviewed or not, by application date, by sex" was prepared by Mr. McKie, and it shows that after October 1, 1977, both males and females were interviewed "at an increased rate," but there was still "great discrepancy" of about 17 percent, between the percentage of male applicants who were interviewed and the percentage of female applicants who were interviewed. Mr. McKie saw gender as a "significant predictor" of whether one would be interviewed or not for the whole period between May, 1977 and May, 1978 (Exhibit #39; Transcript, Vol. V, pages 82–85).

113The statistical evidence also showed that "height and weight" was a significant factor in being interviewed and in being hired (Exhibits #40, #41 and #42; Transcript, Vol. V, pages 85–91, 3). With his "multiple regression" analysis, Mr. McKie determined that "weight" rather than "sex" was the most significant factor, and that, considering everything, "national origin" and "weight" were the two most significant factors in deciding whether interviews would be given. Once weight is in actuality a factor, there is, of course, a significantly disparate impact upon women in seeking employment, even though there is no formal minimal weight requirement. The Respondent did not lead any evidence to suggest that there was any minimum weight standard (in fact, asserted there was not) and accordingly, did not argue that weight was a bona fide occupational qualification.

114There was a great deal of data submitted in respect of the statistical evidence, comprising some nine volumes (Exhibit #16). The data was complex and to some extent introduced into evidence in a confusing manner. Given the ultimate finding in respect of the Complainant's Complaint, it is not necessary to make findings in respect of the statistical evidence, however, it is appropriate to make one finding in this regard.

115The law has been reviewed at length as to the use of statistical evidence in discrimination cases. In my view, the statistical evidence presented in this Inquiry was most significant on the issue of whether a prima facie case of discrimination because of sex under the Code could be made by statistics alone. The prohibited ground of discrimination because of sex was the main ground at issue in so far as the statistical evidence was concerned and in respect of the Complainant's Complaint.

116There was considerable statistical evidence presented on this particular issue, with a good deal of commentary about such evidence through witnesses and through oral and written arguments. Put simply and succinctly, it seems to me the salient statistics on this issue can be set forth as follows.

117Of the group of 595 applicants (Exhibit #26) for which data was available, 172 females (28.9% of all applicants) applied and 423 males (71.09% of all applicants) applied. Of these applicants, 32 females were interviewed (19.51% of total interviewees of 164), and 132 males were interviewed (80.49% of total interviewees). Although females constituted 28.9% of applicants, they constituted only 19.51% of interviewees. The statistics would suggest, at first appearance, that females had a significantly less chance than male applicants of being interviewed simply because they were females. However, the reduction in the percentage of women at the interview stage is coupled with an even more dramatic reduction in the percentage of women at the hiring stage. Only two females (3.92% of the total of 51 hired) were hired, whereas 49 males (96.08% of the 51 hired) were hired. Overall, while females constituted 28.9% of applicants, they comprised only 3.92% of the persons hired. Put another way, 1.1% of female applicants were hired as opposed to 11.58% of male applicants. In my opinion, a prima facie case of discrimination because of sex under the Code was made out by statistics alone in the instant Inquiry. The Respondent's system of recruitment of correctional officers, based upon the statistical evidence, discriminated against females generally, at least for the period of time, 1977 to 1978, examined by the statistical evidence presented in this Inquiry. In my opinion, discrimination because of sex was endemic to the recruitment system at Mimico until Mr. De Grandis took over the position of Superintendent. He did not in any way follow discriminatory practices at Mimico himself, in recruitment or otherwise, and he introduced changes for the better in the recruitment system although this, understandably, took time. Finally, the central recruitment process introduced by the Ministry of Correctional Services since November, 1981, is undoubtedly more professional and objective, and while there was not statistical evidence for recruitment in recent years, the oral evidence suggested that there was considerable improvement.

118Having made the finding that the statistical evidence in itself presented a prima facie case of discrimination because of sex, it is also clear from all the evidence that the Complainant was not personally discriminated against because she was a woman or on any other prohibited ground. The Respondent has successfully rebutted the prima facie case of discrimination because of sex made out by the Commission's statistical evidence, by the oral testimony of the witnesses, in particular, that of Superintendent De Grandis. I have no doubt in finding that he is a truthful witness and that he interviewed the Complainant fairly and objectively, making a decision in his best judgment, and based upon considerable experience, that she was not suitable for the employment position sought, being the position of correctional officer. For these reasons, the Complaint is dismissed.

Footnotes

1. 1One type of situation where a lesser statistical disparity might carry more weight is where the employer places too much emphasis on subjective evaluation of applicants and allows the decision-makers a lot of unguided discretion in hiring decisions. Hiring procedures which rely heavily on the subjective judgments of white male interviewers can easily be used to mask sexually and racially motivated hiring decisions (Phillips, supra, at 1026; Payne, supra, at 24,879; Brown, supra, at 1382–83; Senter, supra, at 529). Some jobs are such that a certain amount of subjective evaluation is necessary. However, an over-emphasis on subjective evaluation and under-emphasis on objective qualifications is suspect particularly where statistics indicate that the discretion is not being exercised in a sexually or racially neutral manner. In such cases, once the plaintiff has shown discrimination by the use of statistics, the employer may generally only rebut this proof by the use of its own statistical evidence (Coe, supra, at 453, fn. 2). For, where subjective decisions are at issue, there is seldom other objective evidence of non-discriminatory use of the discretion. There is also a large subjective element inherent in relatively infrequent decisions to hire highly qualified persons for special positions such as president of a university or chief executive officer of a large corporation. In these situations there are too few people who are qualified and too few hiring decisions made for any useful statistics to be garnered. However, in cases such as the one before this Board of Inquiry the numbers of qualified persons and hiring decisions are sufficient for meaningful statistical analysis of the exercise of discretion (Lamber, Reskin and Dvorkin, "The Relevance of Statistics to Prove Discrimination: A Typology," 34 Hastings L.J. 553 (1983) at 582–84).