284 Assessment Psychology
Columbia, designed the Personal Data Sheet (Woodworth,
1920). The Personal Data Sheet consisted of a written list of
questions concerning presumed symptoms of psychological
disturbance (e.g., “Are you happy most of the time?”), which
were to be answered by checking “Yes” or “No.” Although
intended for use as a screening device to deselect emotionally
unstable draftees, Woodworth’s measure was not completed
in time to serve this purpose. Following the war, however, the
Personal Data Sheet was put to civilian use as a measure of
adjustment, and as such it was the first formal self-report
personality assessment questionnaire to become generally
available.
Although limited in scope and superficial in design,
Woodworth’s measure served as the model on which later
generations of adjustment and personality inventories were
based. Before continuing with that history, there is an his-
torical footnote to World War I that should be noted. The de-
velopment of the Personal Data Sheet as a model for an
enduring tradition in assessment psychology (i.e., personality
inventories), like the development of the Binet-Simon and
Army Alpha before it as models of other enduring traditions
(i.e., individual and group intelligence tests), bears witness to
the impetus of war and public need in evoking formal meth-
ods of psychological assessment. The tides of war inevitably
have their dark side, however, for those caught in the civilian
crossfire as well as for those coming under military attack.
In an event with broad sociopolitical implications, James
McKeen Cattell, after 26 years as a senior faculty member at
Columbia University was, according to Boring (1950,
p. 535), dismissed from his position in 1917 after taking a
pacifist stance with respect to the United States entry into
World War I.
Returning to the history of self-report measures, the next
major development following the Personal Data Sheet was the
publication by Robert Bernreuter (1901–1995) of a new Per-
sonality Inventory (Bernreuter, 1931). Unlike Woodworth’s
measure, which yielded just a single score for overall level
of adjustment, the Bernreuter was a multidimensional self-
report instrument with separate scales for several different
personality characteristics, such as neurotic tendencies, as-
cendance-submission, and introversion-extraversion. This
was the first multidimensional personality assessment mea-
sure to appear and, although the era in which it was widely
used and recognized is long past, the Bernrueter’s place in
history is assured by its having set the stage for a bevy of
similarly designed instruments that came to constitute a
cornerstone of assessment psychology. Among these many
multidimensional personality questionnaires, six currently
prominent instruments are notable for illustrating different
motivations and methodologies that have been involved in
developing such measures: the Minnesota Multiphasic Per-
sonality Inventory (MMPI), the California Psychological
Inventory (CPI), the Millon Clinical Multiaxial Inventory
(MCMI), the Sixteen Personality Factors Questionnaire
(16PF), the NEO Personality Inventory (NEO-PI), and the
Personality Assessment Inventory (PAI).MMPIThe Minnesota Multiphasic Personality Inventory was con-
structed during the late 1930s by Starke Hathaway (1903–
1995), a psychologist, and J. Charnley McKinley, a psy-
chiatrist, while they worked together at the University of
Minnesota hospitals. Hathaway and McKinley undertook this
task for the purpose of developing a group-administered
pencil-and-paper measure that would assist in assigning pa-
tients to diagnostic categories. The measure they produced
was first published in finished form in 1943 (Hathaway &
McKinley, 1943) and has since then become the most widely
used and researched of all personality assessment instru-
ments. The manner in which Hathaway and McKinley con-
structed the MMPI was noteworthy for their total reliance on
empirical keying in the selection of test items. Empirical key-
ing was a radical departure from the logical keying approach
that had characterized construction of the Woodworth and
Bernreuter tests and other early adjustment scales and trait
measures as well. In logical keying, items are selected or
devised on the basis of some reasonable expectation or
subjective impression that they are likely to measure a partic-
ular personality characteristic. Empirical keying, by con-
trast, involves selecting items according to how well in fact
they differentiate among groups of people previously identi-
fied as having various psychological disorders or personality
characteristics.
The original MMPI of Hathaway and McKinley was
expanded over the years by the addition of many new scales
and subscales, and an extensive revision and re-norming
process produced the MMPI-2 (Butcher, Dahlstrom, Graham,
Tellegen, & Kaemmer, 1989) and an adolescent version, the
MMPI-A (Butcher et al., 1992). Having been developed with
patient populations and for clinical purposes, the MMPI/
MMPI-2/MMPI-A is generally regarded as being more suited
for evaluating psychological disturbance than for elucidating
normal variations in personality characteristics. Neverthe-
less, the instrument has proved valuable in a variety of con-
texts and is often used by psychologists doing forensic,
neuropsychological, and personnel evaluations as well as
mental health assessments.