In this study, 13 Ph.D.-level staff members and
16 predoctoral trainees were asked to make judg-
ments regarding the diagnostic status of more than
800 patients based on the patients’MMPI scores.
These judgments were made without any contact
with the patient or any additional information on
the patient. Each judge simply examined the
MMPI profile (scores) for each patient and then
predicted whether the patient was“psychotic”or
“neurotic.” These judgments constituted clinical
predictions because it was left up to each judge as
to how she or he used the MMPI information to
formulate a diagnosis.
In contrast, statistical predictions involved the
application of a variety of algorithms, in which
MMPI scale scores were combined (added or sub-
tracted) in some manner and previously established
cutoff scores for psychosis versus neurosis were
used. In addition, some statistical predictions
involved the application of specified decision rules
based on MMPI high-point codes or other psycho-
metric signs. A total of 65 different quantitatively
based rules were considered.
What were these clinical and statistical predic-
tions compared to in order to assess their accuracy?
In this study, the criterion diagnosis was the psy-
chotic versus neurotic diagnosis provided by each
patient’s hospital or clinic. Thus, the accuracy of
each clinician’s and each statistical algorithm’s pre-
diction was determined by assessing the agreement
between predictions and the actual criterion diag-
noses across all cases.
Table 10-1 presents selected results from this
study. Not all of the statistical indices used are pre-
sented in this table, but it should serve to give the
reader a general impression of the results. First, judges
(clinical prediction) werecorrect, on average, in 62%
of the cases. It is noteworthy that the trainees’accu-
racy rate was comparable to that of doctoral-level
clinicians with more years of clinical experience.
Second, a number (14 in all) of statistical indices
(statistical prediction) were comparable to or outper-
formed the judges with regard to overall accuracy of
prediction. Several diagnostic decision rules based on
simple linear combinations of MMPI scale scores
(e.g., Pt–Sc) outperformed the clinicians.
A variety of additional, updated reviews of the
studies pitting clinical versus statistical prediction
have uniformly demonstrated the superiority of sta-
tistical procedures (e.g., Ægisdóttir et al., 2006;T A B L E 10-1 Accuracy of Clinical Versus Statistical Predictions in Goldberg’s (1965) Study
Source of Prediction Average Accuracy Percentage (n 861)
Clinicians
13 Ph.D.-level staff 62 (range 59 to 65)
16 Trainees 61 (range 55 to 67)
29 Total judges 62 (range 55 to 67)
Statistical Indices
(L Pa Sc) (Hy Pt) 70
Two-point code rules 67
Sc (Hs D Hy) 67
(Pt Sc) 65
High-point code rules 66
(Hy Pa) 61
Pa (Hs D Hy) 62
(Hs Sc) 61
(Pd Pa) (Hs Hy) 63
NOTE: Adapted from Goldberg (1965).
CLINICAL JUDGMENT 291