Clinical psychology has for some time debated
the merits of clinical (subjective) versus statistical
(objective, quantitative) prediction. Research
clearly supports the statistical approach to clinical
prediction. Although intuitively appealing, clinical
prediction is subject to a variety of biases (e.g., race,
social class, and gender), may lead to unwarranted
overconfidence, and is characterized by unreliability
and validity problems. However, objective decision
rules and statistical algorithms are not available for
most of the prediction tasks that face clinicians.
Therefore, clinical judgment and clinical prediction
are often required.
We offer a number of recommendations to
improve the reliability and validity of clinical judg-
ments: (a) Consider all available information, and do
not ignore inconsistent data. (b) Consider clients’or
patients’strengths and assets as well as pathology and
dysfunction. (c) Document all predictions, try to eval-
uate their accuracy, and use this information as feed-
back. (d) Use only structured interviews, structured
rating scales, objective personality tests, and behavioral
assessment methods to gather data. (e) Consider the
client’s situation and environment before making pre-
dictions. (f) Consider base rates and regression effects.
(g) Do not let one’s level of confidence influence pre-
diction. (h) Be aware of and guard against stereotyped
beliefs and illusory correlations.
The clinical report serves as the major form of
communication to convey the findings from a clin-
ician’s assessment and evaluation. The report should
address the referral questions and use language that is
tailored to the person or persons who will be reading
the report. Finally, the report should contain infor-
mation that is detailed and specific to the client and
should avoid vague, Barnum-like statements.KEY TERMS
Barnum effectA term applied in cases where
statements that appear to be valid self-descriptions
in actuality characterize almost everybody.
clinical interpretationA complex, inferential
process in which the clinician considers the
informationathand(e.g.,interviewdata,test
results) to conceptualize the problem and deter-
mine a course of action.
correlatesRelated variables. Clinicians employ-
ing a correlational orientation to patient data focus
on the presumed behavioral, attitudinal, or emo-
tional correlates of specific results.
quantitative or statistical approachAn
approach to clinical judgment and interpretation
that uses formulas and statistical models (already
derived) to make predictions about clinical out-
comes. Once the formulas have been established,
this approach involves no clinical decision making
at all.
referral questionThe question about the patient
that prompted the assessment.
samplesSpecimens of behavior. One orientation
to patient data views these data as samples of a
larger pool of information about the patient.
signsMarkers of underlying characteristics. One
orientation to patient data regards these data as
signs of some underlying state or trait.
stereotyped beliefsFixed beliefs (e.g., about
certain diagnostic signs, about certain demo-
graphically defined groups) that may influence
clinical judgment.
subjective or clinical approachAn approach to
clinical judgment and interpretation that is largely
intuitive and experiential. Subjective or clinical
interpretation requires that the clinician be sensitive
to information from a wide range of sources and
make a series of inductive or deductive generalizations
to link the observations and predict the outcome.CLINICAL JUDGMENT 307