In contemporary psychology, and in psychological assessment in particular, we increasingly encounter the term evidence-based — the demand that our procedures be grounded in empirical evidence. This demand is legitimate and important. At the same time, a second pole exists, labeled with some irony the eminence-based approach: practice grounded in the authority, experience, and personal conviction of respected clinicians. The question I want to pose is whether these two approaches must necessarily stand in opposition.
What the Evidence-Based Approach Means
The concept of evidence-based medicine was introduced in the early 1990s by Gordon Guyatt and developed principally by David Sackett. In its original definition, it does not rest on empirical data alone — it is the integration of three components: the best available evidence, clinical expertise, and the patient's values and preferences. This definition matters precisely because clinical experience is not excluded from the evidence-based approach; it is an integrated part of it. In polemics, however, this fact is often overlooked.
In the context of psychological assessment, an evidence-based approach means primarily using instruments with demonstrated validity and reliability, interpreting results against normative data and empirically established relationships, attending to base rates, and working with concepts such as incremental validity and clinical utility. These are all reasonable and desirable requirements.
What the Eminence-Based Approach Means
The label eminence-based originated as a critical parody of traditional practice, in which the correctness of a procedure was determined by the authority of an experienced clinician rather than by systematically gathered data. In their satirical BMJ piece on alternatives to evidence-based medicine, Isaacs and Fitzgerald (1999) defined eminence-based practice as "making the same mistakes with increasing confidence over an impressively long number of years." It is witty, but one-sided.
In psychological assessment, we should distinguish more carefully. What is labeled the eminence-based approach can take two fundamentally different forms. On one side stands the unjustified assertion of authorities, repeated by tradition and accepted as true because respected people say so — this is genuinely problematic and deserves criticism. On the other stands something quite different: authentic clinical experience, systematically built and corrected, representing a form of knowledge that cannot be fully captured in standard research designs.
What Science Is, and Where Its Limits Lie
To speak meaningfully about an evidence-based approach, we must first ask what "evidence" actually means. In the natural sciences, where the concept originated, the notion of proof has relatively clear contours: an observable and measurable phenomenon, a repeatable experiment, a falsifiable hypothesis. Science in this sense, as Karl Popper formulated it, advances through systematic attempts to refute hypotheses; what cannot be refuted is tentatively accepted as valid.
Psychology, however, occupies a different epistemological position. Its subject matter — the human psyche — is not directly observable, is not stably measurable, and is not fully reproducible. Psychological phenomena are context-bound, influenced by the very act of observation, and their understanding employs cognitive operations different from those used in knowing material objects. Already at the end of the nineteenth century, Wilhelm Dilthey distinguished between explanation (Erklären), proper to the natural sciences, and understanding (Verstehen), proper to the human sciences. Psychological assessment moves at the interface of both: it uses standardized instruments and statistical procedures, but at the same time requires the capacity to understand the unique narrative of a particular person.
Contemporary philosophy of science is also well aware that even the natural sciences are not as "hard" as was once assumed. Thomas Kuhn showed that scientific progress is not a linear accumulation of facts but proceeds through paradigmatic shifts in which the very framework within which facts acquire meaning is transformed. Michael Polanyi introduced the concept of tacit knowledge — we know more than we can tell — and showed that even in the natural sciences, the personal knowledge of the scientist, their intuition and capacity to recognize patterns that are not explicitly formulated, plays a substantial role. In psychological assessment, tacit knowledge is present to an even greater degree.
We must also recognize the methodological limitations inherent in empirical psychological research. Experimental and correlational studies work with averages and tendencies in populations. A result such as "the test has a validity of r = 0.45 against criterion Y" tells us something important about the statistical relationship between two variables, but by itself does not determine how to interpret a particular profile of a particular person. Psychological assessment works at the interface of nomothetic and idiographic approaches: it draws on general regularities but applies them to a unique case. And it is precisely in this step — in the transition from the general to the individual — that clinical judgment is engaged in a way that cannot be algorithmized.
Why These Approaches Need Not Stand in Opposition
If we restrict ourselves entirely and exclusively to what is irrefutably empirically demonstrated, we will be limited in assessment. Assessment is more than the mechanical application of craft procedures. Learning a test manual — knowing the administration instructions, scoring rules, norm tables — is a necessary foundation, but by itself it is not sufficient. Just as knowledge of notation and harmony does not yet make someone a performer, knowledge of a manual does not yet enable the assessor to extract from a test all the information the test offers.
In the assessment process, the personality of the assessor matters significantly — their capacity to establish rapport with the person being examined, sensitivity to nuances of behavior, the ability to integrate diverse information into a coherent whole. This aspect of diagnostic work is neither less scientific nor anti-scientific. It is a different type of knowledge, shaped primarily through transmission from more experienced colleagues, systematic practice under supervision, and continuous self-reflection. In psychological assessment, there exists extensive non-codified knowledge — experiences and attitudes not captured in manuals but transmitted within training and supervision. It includes the ability to recognize when a formal test result is misleading, the art of asking targeted questions during administration, the feel for the qualitative aspects of performance. This is precisely Polanyi's tacit knowledge: the assessor knows more than they can explicitly formulate. Without this layer, assessment remains impoverished.
On Uncertainty, Certainty, and Professional Growth
A certain degree of uncertainty is a natural part of assessment work — at all levels of professional development. In the early phases of a career, this uncertainty is naturally more intense and may provoke the need to seek support in something that offers a sense of unambiguity and certainty. The evidence-based approach can then become attractive not as a methodological framework — which would be entirely justified — but as a promise that assessment work can be rid of ambiguity.
This tendency is understandable. When the assessor does not yet have sufficient feedback from their own clinical practice, it is natural to seek certainty in what is "proven." It becomes problematic, however, when the demand for evidence turns into a demand for absolute certainty — when the assessor is unwilling to draw any conclusion that cannot be supported by irrefutable evidence. Such an approach paralyzes the assessment process, because in psychology absolute certainty is generally unavailable. Psychology is not physics. Our findings are often probabilistic, context-dependent, and culturally conditioned. The maturity of the assessor is shown not by the absence of uncertainty, but by the capacity to work with it — formulating hypotheses, testing them, admitting alternative explanations, and yet arriving at clinically usable conclusions. Paradoxically, it is precisely the capacity to tolerate uncertainty that testifies to professional maturity, not to uncontrolled subjectivism.
At the same time, the opposite risk must be named: an excessive trust in one's own clinical impression that crosses the line from justified clinical judgment to what might be called impressionology — assessment based on impressions without reflection, without verification, without the willingness to subject one's conclusion to criticism. As Paul Meehl warned, clinical judgment is prone to confirmation bias: we tend to notice information that confirms our initial hypothesis and to overlook that which disconfirms it. Clinical experience has value when it is systematically built, corrected by feedback, and kept in contact with the current knowledge of the field.
A Case in Point: The Draw-A-Person Test
The tension between evidence-based and eminence-based approaches can be illustrated well through the Draw-A-Person test (DAP). This instrument has a long tradition in psychological assessment — its origins reach back to the work of Florence Goodenough in the 1920s, and it was later developed as a projective technique for personality evaluation by Karen Machover.
The empirical research on figure drawing yields a sobering picture. As a developmental measure of graphic expression — an instrument for assessing cognitive maturity in children on the basis of the presence, proportions, and detail of body parts in the drawing — it shows an acceptable degree of validity, though with clear limitations: ceiling effects in older children, the influence of socioeconomic background and prior experience with drawing. As a test of personality, however, its validity is inadequately established. Research has repeatedly shown that the linkage of specific drawing signs to personality traits has weak empirical support and is susceptible to illusory correlations, as Chapman and Chapman demonstrated in their classic 1967 study. The broader critique, articulated forcefully by Lilienfeld, Wood, and Garb (2000), is that human figure drawing tests used as projective personality measures fail the standard psychometric criteria of reliability and validity, and that clinicians continue to extract from them inferences the data do not warrant. Even the later attempt to rescue the drawing as a measure of intelligence — the DAP:IQ — did not survive empirical scrutiny; Imuta and colleagues, in a study pointedly titled "Drawing a Close to the Use of Human Figure Drawings as a Projective Measure of Intelligence," concluded that its validity as a screening measure of intelligence was not supported.
It is worth being precise about the evidentiary landscape here, because it is uneven. The Rorschach, after a long period of controversy, has been substantially rehabilitated through the Rorschach Performance Assessment System (R-PAS), which provides a standardized, internationally normed, and empirically grounded administration and scoring framework. The Thematic Apperception Test retains a modest evidence base for thematic and narrative analysis. Drawing tests — the DAP, the House-Tree-Person, and similar instruments — have not undergone a comparable rehabilitation. They sit closer to the eminence pole than to the evidence pole, and any honest assessment of their place in contemporary practice must say so.
And yet an experienced assessor may still derive information from a drawing that has clinical value — not as isolated evidence, and emphatically not as an increment of demonstrated validity, but as one qualitative datum within a broader mosaic. A drawing can generate hypotheses that are subsequently tested with other instruments. It can reveal qualitative aspects — the examinee's approach to the task, the degree of effort invested, signs of anxiety or disorganization — that standardized questionnaires do not capture. To call this incremental validity, however, is to make an empirical claim that the evidence does not support. The honest formulation is narrower and more modest: the drawing is a source of clinical hypotheses and process observation, not a personality measure whose added value has been demonstrated over and above other data.
This is precisely where the dichotomy of evidence-based versus eminence-based proves too simple — but in a way that cuts against the temptation to defend the drawing. A strictly evidence-based position would lead to the rejection of figure drawing as a personality instrument. A purely eminence-based position would allow the drawing a probative value it does not empirically possess. The reasonable assessor moves between the two poles, and in the case of the drawing specifically, moves closer to the evidence pole: aware of the method's limits, unwilling to draw personality conclusions from single signs, and using the drawing only as one hypothesis-generating observation among many.
The Assessor as Instrument
Good psychological assessment requires both: respect for empirical evidence and developed clinical competence. The evidence-based approach tells us which instruments to use and which conclusions are empirically supported. Clinical experience tells us how to use these instruments in the context of a living person — with their story, their motivation, their defenses, and their current life situation.
In psychological assessment, the most important instrument is the assessor themselves — their professional identity, expertise, ethical stance, capacity for self-reflection. Empirical evidence is a necessary correction that prevents clinical judgment from degenerating into impressionology. Clinical experience is a necessary correction that prevents assessment from becoming the mechanical processing of numbers without understanding of the person. The true assessor is not the one who rigidly clings to one pole, but the one who can integrate both — who knows where evidence ends and clinical judgment begins, and who is at the same time aware of where clinical judgment ends and impressionology begins. And, crucially, who is honest about which of their instruments sit on which side of that line. Between these boundaries lies the space in which genuine assessment work takes place.
This article serves educational purposes and does not constitute a diagnostic or methodological guideline. The discussion of the Draw-A-Person test is deliberately simplified; for a more detailed orientation in the psychometric properties of this and related instruments, the reader is referred to the critical reviews listed below.
Selected Sources and Recommended Reading
Chapman, L. J., & Chapman, J. P. (1967). Genesis of popular but erroneous psychodiagnostic observations. Journal of Abnormal Psychology, 72(3), 193–204.
Evidence-Based Medicine Working Group (1992). Evidence-based medicine: A new approach to teaching the practice of medicine. JAMA, 268(17), 2420–2425.
Goodenough, F. L. (1926). Measurement of intelligence by drawings. World Book.
Hunsley, J., Lee, C. M., & Wood, J. M. (2003). Controversial and questionable assessment techniques. In S. O. Lilienfeld, S. J. Lynn, & J. M. Lohr (Eds.), Science and pseudoscience in clinical psychology (pp. 39–76). Guilford Press.
Imuta, K., Scarf, D., Pharo, H., & Hayne, H. (2013). Drawing a close to the use of human figure drawings as a projective measure of intelligence. PLoS ONE, 8(3), e58991.
Isaacs, D., & Fitzgerald, D. (1999). Seven alternatives to evidence based medicine. BMJ, 319(7225), 1618.
Kuhn, T. S. (1962). The structure of scientific revolutions. University of Chicago Press.
Lilienfeld, S. O., Wood, J. M., & Garb, H. N. (2000). The scientific status of projective techniques. Psychological Science in the Public Interest, 1(2), 27–66.
Machover, K. (1949). Personality projection in the drawing of the human figure. Charles C. Thomas.
Meehl, P. E. (1954). Clinical versus statistical prediction: A theoretical analysis and a review of the evidence. University of Minnesota Press.
Meyer, G. J., Viglione, D. J., Mihura, J. L., Erard, R. E., & Erdberg, P. (2011). Rorschach Performance Assessment System: Administration, coding, interpretation, and technical manual. Rorschach Performance Assessment System, LLC.
Mihura, J. L., Meyer, G. J., Dumitrascu, N., & Bombel, G. (2013). The validity of individual Rorschach variables: Systematic reviews and meta-analyses of the comprehensive system. Psychological Bulletin, 139(3), 548–605.
Polanyi, M. (1966). The tacit dimension. Doubleday.
Sackett, D. L., Rosenberg, W. M. C., Gray, J. A. M., Haynes, R. B., & Richardson, W. S. (1996). Evidence based medicine: What it is and what it isn't. BMJ, 312(7023), 71–72.
