The use of graphology as a tool for employee hiring and evaluation

Posted on

Part I

The BCCLA holds that an employee has a right to privacy in at least the respect of not being coerced to reveal information about him—or herself that is not directly relevant to the carrying out of his or her job. However, we hold that this right to privacy is not indefeasible (BCCLA, 1987). It can be outweighed by three broad types of considerations, having to do with (a) the responsibility of the employer to employees other than the one(s) being tested, (b) the employer’s responsibility to the public at large and (c) the employer’s right to select the most competent person for the job in order to maximize his or her own profits.

With respect to (a) and (b), if we maintain that the employer has the right, or even the duty in some cases, to monitor an employee’s performance in these matters, then we could have no objection to the employer getting relevant evidence about how well an employee is likely to perform before an employee begins the job. In fact, if it were possible to get this evidence, we might think that, for example, the employer has a duty to determine in advance that an employee is likely to be prejudiced against minority groups before hiring that person as a personnel manager. With respect to (c), an employer who wishes to maximize his profits from his employees has the right to determine whether an employee possesses the requisite skills and capacities to be effective in the job. Thus, vocational preference tests, aptitude tests, and, perhaps, some personality profiles may be relevant to determining the answers to these questions. Again, if the employer has the right to monitor the effectiveness of an employee currently holding a job, we cannot see a principled reason to object to the employer seeking to gain information about this prior to hiring that person for the job. And, if the employer has the right to monitor the actual work performance of an employee, he or she should have the right to determine the employee’s skills by means of tests before he or she is hired. So, we do not have a blanket opposition to appropriate tests being used on current and future employees.

However, an employee ought not to be expected to give up his or her right to privacy when the test in question really does not provide the relevant information. Nor should a job applicant be tested by means of a procedure that is arbitrary and which does not in fact determine the abilities it claims to determine. It is obviously unfair that a job applicant is passed over because a test result indicates that he lacks characteristic c, when, in fact, he possesses c, but the test does no better than chance in determining whether he possesses it or not. It would be equally unfair if a job applicant were passed over because a test result indicated that another candidate, A, possessed c, when, in fact, A lacked that capacity. For example, if applicant A can type 70 words per minutes and B only 60, it would be unfair to A if B got the job because the typing test indicated that A can type 45 words per minute while B can type 65.

Personnel managers or others responsible for hiring, are generally very much concerned to prevent this sort of unfairness in their personnel decisions. When the skills in question are straightforward, as in the typing example just considered, they can be relied upon to use reliable testing procedures. But hiring and promotion decisions usually involve characteristics which are different from typing speed in two important respects: 1) they are much more important for the future of the institution and 2) they are much more difficult to discern. We have in mind such traits as reliability, honesty, motivation, ability to get along with fellow workers or customers and so on. Because of 1), personnel managers are especially concerned to find out in advance as much as possible about the applicant’s possession of these traits, since mistakes about these traits can be very very costly.(1) And because of 2), they rightly worry that they may be unfair about judging a job applicant on these traits by means of interviews, letters of recommendation, etc., or other subjective criteria.

Orthodox psychologists readily admit that they do not have any `instant answers’ for determining these traits. There are no paper-and-pencil tests that provide even remotely reliable evidence about them. Thus, the conscientious personnel manager really does have a serious worry about his or her objectivity and fairness in making judgments on these matters. It is ironic that some of the personnel managers for municipalities and corporations in B.C., who are most concerned about objectivity and fairness, have fallen prey to a personnel selection tool which, in fact, is most likely to result in unfair personnel decisions. Graphologists (handwriting analysts) who generally lack the modesty of psychologists and the healthy skepticism of those who use the scientific method, promise foolproof evaluations of employees’ intellectual capacities, honesty, reliability, compatibility, sociability, commitment, motivation, not to mention sexual practices and orientation and even physical and mental health! They offer to ensure that fair employment decisions will be made and that institutions will get the best employees. It is no wonder that many personnel managers for municipalities and corporations in B.C. have hired graphologists to assist in these decisions, despite the fact that graphologists routinely fail fair and impartial tests of their exaggerated claims to discern traits that are opaque to orthodox assessment techniques.

This last charge is a serious one. If it is justified, an obvious conclusion suggests itself: that those corporations and municipalities presently using graphology on prospective and current employees should cease this practice as soon as possible, in order to avoid committing an injustice. This will be our recommendation in Section V, but our first task is to defend our charge that graphology simply cannot determine the personality traits that its proponents claim. That is the burden of the next Section II. Section III will offer an explanation for why many otherwise astute personnel managers believe in its efficacy, despite the overwhelming evidence to the contrary. Section IV will raise other concerns about privacy and other matters that we have uncovered in our examination of graphologists practicing in the Lower Mainland, which enhance our case that this practice should be halted.

Part II

A brief history of graphology makes it apparent that the discipline was not founded by people adept at using the scientific method to determine correlations between aspects of handwriting and personality traits. The `Trait School’, originating in France in the mid-1800s, claimed to identify personality traits with individual attributes of handwriting (e.g., slants, widths of strokes, etc.) (Klimoski & Rafaeli [1983], p. 192). The practitioners of this school offered no rationale for their claimed correspondences; they simply asserted them. The same reliance on the ancient principle of `sympathetic magic’ (`like’ begets `like’) is apparent in the rationale for handwriting analysis as in other alleged character reading systems such as astrology (“Mars is connected with violence, war, and so on, because, like blood, it appears red”). Likewise, in graphology, people whose lines slant upwards are `upbeat’ people. The `Gestalt School’(2) arose in Germany in 1910 as a response to the earlier school (Klimoski & Rafaeli [1983]). One would expect that the opposition would be founded on an appreciation of new statistical methods for establishing correlations, of the sort being introduced by Galton and Pearson in England, but such was not the case. Instead, the Gestalt people defended something even more vague, holding that graphological interpretations should be based on the handwriting as a whole, allowing room for intuition. Why not have the best of both worlds?, one might ask; and Bunker, in 1929, founded a school called Graphoanalysis to do just that.(3) Again, this movement merely widened the opportunities for handwriting analysts to say whatever sounded plausible. There was no attempt by this system to use the better statistical techniques of, e.g., Spearman and Thurstone, whose correlational and factor analytic techniques were used by this time by all orthodox scientists in the field of differential psychology (the study of individual variations in aptitudes and personality).

It would be unfair to browbeat modern graphologists with the lack of rigour displayed by their forebears if the moderns had learned from their mistakes and grown, in the way that present day science has outgrown embarrassing moments in its past. But this simply has not happened with graphology. In fact, one of the surest symptoms of a pseudo-science is this lack of progress over a long period of time.(4) Witness one of graphology’s modern proponents, Jane Patterson (1976), appealing to the same principles of sympathetic magic as her forebears: “People whose writing slants to the right come forward to meet other people” (People with a backhand are reticent) or “People who stick to the copybook dimensions have a good sense of proportion.” These kinds of puns are not even funny, let alone theoretical statements referring to scientifically established correlations.

No modern graphologist has attempted to determine correlations between features of handwriting and personality traits in the way modern psychology would proceed.(5) Roughly, such a procedure would look like this: a sufficiently large and representative sample of subjects to allow valid statistical comparisons would be drawn. Graphologists would identify certain features of handwriting (such as positions of the dot at the `i’, or whatever), and form their hypotheses about which personality traits are correlated with them. Subjects would be given personality tests with reasonable validity and reliability coefficients, administered and scored blindly, and correlations with graphologists’ predictions computed. In addition, longitudinal studies, again double-blind, would be conducted, to see if these traits could be discerned in the subjects by independent observers using objective criteria. Needless to say, such studies have not be done by graphologists. To their credit, two local graphologists, Aron Printz and Robyn Smith (vice president of the American Handwriting Analysis Foundation) have admitted to these writers that graphology is not a science, but an art. We hope that they are equally candid with their prospective clients. To be fair, John Russell and Dale Beyerstein heard Mr Printz make this admission in front of Chilliwack Council after that municipality had spent $11,700 for his services.

But even more interesting than the question whether graphology is a science is the question whether it works. One would hope that the graphologists themselves would have enquired into this. But Michael Moore (Moore, 1985) contacted eight of the major organizations engaged in graphology worldwide (in the US, Israel and Italy), requesting “controlled studies in which the results of handwriting analysis are shown to predict employee characteristics”. Four associations did not bother to respond. Of the four that did, none could produce such a study, but one forwarded a non-technical article from Playboy magazine. Nonetheless, respondents had a lot of other things to talk about. The Israeli Graphological Institute referred Moore to Allport and Vernon’s Studies in Expressive Movement (1933), either not understanding that this work provides a negative evaluation of graphology, or hoping that Moore wouldn’t check the reference.

The one study (originally) sponsored by graphologists that meets the canons of science (Rafaeli & Klimosky, 1983), not only showed no evidence for graphology’s validity, but also demonstrated the lengths to which graphologists are prepared to go to suppress negative evidence. The American Association of Handwriting Analysts(6) originally agreed to the protocols of the study; but when they saw that the results would be embarrassing to them, threatened a lawsuit if the authors published the results. This, despite one bit of evidence in favour of graphology: viz., that graphologists, trained in the same school showed some semblance of inter-rater reliability (i.e., that they made similar judgments from the handwriting samples). Even this basic requisite has not been found in studies of graphologists such as those of Louis Goldberg (1987) and Abraham Jansen (1973).

But at least they did better than the American Handwriting Analysis Foundation. This group was recently asked to participate in a study designed by Nancy Higgins and Alan Davison of the B.C. Skeptics to test local graphologists (B.C. Skeptics, 1988). Their vice president, Robyn Smith, refused, on behalf of the Foundation, to participate from the outset, telling one of the authors that “it would not be fair to test it [graphology] against chance” (D. Beyerstein, 1988, p. 2). Perhaps if she knew something about the scientific method, she would not have refused (or at least not on those terms). It was the B.C. Skeptics’ intention to get the participation of Aron Printz; but he also categorically refused (ibid.). His reasons had to do with his admission that graphology is not scientific. But this is not to the point: the test was designed to show whether graphology worked, not whether it is scientific.

The studies listed in the Bibliography were designed to test whether graphologists can actually predict personality traits from handwriting samples.(7) We must remember that graphologists claim to predict these traits from the style of handwriting alone, and not from the information contained in the handwriting sample. We must remember also that a job application letter used by the graphologist invariably includes a gold mine of information about the writer. It is almost certainly the case, therefore, that the information contained in such letter can be, and is, used by the graphologist in constructing a personality profile of the applicant. Consider how well someone who can read can do at posing as a graphologist:

Candidate Writes ‘Graphologist’ Predicts
“I was born in London on
May 9, 1937… “
“… shows concern about his retirement; …more concerned than most people about loud noises… “
“I received my Bsc (Hons) from UBC in 1979, with a major in Physics…” “… shows by the formation of her ‘t’ and ‘f’ that she has a good head for figures, and is suitable for scientific pursuits….”
“Since I graduated from high school in 1966 I have been in the Warehouse… rising last year to Floor Manager…” “… shows by the careful crossing of his ‘t’s that he is a stable personality; would probably show a great deal of loyalty to the firm….”

But note also that the obvious information is not all there is to be mined from such letters. Such cues as spelling ‘centre’ ‘center’, spelling mistakes, grammatical errors, vocabulary, etc., reveal a great deal about a person, but officially this is not the information graphologists are supposed to be using. Jansen (1973) reports that Dutch psychologists, working from typewritten transcripts of the handwriting samples used by graphologists, made predictions that had the same validity as those made by graphologists. An Israeli psychologist, serving as a control to graphologists, outperformed (though not significantly) the graphologists in predicting supervisors’ rankings of bank employees who had submitted a brief biographical sketch when they applied for their jobs (Ben-Shakhar, et al [1986], p. 647). Many other studies report the same finding: when there is content information in the handwriting samples, non-graphologists do as well or better than the graphologists in predicting traits.

On the other hand, in properly controlled, blind studies, where the handwriting samples contain no content that could provide non-graphological information upon which to base a prediction (e.g., a piece copied from a magazine), graphologists do no better than chance at predicting the personality traits (Rafaeli & Klimosky [1983], Ben-Shakhar [1986], Karnes [1988], Jansen [1973]). There is also ample evidence (Karnes, 1988) that a randomly-chosen graphologist’s report will be accepted overwhelmingly as an excellent description of themselves by a large group who think it was done individually for them.

The only evidence that graphologists really can discern complex personality traits from handwriting comes from their own reports, given in their sales pitches, or in privately published books or articles. As we mentioned in Note 4, this is typical of the “evidence” for a pseudoscience, and not the evidence that a conscientious personnel manager would want to have before using an assessment device to make judgments concerning a person’s life prospects. On the other hand, one of the authors of this brief has canvassed the scientific literature on graphology, and has not found even one reputable text in psychological measurement that even mentions graphology, except to denounce it. Most do not even consider it worthy of mention. The criticisms of the serious methodological flaws and empirical falsehoods of graphology come from those scientists who take time out from their own research to debunk pseudoscience.(8) The B.C. Skeptics has repeatedly challenged graphologers, publicly and in private, to answer these criticisms. They have received no scientifically acceptable answers from the graphological community whatsoever.

There is one last line of defense offered for graphology that needs rebutting: it might be maintained that all these studies showing graphologists to fail to live up to their promises are done by psychologists their competitors. Might they not have a vested interest, making them go out of their way to discredit the opposition? This explanation may appeal to people who like conspiracy theories or who hate psychologists. But for those who hate psychologists, there is an argument that makes them look even worse: If their studies showed the graphologists to be on to something, the psychologists would have no need to hide this fact. Even better than eliminating the opposition, from their point of view, would be to steal their hard-won insights, and use them themselves. After all, this is what has happened to hypnotism. Long considered to be on the fringe of psychological practice(9), some psychologists now use it, and have no need to keep its supposed uses a secret. So, this last ditch attempt to save graphology is like water: it boils down to nothing.

Part III

If we are right in maintaining that graphology is worthless as a predictor of these personality traits, why is it that so many personnel managers swear by it? Part of the explanation is that they are not informed of the scientific literature demonstrating the inefficacy of graphology; and are persuaded by testimonials of their colleagues, who are equally uninformed. Another reason, alluded to in Section I, is that they are desperately searching for some tool that will provide them with evidence about employees’ reliability, honesty, commitment, and other factors which orthodox psychologists rightly do not claim high success rates in discerning. But the most important reason has to do with a phenomenon variously called by psychologists personal validation (Forer [1949]), subjective validation (Marks and Kammann [1980], pp. 174-199) and the Barnum Effect (Snyder et al [1977]).(10)

In order to see how subjective validation works, imagine the following scenario: Someone whose opinion and credentials you trust gives you a paper-and-pencil psychological test. After you have done the test, this person provides you with the following personality profile, which she informs you was generated from the test you took:

You have a need for other people to like and admire you, and yet you tend to be critical of yourself. While you have some personality weaknesses you are generally able to compensate for them. You have considerable unused capacity that you have not turned to your advantage. Disciplined and self-controlled on the outside, you tend to be worrisome and insecure on the inside. At times you have serious doubts as to whether you have made the right decision or done the right thing. You prefer a certain amount of change and variety and become dissatisfied when hemmed in by restrictions and limitations. You also pride yourself as an independent thinker; and do not accept others’ statements without satisfactory proof. But you have found it unwise to be too frank in revealing yourself to others. At times you are extroverted, affable, and sociable, while at other times you are introverted, wary, and reserved. Some of your aspirations tend to be rather unrealistic.

You are asked to rate how well this profile describes your own personality, on the following 5-point scale:

5. Excellent

4. Good

3. Fair

2. Poor

1. Wrong

If you are like hundreds of thousands of university students over the past 40 years who have participated in the experiment described below, you will rate this profile highly—the average rating when this experiment has been done hovers around 4.2 on this scale. Forer (1949) first did this experiment on a psychology class in 1948, getting an average ranking of 4.26. The present authors have both done this experiment with their classes, systematically getting similar results. The key element, however, is that once students have given their ratings, it is revealed to them that they all have received the same profile! The “tests” that they took supposedly to generate this profile were simply thrown in the wastebasket. The purpose of this part of the exercise is merely to put on a convincing show that some reliable measurement technique is being used.

These experimental results demonstrate the power of subjective validation: how any personality sketch, so long as it is sufficiently vague that the reader can read meaning into it, provided he has a desire to do so, will receive a favourable response from the majority of subjects. It does not even matter what deception is practiced on them whether they are told that the personality profile is generated from a psychological test, a sample of their handwriting, their astrological sun sign, their biorhythm the average ranking remains unchanged. Forer,in fact, originally performed this experiment to demonstrate his claim that a Los Angeles nightclub graphologist was peddling statements that were true of everybody but cast in a formed that seemed to be unique to the individual. But this notion equally explains how people can come to believe that other pseudosciences provide accurate information about them.

Forer (1949) interprets the tendency to be fooled this way as a sign of gullibility; but it is in fact a sign of intelligence to be able to read meanings into vague sentences it is a skill we use constantly in interpreting verbal communication. But humans are so good at this that they sometimes read meanings into passages where there isn’t any that was originally intended by the author.

Also, as Kammann and Marks (1981, pp. 189-90) note, the above passage consists of near opposite remarks (“While you have personality weaknesses you are generally able to compensate for them [i.e., you have strengths]”). If one remark does not apply, then the other will; and most of us are complex enough creatures that both will apply, if we think of different times in our lives, or how we behave in a variety of circumstances.

Another secret in supplying a reading that will be accepted by the subject is sagely noted by Hyman (1981b, p. 79): tell him what he would like to be true about himself. However, don’t be too obvious about it: the successful deceiver is well advised to use the subtle forms of flattery to be found in the above personality profile.

With this in mind, we are ready to return to the question with which we began this section: why do so many personnel managers find graphologers’ assessments of job applicants so accurate? Part of the answer is that successful graphologists usually begin their sales pitches with a free personality assessment of the personnel officers themselves! The production of a personality profile with the features of Forer’s can be expected to elicit the admiration of a personnel manager for the graphologist’s acumen, and to disarm any skepticism there might initially be about the validity of the method.(11)

But the defender of graphology will respond that the personnel managers’ positive assessments of graphology are based not (only) on graphologers’ assessments of themselves, but (also) on the graphologers’ profiles of the job applicants. Quite so. But we have several reasons to take these testimonials to graphology with a grain of salt. First of all, personnel managers favourable to graphology most often cite as their main reason the fact that they are perfectly satisfied with the employees selected by this method. Now, while we have no reason to doubt that they should be satisfied with them, we must point out that this does not settle the issue. Personnel managers, when questioned about graphology, defensively point out that they use other selection tools besides graphology. And, as mentioned at the beginning of this section, graphology promises information about those traits such as honesty, which can be corroborated only by long term observations of the employee. So, while the personnel manager can evaluate these traits in the person in fact hired, he necessarily has no independent evidence about the people who were not hired. How can he compare the person hired to the ones rejected on the basis of graphology? Obviously he cannot.

Second, there is the self-fulfilling prophecy aspect. If managers think that an employee will do well, as they will do if they have received a glowing report on the employee they hire from the graphologist, then they are likely to excuse mistakes, or subtly help them over the beginning hurdles (after all, this is worth doing for a first-rate employee!). Also, if only those applicants that are on the short list are examined by the graphologist, they have already been screened for the traits relevant to successful job performance. Thus it is likely that anyone on the short list would have done approximately as well in the long run on the job.

Third, it is no surprise that a personnel manager defends his use of graphology when questioned by those more committed to the scientific method, especially when the issue of spending the municipality’s or the corporation’s money is raised. Avoiding ridicule serves as a powerful motivation to avoid cognitive dissonance.(12)

Fourth, we have evidence, presented by Aron Printz (in his appearance before Chilliwack council, mentioned previously), for an hypothesis involving subjective validation to explain graphologers’ abilities to choose employees that will satisfy personnel managers. Mr. Printz mentioned that he always speaks to the personnel manager before analyzing the candidates’ handwriting samples, to ensure that he is looking for the traits that concern the personnel manager. Thus, he has a very good idea of what the personnel manager is looking for in the content of the handwriting samples. As we have previously mentioned, any worthwhile letter from an applicant should contain enough information about the writer to permit shrewd guesses. Printz also asks in advance to know the sex(13) age, handedness, education and occupation of subject. Studies have shown that people, given this information alone, can and do score above chance in attributing many traits. And, as a result of this conversation, any competent “fisher” can discern what the personnel manager is already thinking about the respective job applicants. The personnel manager in most cases will express concerns he has in general about different types of job applicants (“Would someone with a Master’s degree get bored by a job like this and leave soon?”; “Would a man be prepared to work under our female mayor?”), and perhaps even his concerns about a particular candidate (“This guy claims to have left a more prestigious job than the one we are offering. Why did he leave? Would he resent the cut in salary?”). Thus, a graphologist with any smarts probably knows just whom the personnel manager favours before he even looks at the candidates’ handwriting. And even if he does not know exactly, with judiciously vague wording the graphologer can send in a report that will allow the personnel manager to read in what he or she wants to hear. Remember the point made in Section II, that the graphologer knows whatever the personnel manager knows about the job applicant from the covering letter. So, stating something vague in the profile that can be read as the personnel manager sees fit is not that difficult a job.

Part IV

Given that this is so (and we think we have provided enough evidence to shift the burden of proof onto the defender of graphology, where it should have been in the first place), this gives rise to a serious worry about the use of graphology for hiring. We argued in Section I that the major reason that personnel directors give for using graphology is that it would, if valid, provide them with independent evidence about the candidate, which would counter their own subjective biases. But, if we are right, graphology does just the opposite it makes those biases more decisive in the hiring process. Most personnel directors are good at recognizing their biases and discounting them; or recognizing that their reason for not ranking a given candidate very highly is just a hunch, one that they would discount if it were weighed against objective evidence such as university grades, number of years experience in the job, etc. But if that same hunch is now fed back to them as something discerned by the “reliable” method of graphology, they will think their hunch is confirmed, and now be prepared to use it to counter more reliable evidence. So, the very thing they are trying to avoid is now a part of their hiring procedure.

In addition to the unfairness involved in using an unreliable testing procedure, most municipalities and companies compound the problem by violating their job applicants’ right to privacy. With most testing procedures or with most information gathering methods, the job applicant is aware that a test is to be conducted, or gives permission for the information to be gathered. Thus, is the applicant does not wish the employer to have that information, he or she can refuse to apply, or refuse permission for the information to be gathered at the risk of not being considered. But none of the municipalities that until recently were using graphology, and neither Langley nor Chilliwack, which continue to use it, inform(ed) job applicants that they will be subjected to it. We suspect that fear of embarrassment has a lot to do with this, since the standard excuse for not informing applicants, or obtaining their permission for graphology simply will not stand up. If graphology provides information about the subject’s true personality lurking below societal conditioning and the conscious awareness of the subject, as its defenders claim, then applicants could not disguise their handwriting from the discerning graphologer; so there is not reason not to inform applicants.

Thus, we recommend to those who decide to continue this practice that they at least inform their applicants that they will be subjected to it. They should also, of course, be more forthright than they have been about informing their ratepayers that tax dollars are being spent on this scientifically questionable service.

Secondly, we have concerns about the degree of confidentiality that prospective job applicants can expect from at least one Vancouver graphologer, Aron Printz. In a meeting with the Chilliwack Council (fortunately closed to the press), Mr Printz’s argument for why his services should be continued consisted of a series of testimonials from previous customers. We assume that he had their permission to use their names. However, in describing a previous job search that he did for Chilliwack, he mentioned the names of one applicant who got the job and another who did not. We doubt very much that he had received or even asked for their permission to use their names in this context. This would not be a problem if he were speaking just to Council members, since these people are ultimately responsible for personnel decisions. However, two members of the B.C. Civil Liberties Association were also present at this meeting, and it is a breach of confidentiality to mention these people’s records in front of outsiders. Mr. Printz did not even have the concern about confidentiality to ask the BCCLA representatives to keep these remarks in confidence. If municipalities or corporations that continue to use graphology are unconcerned about the other issues we have raised, we hope that they will at least try to ensure that the names of their applicants are kept confidential by their graphologists in the future.

Part V

In this brief we have raised four arguments against the use of graphology: that it has simply not been demonstrated to work; that B.C. graphologers refuse to demonstrate that it works under fair conditions (Section II); that the use of graphology is likely to make personnel directors’ subjective assessments assume more weight than they otherwise would, and that employees are presently subjected to a testing procedure without their knowledge or permission; and our worries about the understanding of the duty to respect confidentiality of at least one graphologer (Section IV). We think that these reasons are more than sufficient for any municipality or company presently using graphology to abandon the practice.

Municipalities have a special obligation to use fair hiring practices. First of all, they fall under the Charter of Rights and Freedoms, which imposes constraints on municipalities to observe fair hiring practices and to avoid arbitrary ones; as well as requiring them to try to preserve confidentiality and the privacy of those who apply for jobs with them. Secondly, a more important consideration is that municipalities represent their citizens; who want to see fair and non-discriminatory policies carried out by their representatives. Thirdly, municipalities’ hiring policies are in the public eye more than those of private companies; and the former ought to serve as a model worthy of emulation by the latter. For these reasons, five municipalities, Richmond, White Rock, Port Moody, Abbotsford and Matsqui have abandoned their use of graphology in recent months. Only Langley and Chilliwack have resolved to continue their use of this arbitrary and worthless screening tool, despite having heard the case presented above by the BCCLA. We hope that further reflection on their part will result in a change in their policies. If not, the chances are good that they will severely damage the career prospects of someone; and that person may have legal redress.

Private firms do not fall under the ambit of the Charter; but ought to consider the moral arguments against the use of graphology nonetheless. The use of this tool has been increasing in the private sector in the past few years, we think largely because personnel managers in these firms have been exposed only to the sales pitches of the graphologers, and have not heard the contrary view. Much media attention has been drawn to this practice over the last few months, and therefore we hope to see a decline in the use of graphology. The BCCLA, in cooperation with the B.C. Skeptics, will endeavour to provide the arguments contained in this brief to personnel directors of these firms.

References and bibliography

ALLPORT, Gordon & P.E. VERNON (1933): Studies in Expressive Movement, New York, Macmillan.

BEYERSTEIN, B. (1987): “Graphology: Write or Wrong?”, the Rational Enquirer (newsletter of the BC Skeptics), Vol. 1, No. 1 (June), p. 1.

BEYERSTEIN, D. (1988) “B.C. Skeptics Cooperate with BCCLA to End Graphology”, the Rational Enquirer (newsletter of the BC Skeptics), Vol. 1, No. 3 (March), pp. 1-2.

BCCLA (1987): Psychological Testing and Privacy. Available from the Association.

B.C. SKEPTICS (1988): “Graphology Revisited”, the Rational Enquirer (newsletter of the BC Skeptics), Vol. 1, No. 3, (January), p. 6.

BEN-SHAKHAR, Gershon, M. BAR-HILLEL, Y. BILU, E. BEN-ABBA & A. FLUG (1986): “Can Graphology Predict Occupational Success? Two Empirical Studies and Some Methodological Ruminations”, Journal of Applied Psychology, Vol. 71, pp. 645-53.

CROWN, D. (1987): “A Review of Scientific Aspects of Graphology: A Handbook”, Journal of Forensic Sciences, Vol II, No. 1, pp. 287-8.

CRUMBAUGH, J.C. & E. STOCKHOLM (1977): “Validation of Graphoanalysis by global or holistic method”, Perceptual and Motor Skills, Vol. 44, pp. 403-10.

FESTINGER, L. REIKEN, H.W. & SCHACTER, S. (1956): When Prophesy Fails, Minneapolis, University of Minnesota Press.

FORER, B.R. (1949): “The Fallacy of Personal Validation: A Classroom Demonstration of Gullibility”, Journal of Abnormal Psychology, Vol. 44, pp. 118-23.

GOLDBERG, L. (1987): in Nevo, op. cit.

HYMAN, Ray (1981a): “The Psychic Reading”, Annals of the New York Academy of Sciences, Vol. 364, pp. 169-81.

HYMAN, Ray (1981b): “Cold Reading: How to Convince Strangers that You Know All About Them”, in Kendrick Frazier (ed): Paranormal Borderlands of Science, Buffalo, NY, Prometheus Books.

JANSEN, A. (1973): Validation of Graphological Judgments: An Experimental Study, The Hague, Mouten.

KARNES, Edward (1986): “Graphoanalytic and Psychometric Personality Profiles: Validity and Barnum Effects”, in press.

KLIMOSKY, Richard & Anat RAFAELI (1983): “Inferring Personal Qualities Through Handwriting Analysis”, Journal of Occupational Psychology, Vol. 56, pp. 191-202.

MARKS, David & Richard KAMMANN (1980): The Psychology of the Psychic, Buffalo, Prometheus Books.

MOORE, Michael (1985): “About the Sad State of Scientific Graphology”, Psychological Documents, Vol. 15, No. 2 (MS #2676).

NEVO, B. (ed.) (1986): Scientific Aspects of Graphology A Handbook, Springfield, Ill., Charles C. Thomas.

PATTERSON, J. (1976): Interpreting Handwriting, New York, McKay Press.

RAFAELI, A. & R.J. KLIMOSKI (1983): “Predicting Sales Through Handwriting Analysis”, Journal of Applied Psychology, Vol. 68, pp. 212-17.

SNYDER, C.R., R.J. SHENEKL & C.R. LOWERY (1977): “Acceptance of Personality Intepretations: The ’Barnum Effect’ and Beyond”, Journal of Consulting and Clinical Psychology, Vol. 45, pp. 118-23.

SUNDBERG, N.D. (1955): “The Acceptability of ’Fake’ versus ’Bona Fide’ Personality Test Interpretations”, Journal of Abnormal Social Psychology, Vol. 50, pp. 145-7.


1. Not only in terms of the consequences of dishonesty such as financial loss and embarrassment, for example; but it is often difficult to remove an employee once hired: union agreements, morale problems amongst the remaining employees and so on make removing an employee a trying process which all institutions have good reason to prevent if possible.

2. Not to be confused with the reputable school of academic psychology by the same name.

3. “Graphoanalysis” is the registered trade name of the system taught by the descendants of Bunker, the International Graphoanalysis Society Inc. As far as we know, none of the major B.C. graphologists are followers of this system.

4. Others include that the textbooks that are read by “specialists” in the field are the same ones to inform laypeople, and are readily available in suburban bookstores; and that the “discipline” is taught in short noncredit night school courses such as those offered by the Vancouver School Board. Both of these criteria are met by graphology. (Of course, graphologers here maintain that in Europe such courses are offered in bone fide psychology courses, and there are Chairs of Graphology in major universities. But they offer no proof. The B.C. Skeptics is in the process of checking our these assertions, so far with no success.) In addition, any field where positive studies are almost always self-published and critical attacks on these tracts appear in the most prestigious refereed scientific journals should be viewed with suspicion; and graphology scores as a pseudoscience on this criterion as well. Finally, graphology offers no coherent theoretical base, has numerous factual disputes, and its empirical findings are shaky, when they are offered at all. For a comprehensive review, see Nevo (1986).

5. Crumbaugh (1977) may be the closest to an exception here.

6. Not to be confused with Robyn Smith’s group, mentioned supra.

7. Note the open-mindedness of the scientists prepared to take the time to do these studies. Graphology has an incredibly low a priori probability for anyone who has the remotest understanding of neurophysiology and physiology. Is it probable (as the graphologists must be maintaining if their theory makes the slightest amount of sense), that traits such as promiscuousness or honesty (already complex dispositions, not simple things located in a few localized neurons in the brain), could have a somatic representation that is channeled into the motor cortex, down the pyramidal tracts, and out the alpha motor neurons to produce a unique writing style unique to all honest or promiscuous people whether they are left- or right-handed, male or female, taught to write by the Maclean’s Method in Vancouver in 1928 or the system taught in Topeka, Kansas in 1973…?

8. The Committee for the Scientific Investigation of the Paranormal (CSICIOP), affiliated with the B.C. Skeptics, has slated a session on graphology at its next conference on pseudoscience, in September, 1988.

9. And still considered to be so by many psychologists. But this is not to the present point. Our point is that those psychologists who do think there is some value to hypnotism use it; and given their superior skills, other knowledge, and research, do not have to fear the competition from old fashioned Mesmerists.

10. This term comes from the motto of his circus, “We have something for everyone”. However, it could just as well refer to two other sayings attributed to Barnum, “There’s a sucker born every minute”, and “Never give a sucker an even break”.

11. Note that we are not maintaining that graphologists are necessarily aware that this is what they are doing. In fact, graphologists are usually ignorant of the psychological literature, so are likely to be unaware of the Barnum Effect. Thus, it is compatible with our hypothesis that graphologers are unconsciously using the Barnum Effect, and are deceiving themselves, rather than intentionally deceiving their clients. But either way, the result remains the same: the client’s satisfaction with the reading remains a product of the flattering reading and vague wording, and is not evidence that the graphologist has discerned anything of significance about the client by the use of graphology.

12. For an account of the lengths to which people will go to defend beliefs they have committed their reputations to, cf. Festinger et al (1956).

13. It is interesting that he claims to be able to discern people’s sexual preference and orientation from their handwriting, but not their sex.