RESEARCH ARTICLE
Psychological Test Profiles of USAF Pilots Before
Training vs. Type Aircraft Flown
James E. Boyd, John C. Patterson, and Bill T. Thompson
BOYD JE, PATTERSON JC, THOMPSON BT. Psychological test profiles
of USAF pilots before training vs. type aircraft flown. Aviat Space
Environ Med 2005; 76:463– 8.
Background: Student pilots in the USAF are selected for fighter,
bomber, or airlift/tanker tracks after basic flight training. This selection
needs to be accurate in order to save time and training costs. The
objective of this study was to determine whether significant psychological differences exist between pilots flying different types of aircraft and
whether these differences could predict who will become a fighter pilot
(FP) vs. a bomber pilot (BP) or airlift/tanker pilot (AP). Methods: Pilots
who took the Multidimensional Aptitude Battery (MAB) and NEO Personality Inventory Revised (NEO-PI-R) were linked to their aircraft type
using primary USAF specialty codes. The data for 2105 pilots was
analyzed using MANOVA and Bonferroni post hoc analysis to evaluate
for relationships between test results and airframe assignment. Results: A
statistically significant difference was found between FP and AP pilot
means on all segments of the MAB and portions of the NEO-PI-R. The
mean scores of the FP group were higher on all IQ facets of the MAB. On
the NEO-PI-R, the FP group scored lower on agreeableness and higher
on conscientiousness. Discussion: The homogeneity of the pilot population gives the statistical difference in scores limited practical value for
predicting which aircraft a pilot is best suited to fly. However, scores on
these tests clearly could be a useful adjunct, along with flight training
grades and personal desires, in determining a student pilot’s potential for
success in the multi-tasking environment of the fighter pilot.
Keywords: multidimensional aptitude battery, MAB, NEO PI-R, medical
flight screening.
T
HE U.S. AIR FORCE flies a wide range of aircraft
types to support a variety of missions, each of
which places its own unique demands on the pilots that
fly them. It takes a significant investment of time and
money to train the pilots that fly these aircraft. Matching pilots to the aircraft type they are best suited to fly
has long been the goal of any flight-training program.
Historically, this aircraft matching process has primarily been based on each pilot’s flying skills and desires
(5).
Psychologists and aviation medical specialists have
long attempted to ascertain whether any unique personality traits in pilots could be used in pilot selection
and training (1,2,5,14). U.S. Air Force and U.S. Army
studies have revealed three distinct personality subtypes among experienced and successful military pilots
(10,11). The most prevalent type (58%) was identified as
achievement-oriented, dominant, and affiliative, and
this combination was labeled as “typical” military pilots, who are possessed of a structured, matter-of-fact
approach to problem solving. The second type (21%), in
addition to the previously listed traits, were more aggressive, dominant, exhibitionistic, and self-aggrandiz-
Aviation, Space, and Environmental Medicine • Vol. 76, No. 5 • May 2005
ing and were typified as having the “right stuff.” The
third type (21%), typified as having the “wrong stuff,”
were described as cautious, compulsive, socially retiring, and the least affiliative and achieving. However, as
in earlier German studies, none of these personality
types proved to be predictive of success as a military
pilot (6,14).
None of the previous studies have definitively addressed whether or not a “fighter pilot” personality
exists that could be identified by current psychological
testing. Is the often-exhibited fighter persona due to
inherent qualities, which have led them to gravitate
toward a common profession and aircraft type, or is it
due to the various environmental or squadron cultures
that have evolved over the years? Can psychological
testing prior to flight training identify measurable differences between pilots who eventually end up flying
fighter, bomber, and airlift/tanker aircraft? If differences do exist, are they significant enough to create a
predictive model that would be useful in determining
to which aircraft type a pilot is best suited?
Fighter pilots (FP) often must fly their aircraft to the
limits of their performance in aerial combat while simultaneously operating radar as well as offensive and
defensive weapon systems. These pilots need to be
self-reliant and capable of highly functioning in stressful, multitasking situations. While these traits are
present in these individuals prior to flight training, a
self-confident attitude is highly developed during
fighter pilot training.
The flying skills of airlift/tanker pilots (AP) are
somewhat different. They function as part of a two-pilot
crew and rarely need to push their aircraft to the limits
of its flight envelope. Given that all aircrew members
are stressed when flying in a high threat combat environment, the critical tasks of multiplace aircraft are
divided among the pilots and other crewmembers.
From the USAF School of Aerospace Medicine, Brooks City-Base,
TX.
This manuscript was received for review in March 2004. It was
accepted for publication in February 2005.
Address reprint requests to: James E. Boyd, M.D., M.P.H., Chief of
Aeromedical Services, MDOS/SGOAF, 16th Medical Group, A-113
Lielmanis Ave., Hurlburt Field, FL 32544; james.boyd@hurlburt.
af.mil.
Reprint & Copyright © by Aerospace Medical Association, Alexandria, VA.
463
PSYCHOLOGICAL TEST PROFILES—BOYD ET AL.
For more than 30 yr, the majority of USAF pilot
trainees followed a common pathway; at graduation
these all-purpose pilots could be assigned to any aircraft, including fighters, bombers, and airlift or tanker
airframes. Undergraduate pilot training (UPT) at the
basic level was completed in the T-37 followed by advanced training in the T-38. During the 1990s, the USAF
gradually adopted a system called Specialized Undergraduate Pilot Training (SUPT) in which the prospective pilots are selected into either fighter or airlift/
tanker aircraft after completing their initial phase of
flight training. Bomber pilots (BP) may still be selected
from either training track. Aircraft track selection must
now occur much sooner; at the end of the basic phase in
either the T-37 or newer T-6, pilots are selected for
either a fighter or airlift/tanker training track. Advanced training then continues in the T-38 for FP or the
T-1 for AP. BP may train in either track but train predominately in the T-1. As before, the students are
awarded their pilot wings after graduation from the
advanced phase.
The selection of a prospective pilot to an aircraftspecific track is based on the current needs of the USAF,
individual pilot skills, aptitude, and personal desires.
Student pilots must now decide which type of aircraft
they prefer before they have gained much flying experience. Also, instructors must critically evaluate trainees’ skills earlier for the purpose of matching them to
the type of aircraft that best matches their aptitude.
Currently the USAF requires all pilot candidates to
undergo Medical Flight Screening (MFS) at Brooks
City-Base, TX, or the USAF Academy in Colorado
Springs, CO, before student pilot training (9). Since
1994, the MFS has consisted of an echocardiogram, a
complete ophthalmology examination, neuropsychological baseline testing, and, more recently, anthropometry.
MFS psychological testing has four components: the
Cog Screen, a measure of learning and retention ability;
the Multidimensional Aptitude Battery (MAB), a broadbased intelligence test; the NEO Personality Inventory–
Revised (NEO-PI-R); and beginning in 1995, the Armstrong Laboratory Aviator Personality Survey (ALAPS)
(4,7,8,13). All candidates are required to complete the
Cog Screen and MAB to provide a neuropsychological
baseline, while the NEO-PI-R and ALAPS are optional
and are used primarily for research. None of these
psychological tests is currently used to identify pathology to screen for poor or unfit pilot candidates.
The critical process of selecting a general type of
aircraft a prospective pilot will fly now occurs earlier in
the training pipeline. Any additional objective data that
could be used to assist in this selection would prove
valuable in helping assign them to the type of aircraft to
which they are psychologically best suited. This couls
possibly lead to a long-term pilot satisfaction and better
retention as a military pilot. The purpose of this study
was to see whether significant psychological differences
exist between FP, AP, and BP using current baseline
MFS neuropsychological testing. If differences are
found, could they be used to predict into which type of
aircraft a pilot ultimately flies?
464
METHODS
This retrospective cohort study was designed to determine whether there were statistically significant differences in psychological test results among three broad
groups of USAF pilots: those who fly fighter, bomber,
and airlift/tanker aircraft. Subjects were not separated
either by age, race, or gender. Breaking them out as
separate groups for this study was deemed impractical
and would have added complexity. Only MAB and
NEO-PI-R test scores were used because they have
well-established norms and are used most often in clinical settings.
The inclusion criteria were: completion of UPT/SUPT
and assignment to a fighter, bomber, or airlift/tanker
aircraft; completion of the MAB and NEO-PI-R neuropsychological tests; and a signed informed consent for
the above tests. Exclusion criteria were: failure to complete UPT/SUPT; assignment to an aircraft other than a
fighter, bomber, or airlift/tanker; absence of valid psychological test scores; or lack of a signed consent form.
Pilots in the USAF Reserves and Air National Guard
were excluded because data on aircraft type were unavailable. Using the Military Personnel Center database, pilots who met the above criteria were identified
by social security number and then matched to their
MFS test data. Once the databases were merged, all
personal identifying information was removed. Type of
aircraft assignment was determined and pilots were
grouped by primary USAF specialty code: i.e.,
Fighter ⫽ 11FXX; Bomber ⫽ 11BXX; and Airlift/
Tanker ⫽ 11AXX.
As of December 2000, a total of 8304 prospective
pilots had taken the psychological portions of the MFS
and could reasonably be expected to have completed
flight training and assignment to a primary aircraft.
However, 1144 of these pilots were excluded because
they were in the USAF Reserves or Air National Guard.
A total of 342 (4.3%) of MFS applicants were physically
disqualified and did not enter or complete flight training. Although there were no specific data available to
identify students who did not complete flight training,
average attrition rates indicate that approximately 700
active duty UPT/SUPT students (8 –10%) did not successfully complete pilot training. Also excluded were
approximately 600 prospective students who did not
take the NEO-PI-R prior to its addition in 1994.
This left a pool of 5518 pilots who could be included
in the study. Of these, 3413 MFS pilot candidates failed
to meet one or more of the above criteria or had insufficient information in the Military Personnel Center
database. This could be due to several reasons, including data entry errors, flying another category of aircraft
(i.e., helicopter), or still flying training aircraft while
assigned as First Assignment Instructor Pilots. A total
of 2105 (38%) pilots remained who met all inclusion
criteria and were successfully identified and matched.
These included 870 FP (mean age 23.3, SD 2.3), 159 BP
(mean age 23.5, SD 2.1), and 1076 AP (mean age 24.0, SD
2.5) (ages at time of testing). All had been briefed on the
purpose of the tests and had their questions answered.
It was emphasized that the neuropsychological tests
would not affect their selection status and the personAviation, Space, and Environmental Medicine • Vol. 76, No. 5 • May 2005
PSYCHOLOGICAL TEST PROFILES—BOYD ET AL.
Fig. 1. Comparison of IQ scores for USAF airlift/tanker (A/T), bomber (B), and fighter (F) pilots. VIQ ⫽ verbal IQ; PIQ ⫽ performance IQ; FSIQ ⫽
full-scale IQ.
ality testing was voluntary, as was consent to have their
results used for research. All had passed a Flying Class
I physical examination and completed UPT or SUPT.
The MAB is equivalent to the Wechsler Adult Intelligence Scale and is computer generated and scored
(7,12). Scores are reported as verbal and performance IQ
scales. The components of the verbal scale are information, comprehension, arithmetic, similarities, and vocabulary. The components of the performance scale are
digit symbol, picture completion, spatial thinking, picture arrangement, and object assembly. The scores of
each subtest for the general population norms are
scaled to a mean of 50 with a standard deviation of 10
(3,7,11). Verbal and performance scores are combined
for the full-scale IQ score. The general population norm
IQ scores are scaled to a mean of 100 with a standard
deviation of 15, which have summary score reliabilities
in the range from 0.94 – 0.98 (7,11). For this study the
verbal (VIQ), performance (PIQ) and full-scale IQ
(FSIQ) scores were primarily used for comparison.
The NEO-PI-R is a test that focuses on identifying
normal personality traits rather than pathology (4). The
validity of the NEO-PI-R has been evaluated extensively and is summarized in the test manual. It has 240
statements to which the subject responds to on a Likert
scale of 1–5. 1 equals “strongly disagree” and 5 equals
“strongly agree.” The test is self-paced and computeradministered using a standardized set of instructions
and is scored automatically when completed. The
scores are divided into five domains (neuroticism, extraversion, openness, agreeableness, and conscientiousness) each with six component facets that comprise each
domain. Raw scores are commonly converted and reported in a T-distribution compared with the general
population norms. For the purposes of this study the
NEO-PI-R raw scores were used. All domain and facet
scores were compared to seek subtle differences in personality that might have occurred between the three
airframe groups of interest.
Aviation, Space, and Environmental Medicine • Vol. 76, No. 5 • May 2005
RESULTS
The raw scores were subjected to a preliminary basic
descriptive analysis. Means, standard deviations, minimum and maximum ranges, skewness, and kurtosis
indices were calculated for each group. A multivariate
analysis of variance (MANOVA) was required to further analyze the data because of both multiple dependent (all MAB and NEO-PI-R scores) and independent
variables (primary USAF specialty code). A Bonferroni
post hoc analysis was used to assess specific differences
between the levels of the independent variables. Consistent statistically significant differences in the means
were found for all IQ scores and component sections of
the MAB as well as on some domain and facet portions
of the NEO-PI-R between FP and AP (Fig. 1 and 2)(See
also Tables A and B online*).
The mean MAB FSIQ scores were 122.0, 120.4 and
119.6 for FP, BP, and AP, respectively. The FP mean
score was the highest of the VIQ, PIQ, and FSIQ on the
MAB. There was a consistent, statistically significant
trend in each of the 10 sub-component scores of the
MAB between the FP and AP (See Table C online**).
When comparing the FP group to the AP and BP, it was
again statistically higher. On the FP and AP comparison
all the results were p ⱕ 0.000 for VIQ, PIQ, and FSIQ.
When comparing FP to BP the ranges were p ⬍ 0.025 for
VIQ, p ⬍ 0.004 for PIQ, and p ⬍ 0.025 for FSIQ. However, though the means for FP were higher and were
statistically significant, there was considerable overlap
of the ranges for all groups (Fig. 1).
The FSIQ standard deviation (SD) of 6.3 was smallest
for the fighter group followed by the airlift/tanker
group with a SD of 6.7, and the BP having the largest SD
* For Tables A and B, see http://www.asma.org/ and click on the
link for the journal.
** For Table C, see http://www.asma.org/ and click on the link for
the journal.
465
PSYCHOLOGICAL TEST PROFILES—BOYD ET AL.
Fig. 2. Comparison of NEO-PI-R domain scores for airlift (A), bomber (B), and fighter (F) pilots. NEU ⫽ Neuroticism; EXT ⫽ Extraversion; OPN ⫽
Openness; AGR ⫽ Agreeableness; CON ⫽ Conscientiousness.
of 6.9. These data show that successful FP are a more
similar group when compared with the other two
groups. But again the overlap in scores indicates that
the three groups are much more similar to each other
than different (Fig. 1).
The NEO-PI-R personality results revealed statistically significant differences between the FP and AP in
two domains of the test, agreeableness and conscientiousness. The FP agreeableness scores were lower with
a mean difference of 1.9 (p ⱕ 0.000), and their conscientiousness scores were higher with a mean difference
of 1.6 (p ⱕ 0.002) (Fig. 2). On 9 of the 30 facet scores that
comprise the 5 major domains, there were again statistically significant differences between the FP and AP.
The facets that revealed significant differences with the
FP scoring higher were: assertiveness 2.0 (p ⱕ 0.000);
activity 2.5 (p ⱕ 0.000); competence 1.7 (p ⱕ 0.000); and
achievement striving 2.3 (p ⱕ 0.000). The facets on
which the FP scored lower were: anxiety 1.8 (p ⱕ 0.000);
self-consciousness 1.6 (p ⱕ 0.002); vulnerability 1.9 (p ⱕ
0.000); warmth 1.3 (p ⱕ 0.014); and tender-mindedness
1.3 (p ⱕ 0.026) (See Table D online†). The BP had no
significant trend and scored above, between, and below
the FP and AP.
DISCUSSION
Compared with the general population, all three
groups of pilots scored above the mean for intelligence
on the MAB, with a smaller standard deviation (14)
(Fig. 1). In general the IQ scores show that the majority
of BP and AP scored as well as or better than many of
those flying fighters and can be seen by the large overlap in MAB scores among all pilot groups. However, on
all IQ scores and component sections of the MAB, the
†
For Table D, see http://www.asma.org/ and click on the link for
the journal.
466
FP mean was statistically higher than the AP. The AP
group had the lowest scores in all areas of the MAB
tests with their lowest FSIQ score seven points below
that of the FP.
It does appear that the lowest FP FSIQ score may be
used as a minimum cutoff for entry into fighter training.
Any pilot scoring below it may not be the best candidate for fighter aircraft. This information could be used
to identify pilot candidates who would likely have difficulty in mastering the complex fighter environment.
The scores could be useful for both student pilots and
their commanders by helping assess their suitability for
an aircraft type when making aircraft track selections.
For instance, student pilots could use their scores to
assess how they compare with the average successful
FP and use the information to help decide on an aircraft
track to pursue. If their scores were above or near the
mean they could feel comfortable that they could succeed in fighters. Conversely, if they were to score significantly below the FP mean, they could conclude that
it may be more challenging for them to succeed in
fighter aircraft. Commanders could also use the information in considering which aircraft a student pilot
may be best suited to fly. It would not, nor should it be
a primary determinant in this decision.
Only a few FP actually scored higher than the top AP
in VIQ, PIQ, and FSIQ results. With a homogeneous
pilot population such as this, statistical differences,
while significant, are difficult to interpret, much less
apply. These results show that pilots of all aircraft types
are more alike than they are different. Candidates who
take these tests already have a strong desire to be pilots.
All are college graduates who have scored well on tests
such as the USAF Officer Qualifications Test, which
contains some aviation-oriented questions.
The NEO-PI-R personality results are more challenging to interpret. In 2 domains, agreeableness and conscientiousness, as well as 9 of 30 facets, statistically
Aviation, Space, and Environmental Medicine • Vol. 76, No. 5 • May 2005
PSYCHOLOGICAL TEST PROFILES—BOYD ET AL.
significant differences in the scores emerge between the
FP and AP. A low agreeableness score indicates a tendency toward a more antagonistic personality with the
individual being more willing to fight for his own interests if the need arose. Also, the lower scores indicate
a tendency toward the skeptical and critical thinking
skills needed for an accurate critical analysis in fastpaced situations, traits that would be a virtue in the
fighter environment (4).
FP had higher conscientiousness scores, which indicates individuals who are typically purposeful, strongwilled, and determined, and thus implies a tendency
toward higher academic and occupational achievement
(4). High conscientiousness is a primary component of
character and high scorers are more scrupulous, punctual, and reliable in their work habits and achievement
of goals. Although agreeableness and conscientiousness
domains differed significantly between FP and AP, all
of the domain scores again show a significant overlap
among all pilot groups (Fig. 2).
IQ and personality tests do have limitations and not
all areas critical for success are measured. These tests
primarily measure potential rather than actual performance. Success as a pilot is largely dependent on three
factors. These are mental and physical ability, emotional stability, and personal motivation. Psychological
tests currently available, such as the MAB and NEOPI-R, all give a fairly accurate portrayal of a prospective
pilot’s ability and stability. Motivation is the critical
driving force for success in any endeavor, but it has
been more elusive and difficult to assess. A motivated
individual with borderline IQ scores compared with his
peers can use hard work and dedication to become a
successful, skillful, and even superior pilot. Another
individual with a higher IQ may lose his or her desire to
fly and either drop out or continue with marginal performance.
Several factors may have confounded these results.
Only 38% of those who completed the MFS testing and
were eligible for inclusion in this study were successfully matched to an aircraft type in the Air Force Personnel Center database. Due to USAF needs, a pilot has
often been selected for an aircraft assignment that did
not match his/her desires. There were also differences
in instructor ability and grading criteria during UPT/
SUPT and the aircraft track selection process. Whether
due to aircraft availability, piloting skills, instructor
grading, or a combination of each, no records were
available to identify the basis for selection to an aircraft
type.
In the future, a similar study should be performed by
analyzing the remaining MFS data (ALAPS and Cog
screen tests). By adding the data gleaned from those
tests it may be possible to further define the performance parameters needed for success in a particular
aircraft type and achieve an accurate predictive model.
Demographics such as gender, age, and race, which
were not used in this study, could be relevant factors
for future study. Future research could also compare
pilots who have successfully completed flight training
with those who failed to complete UPT/SUPT.
It would be interesting to do a study comparing MFS
Aviation, Space, and Environmental Medicine • Vol. 76, No. 5 • May 2005
scores with each pilot’s aircraft preference after he or
she gains some flight experience and just prior to his/
her selection for an aircraft training track. This would
help reduce the confounding effect that arises when
pilots are selected for a cockpit that was not their primary choice. This would be labor intensive and involve
either contacting each pilot individually or obtaining a
copy of any aircraft preference forms (“dream sheet”)
from their service record and matching the results to the
MFS scores. If this could be achieved, the authors believe that it would yield greater accuracy and possibly
show even more differences between those who prefer
to fly fighters vs. those who wish to fly other aircraft. It
would also be useful to compare MAB scores with
primary flight training grades, if they were accessible,
to verify whether those who had the lowest IQ scores
were indeed the ones performing the worst.
CONCLUSIONS
There are statistically significant differences in the
mean scores on the MAB and NEO-PI-R. On the MAB,
the FP group had higher means than the AP group on
all portions of the test. On the NEO-PI-R domains the
fighter group means were lower in agreeableness and
higher in conscientiousness than the AP. These two
domains suggest that the FP group has a tendency
toward narcissism with a willingness to fight if needed
and are purposeful, strong-willed, determined, and successful. All are qualities that are essential in a fast-paced
combat environment.
However, because of the homogenous nature of the
pilot population as indicated by the data, a prediction
model cannot be developed at the present time. There
are very few differences in the pilot groups on these
tests as seen by the overall marked similarity of the
pilot’s scores. Though statistically significant, not
enough difference in the MAB and NEO scores is seen
to predict into which aircraft type an individual should
be assigned given the wide overlap in scores. However,
all FP scored higher than some of the lower scoring
pilots in both the BP and AP in each area of the MAB
(Fig. 1). These FSIQ test scores clearly could be used as
a useful adjunct, along with flight training grades and
personal desires, in determining a student pilot’s potential for success in the multitasking environment of
the fighter pilot.
ACKNOWLEDGMENTS
The authors would like to acknowledge the generous assistance of
the following agencies and individuals in the preparation of this
manuscript: the USAF School of Aerospace Medicine Department of
Graduate Medical Education, the USAF Residency in Aerospace Medicine, and the USAF Aeromedical Consult Service. Special thanks go
to Pamela D. Smith, M.D., M.P.H., Chief of Aerospace Medicine, 15
ADS, Hickam, HI, for her advice and earlier unpublished work on the
relation of NEO-PI-R anxiety and depression scores to UPT class
ranking which provided valuable background material for this study.
The views expressed in this article are those of the authors and do
not reflect the official policy or position of the Department of the U.S.
Air Force, U.S. Department of Defense, nor the United States Government.
REFERENCES
1. Callister J, King R, Retzlaff P. Cognitive assessment of USAF pilot
training candidates. Aviat Space Environ Med 1996; 67:1124 –9.
467
PSYCHOLOGICAL TEST PROFILES—BOYD ET AL.
2. Callister J, King R, Retzlaff P, Marsh R. NEO-PI-R profiles of male
and female U.S. Air Force pilots. Mil Med 1999; 164:885–90.
3. Carretta T, Retzlaff P, Callister J, King R. A comparison of two
U.S. Air Force pilot aptitude tests. Aviat Space Environ Med
1998; 69:931–5.
4. Costa PT, McCrae RR. Professional manual revised NEO Personality Inventory (NEO-PI-R) and NEO Five-Factor Inventory
(NEO-FFI). Odessa, FL: Psychological Assessment Resources,
Inc.; 1992.
5. Flanagan JC. The selection and classification program for aviation
cadets (aircrew-bombardiers, pilots, and navigators). The Air
Surgeon’s Bulletin; 1942:229 –39.
6. German aviation medicine: World War II. Washington, DC: Department of the Air Force; 1950:1027–50.
7. Jackson DN. Manual for the Multidimensional Aptitude Battery.
London: Ontario, Canada; 1984.
8. Kay GG. COGSCREEN: professional manual. Odessa, FL: Psychological Assessment Resources; 1995.
468
9. King RE, Flynn CF. Defining and measuring the “right stuff”:
neuropsychiatrically enhanced flight screening (N-EFS). Aviat
Space Environ Med 1995; 66:951– 6.
10. Picano JJ. Personality types among experienced military pilots.
Aviat Space Environ Med 1991; 62:517–20.
11. Retzlaff PD, Gibertini M. Objective psychological testing of U.S.
Air Force officers in pilot training. Aviat Space Environ Med
1988; 59:661–3.
12. Retzlaff PD, King RE, Callister JD. Comparison of a computerized
version of the paper/pencil Version of the Multidimensional
Aptitude Battery (MAB) (AL/AO TR-1995– 0121). Washington,
DC: U.S. Government Printing Office; 1995.
13. Retzlaff P, King RE, Callister JD, et al. The Armstrong Laboratory
Aviation Personality Survey (ALAPS): development, norming,
and validation. Mil Med 2002; 167:1–7.
14. Siem F. Predictive validity of an automated personality inventory
for Air Force pilot selection. Brooks AFB, TX: U.S. Air Force.
Report No.: AD-A226731, AFHRL-RP-90 –55.
Aviation, Space, and Environmental Medicine • Vol. 76, No. 5 • May 2005
PSYCHOLOGICAL TEST PROFILES—BOYD ET AL.
TABLE A. MULTIDIMENSIONAL APTITUDE BATTERY VIQ, PIQ, FSIQ.
Aircraft
Category/IQ
Range
Minimum
Maximum
Mean
SD
Airlift/Tanker VIQ
Bomber VIQ
Fighter VIQ
Airlift/Tanker PIQ
Bomber PIQ
Fighter PIQ
Airlift/Tanker FSIQ
Bomber FSIQ
Fighter FSIQ
47
38
44
54
41
56
47
36
45
92
98
96
87
96
90
91
102
98
139
136
140
141
137
146
138
138
143
118.59
119.66
120.59
117.88
118.40
120.48
119.57
120.43
121.99
6.94
6.92
6.77
8.42
8.53
8.08
6.71
6.90
6.30
VIQ ⫽ Verbal IQ; PIQ ⫽ Performance IQ; FSIQ ⫽ Full-Scale IQ.
TABLE B. REVISED NEO PERSONALITY INVENTORY (NEO-PI-R) DOMAIN (LETTERED) AND FACET
(ALPHANUMERIC) MEANS (RAW SCORE) AND SD.
Fighter (N ⫽ 870)
Domains/Facets
NEUROTICISM
Anxiety
Angry Hostility
Depression
Self-consciousness
Impulsiveness
Vulnerability
EXTRAVERSION
Warmth
Gregariousness
Assertiveness
Activity
Excitement-Seeking
Positive Emotions
OPENNESS
Fantasy
Aesthetics
Feelings
Actions
Ideas
Values
AGREEABLENESS
Trust
Straightforwardness
Altruism
Compliance
Modesty
Tender-Mindedness
CONSCIENTIOUSNESS
Competence
Order
Dutifulness
Achievement Striving
Self-Discipline
Deliberation
(N)
(N1)
(N2)
(N3)
(N4)
(N5)
(N6)
(E)
(E1)
(E2)
(E3)
(E4)
(E5)
(E6)
(O)
(O1)
(O2)
(O3)
(O4)
(O5)
(O6)
(A)
(A1)
(A2)
(A3)
(A4)
(A5)
(A6)
(C)
(C1)
(C2)
(C3)
(C4)
(C5)
(C6)
Airlift/Tanker
(N ⫽ 1076)
Bomber (N ⫽ 159)
Mean
SD
Mean
SD
Mean
SD
p*
45.81
46.34
48.37
46.32
45.73
48.05
41.30
57.27
51.12
54.77
59.23
59.59
61.73
55.06
50.70
52.61
48.86
51.92
52.24
54.85
46.46
43.45
50.48
48.64
51.67
46.01
46.72
45.35
55.39
56.81
50.21
52.89
60.22
53.07
50.32
9.49
9.30
10.40
7.95
9.57
11.28
8.52
9.49
9.65
10.04
9.06
8.40
8.53
9.61
10.93
10.99
11.09
11.21
10.55
10.71
10.82
11.03
10.15
10.01
10.14
11.80
10.50
10.24
10.20
8.86
10.46
8.82
9.15
9.53
10.35
47.16
46.69
47.63
46.18
46.87
47.80
42.08
58.01
51.04
54.57
58.02
58.78
61.51
54.30
50.67
52.49
48.98
52.49
54.36
53.86
45.10
44.94
49.67
48.53
54.03
46.96
47.13
48.01
55.51
54.91
51.57
54.14
60.29
53.09
51.17
9.27
8.72
9.72
7.57
10.04
10.46
8.49
10.56
10.54
10.61
9.63
9.96
7.87
10.31
9.32
11.01
10.22
9.50
9.41
10.46
10.00
11.07
11.08
10.63
9.84
12.02
12.18
9.46
9.97
8.68
10.68
8.67
9.85
8.08
9.14
46.76
48.15
48.02
46.65
47.30
48.55
43.21
57.75
52.39
55.46
57.20
57.12
61.23
55.13
51.06
52.96
49.86
53.17
52.43
53.85
47.33
45.33
50.17
48.13
52.80
47.16
47.75
46.60
53.83
55.06
50.41
52.07
57.95
52.01
50.47
9.32
9.30
9.72
7.95
9.66
10.81
8.42
9.04
9.25
9.86
9.03
8.89
8.21
9.84
9.90
10.28
10.52
11.26
10.39
10.42
10.37
10.66
10.47
10.33
10.26
10.93
10.66
10.05
10.02
9.04
10.87
9.19
9.40
9.60
10.09
0.083
0.000**
1.000
1.000
0.002**
1.000
0.000**
0.798
0.014**
0.431
0.000**
0.000**
0.601
1.000
1.000
1.000
0.139
0.055
1.000
0.128
0.238
0.000**
1.000
0.897
0.054
0.090
0.123
0.026**
0.002**
0.000**
1.000
0.165
0.000**
0.055
1.000
*p-value comparing fighter and airlift/tanker pilots: Bonferroni ␣ adjustment for multiple comparisons.
**The mean difference is significant at the 0.05 level.
TABLE C. MULTIDIMENSIONAL APTITUDE BATTERY VIQ, PIQ, FSIQ.
Fighter vs Airlift/Tanker
Fighter vs Bomber
IQ
Mean Difference
p*
Mean Difference
p*
VIQ
PIQ
FSIQ
2.014
2.899
2.561
0.000**
0.000**
0.000**
1.629
2.477
2.165
0.025**
0.004**
0.001**
VIQ ⫽ Verbal IQ; PIQ ⫽ Performance IQ; FSIQ ⫽ Full-Scale IQ.
*p-value: Bonferroni ␣ adjustment for multiple comparisons.
**The mean difference is significant at the 0.05 level.
PSYCHOLOGICAL TEST PROFILES—BOYD ET AL.
TABLE D. REVISED NEO PERSONALITY INVENTORY (NEO-PI-R) DOMAIN (LETTERED) AND FACET (ALPHANUMERIC) MEAN
DIFFERENCES (RAW SCORE) AND P-VALUE.
Fighter vs Airlift/Tanker
Domains/Facets
NEUROTICISM
Anxiety
Angry Hostility
Depression
Self-consciousness
Impulsiveness
Vulnerability
EXTRAVERSION
Warmth
Gregariousness
Assertiveness
Activity
Excitement-Seeking
Positive Emotions
OPENNESS
Fantasy
Aesthetics
Feelings
Actions
Ideas
Values
AGREEABLENESS
Trust
Straightforwardness
Altruism
Compliance
Modesty
Tender-Mindedness
CONSCIENTIOUSNESS
Competence
Order
Dutifulness
Achievement Striving
Self-Discipline
Deliberation
Fighter vs Bomber
Mean Difference
p*
Mean Difference
p*
0.094
1.802
0.361
0.329
1.569
0.497
1.916
0.474
1.269
0.688
2.027
2.472
0.502
0.007
0.368
0.351
1.006
1.241
0.193
1.007
0.869
1.877
0.307
0.502
1.139
1.157
1.033
1.249
1.554
1.742
0.197
0.819
2.270
1.057
0.146
0.083
0.000**
1.000
1.000
0.002**
1.000
0.000**
0.798
0.014**
0.431
0.000**
0.000**
0.601
1.000
1.000
1.000
0.139
0.055
1.000
0.128
0.238
0.000**
1.000
0.897
0.054
0.090
0.123
0.026**
0.002**
0.000**
1.000
0.165
0.000**
0.055
1.000
1.349
0.345
0.751
0.139
1.145
0.254
0.781
0.732
0.008
0.201
1.216
0.814
0.224
0.769
0.003
0.117
0.126
0.568
2.121
0.991
1.365
1.494
0.805
0.106
2.365
0.956
0.408
2.660
0.122
1.898
1.357
1.250
0.006
0.002
0.847
0.288
1.000
1.000
1.000
0.637
1.000
0.994
1.000
1.000
1.000
0.478
0.992
1.000
1.000
1.000
1.000
1.000
1.000
0.095
0.965
0.514
0.331
1.000
1.000
0.044**
1.000
1.000
0.017**
1.000
0.078
0.549
0.437
1.000
1.000
1.000
(N)
(N1)
(N2)
(N3)
(N4)
(N5)
(N6)
(E)
(E1)
(E2)
(E3)
(E4)
(E5)
(E6)
(O)
(O1)
(O2)
(O3)
(O4)
(O5)
(O6)
(A)
(A1)
(A2)
(A3)
(A4)
(A5)
(A6)
(C)
(C1)
(C2)
(C3)
(C4)
(C5)
(C6)
*p-value: Bonferroni ␣ adjustment for multiple comparisons.
**The mean difference is significant at the 0.05 level.