L I E> R:-ARiY:
Return this book on or before the
Latest Date stamped below.
University of Illinois Library
t -*J 13a
I -4 I9B8
2 8 1985
AERONAUTICS BULLETIN : NUMBER EIGHT
UNIVERSITY OF ILLINOIS
THE INSTITUTE OF AVIATION
Evaluation of the School Link
Special Methods of Instruction
Ten-Hour Private Pilot Flight-Training Program
Ralph E. Flexman, William G. Matheny, and Edward L. Brown
UNIVERSITY OF ILLINOIS BULLETIN
VOLUME 47, NUMBER 80; JULY, 1950. Published seven times each month by the University of Illinois. Entered
as second-class matter December 11, 1912, at the post office at Urbana, Illinois, under the Act of August 24,
1912. Office of Publication, 358 Administration Building, Urbana, Illinois.
THE INSTITUTE OF AVIATION, established in 1945 as the Institute of
Aeronautics, is operated as the administrative agency responsible for the
fostering and correlation of the educational and research activities related
to aviation in all parts of the University. Other functions include academic
instruction, flight training, management of the University of Illinois Airport,
and aeronautical research.
In connection with the latter function, the Institute issues two types of
publications . . . first, a group of reports on research results, and sec-
ond, a series of bulletins on aviation subjects of an extension service
nature to the citizens of the State.
The following publications have been issued:
Bulletin One: Municipal Airport Management, Leslie A. Bryan, 1947.
(Out of print)
Bulletin Two: Landscape Planting for Airports,
Florence B. Robinson, 1948.
Bulletin Three: Labor Relations in the Air Transport Industry Under the
Amended Railway Labor Act, E. B. McNatt, 1948.
Bulletin Four: Airport Zoning, J. Nelson Young, 1948. (Out of print)
Bulletin Five: Evaluation of the School Link as an Aid in Primary
Flight Instruction, A. C. Williams, Jr., and
Ralph E. Flexman, 1949.
Bulletin Six: Lightplane Tires on Turf and Concrete,
Leslie A. Bryan, 1949.
Bulletin Seven: Light Aircraft Operating Costs, Leslie A. Bryan, 1949.
Bulletin Eight: Evaluation of the School Link and Special Methods of
Instruction in a Ten-Hour Private Pilot Flight-Training
Program, Ralph E. Flexman, William G. Matheny, and
Edward L. Brown, 1950.
Publications of the Institute of Aviation will be sent free of charge
UNIVERSITY OF ILLINOIS
THE INSTITUTE OF AVIATION
Leslie A. Bryan, Ph.D., LL.B., Director
Bernice Schrader, A.M., Editor
Evaluation of the School Link
Special Methods of Instruction
Ten-Hour Private Pilot Flight-Training Program
Ralph E. Flexman, William G. Matheny, and
Edward L. Brown
Published by the University of Illinois, Urbana
This monograph is the eighth in a series of bulletins on aviation
subjects which are being issued from time to time by our Institute
of Aviation. It reports the significant results of another research ex-
periment in the use of a synthetic training aid, the School Link, in the
training of pilots.
The authors are trained in psychology and are also certified pilots.
Mr. Ralph E. Flexman has been a flight instructor in the Institute of
Aviation since 1945 and previously served in World War II as an
Air Force flight instructor. Mr. Edward L. Brown, likewise, has had
long experience in both civil and military flight instruction. Dr. William
G. Matheny is a member of the instructional staff of the University of
Illinois Department of Psychology
Mr. Flexman, Mr. Brown, and Dr. Matheny wish to express their
appreciation to Link Aviation, Inc., of Binghampton, New York, for
furnishing and maintaining the two specially modified School Links
and for sharing the expense of the experiment; to the personnel of the
Institute of Aviation and the Aviation Psychology Laboratory, De-
partment of Psychology, for their assistance with the experiment and
in the writing of this report; to Mr. William Plunkett and Mr. David
Kress, Civil Aeronautics Administration Air Safety Agents, and to
Mr. Earl Taynor and Hr. Harold Hayes, Private Pilot Flight Ex-
The authors gratefully acknowledge the assistance given by the
following instructors who had the greatest load in carrying out the ex-
periment: Misses Dora Dougherty and Elizabeth Seely, and Messrs.
Donald Pardue, Scott Hasler, Howard Burnett, John Howarth, and
The Institute of Aviation is glad to make available the informa-
tion contained in this monograph. The Institute is using these and
other experimental results to continue to improve and modify its own
curriculum for the instruction of beginning pilots. It is hoped that
others may derive help of a similar nature from this study.
In this monograph, as in all publications of the Institute, the
authors have had complete freedom to express their opinions, with the
understanding that they assume sole responsibility therefor.
., , rt _ Leslie A. Bryan
April, 1950 n . 4
The private pilot is a "maneuver doer." His rating testifies that he
has performed certain maneuvers proficiently and, therefore, may enjoy
all of the privileges of the private pilot. The tacit assumption is that
performance of these maneuvers to a standard level of proficiency
makes him competent to operate an aircraft under the full range of
conditions allowed by his ratings. Unfortunately, these maneuvers de-
mand only a small part of the over-all knowledge and ability necessary
for the private pilot to realize efficient utilization of his aircraft.
Many groups who are concerned with pilot training are aware that
the proficiency of the new private pilot is limited and attribute this
condition to the present training program and instructional techniques.
Attempts have been made to resolve this situation by revising the
Private Pilot Flight Test, eliminating certain time-consuming and, per-
haps, unnecessary maneuvers from the flight syllabus, assigning a
greater portion of the flight-training to cross-country flying, and using
Each method has certain advantages and may add to the number of
skills required by the truly proficient pilot, but it is unlikely that any
one method would enable the private pilot to acquire the diversity of
skills that he needs. The desirable solution would be a flight-training
program that encompassed the best of all of these methods. Even with
such a program, if present training methods and techniques were used,
it is not probable that increased proficiency could be attained in the mini-
mum of thirty-five hours required by the CAA for the training of a
private pilot. Also, it would not be desirable to increase the required
amount of flight-training time. The most logical solution seems to lie
in the development of new training methods that would permit the
elementary skills to be taught in a fraction of the time now necessary.
The remaining hours then would be free for instruction in the addi-
tional skills the private pilot needs for the normal use of his airplane.
The present experiment was conducted by the Aviation Psychology
Laboratory of the Department of Psychology and the Institute of
Aviation, both of the University of Illinois. The aim of the experi-
ment was to train students to pass the Private Pilot Flight Test after
only ten hours of flight time. Previous experience with the Link trainer
and new methods of instruction suggested that the ten-hour program
was a reasonable goal. The study was designed also to assess the relative
contribution of each of these two new techniques, that is, the use of
a synthetic trainer and the use of revised instructional methods.
4 SCHOOL LINK EVALUATION AND INSTRUCTION
In view of the interest and research in the general area of syn-
thetic flight trainers and to place the present experiment in its proper
perspective, it seems advisable to present a brief summary of the more
important experiments in this area.
A recent report by Mahler and Bennett 1 includes a review of several
studies in the area of synthetic flight training. This review presents
the following findings: In 1939 E. W. Conlon, under the auspices of
the National Research Council, conducted a study of the use of the
Link Instrument Trainer in contact flight instruction in conjunction
with the Civilian Pilot Training Program. Conlon offered the tentative
conclusion that five to seven hours of contact instruction in the Link
"teaches a correlation of controls, and so forth, equivalent, roughly, to
three hours of flight instruction." 2
E. B. Greene's study, reported in 1941, involving the use of the
Link Instrument Trainer in contact flight instruction, indicated that
"Link training seemed to have given the men unusually good coordi-
nation and calmness in the air." 3 It is estimated that there was a saving
of about 2Yi hours in air time prior to solo as a result of the six hours
of Link training.
C. W. Crannell, Greene, and H. F. Chamberlain did another study
in 1941 similar in nature to the other two cited. These authors con-
cluded that "the consistency with which the groups with more Link
time are found to be superior throughout the battery of ratings and
other tests of flight competency suggests that some real differences in
ability may be present." 4
The war years lent urgency to the need for further knowledge in
this field. In 1942, a synthetic contact flight-training study was initiated
at the Naval Reserve Aviation Base, Long Beach, California. During
this six-month study, 146 students received synthetic training. The
final report of this project concluded that the synthetic preflight train-
ing speeded up pilot training by reducing the number of dual instruction
hours required to solo and also decreased the number of students re-
ceiving down checks (failure) on the "safe-for-solo" check flight.
A more elaborate project was begun in 1943 at the Naval Flight
1 W. R. Mahler, and G. K. Bennett, Special Devices in Primary Flight Train-
ing: Their Training and Selection Value. Port Washington, New York: Special
Devices Center, Office of Naval Research, Technical Report — SDC 151-1-18,
August 31, 1949.
2 As cited by Mahler and Bennett, p. 8.
* Ibid., p. 9.
THE UNIVERSITY OF ILLINOIS 5
Preparatory School, William Jewell College, Liberty, Missouri. The
project ran for eighteen months. The subjects in the experimental
group received synthetic training in a Link Instrument Trainer which
was converted to contact use by removing the cockpit cover and placing
a cyclorama around the trainer. The control group did not receive any
synthetic flight instruction. The study was divided into two phases:
the first dealt with the preparatory school program and the second
with flight training in light airplanes in War Training Service schools.
In both phases the experimental students held a slight advantage over
the control students in their readiness for solo (that is, a student
capable of soloing before the end of the required number of flights),
actual solo time, and flight grades. Although the differences were not
statistically significant, the students themselves felt that the synthetic
training was a definite asset, and the flight instructors preferred to
teach students who had received such training.
Data taken from an unpublished Navy report shows that early in
1945 another experimental synthetic flight-training (primary) program
was undertaken at the Naval Air Station, Memphis, Tennessee. This
program made use of such synthetic devices as a rudder trainer and a
taxi trainer, as well as a contact adaptation of the Link Instrument
Trainer equipped with cyclorama. In the experiment 306 cadets were
paired, and one of each of the pairs was placed in the experimental
group. The remaining 153 students made up the control group. The
subjects were followed through flight stage "A" for data compilation.
An analysis of the tabulated flight grades revealed that the experimental
group which participated in the synthetic flight program received more
G (good) grades on practically all items than did the nonparticipating
These studies are only a few of the total number conducted during
the years from 1929 to 1945. Both the Royal Air Force of England and
the Royal Canadian Air Force investigated the role of synthetic flight-
training devices in their flight-training programs. In addition, a limited
number of colleges having flight-training programs, such as Stephens
College at Columbia, Missouri, experimented with converted Link In-
strument Trainers in order to utilize them in their primary flight-
It is important to note that the synthetic flight trainers used in the
experiments described thus far were designed for the teaching of in-
strument flight procedures. They were adapted for contact flight train-
ing by removing the cockpit cover so the students were able to fly the
trainer by some reference other than the flight instruments.
6 SCHOOL LINK EVALUATION AND INSTRUCTION
The use of converted instrument flight trainers as contact flight
trainers in these experiments imposed certain limitations. The in-
structor was unable to measure student performance on many contact
flight maneuvers. Control of the trainers by reference to other than
the flight instruments was difficult due to the configuration of the
trainers; the flight characteristics as portrayed by the flight instruments
were not accurately simulated in the actual movements of the trainers;
and the cockpit design was not comparable to that of the aircraft used
for flight instruction. These limitations probably influenced the amount
of transfer of training that could occur, for only a small portion of
the primary flight syllabus could be taught in the ground trainer.
Link Aviation, Inc., later designed a classroom demonstration de-
vice known as the School Link. This device seemed to approximate
more closely the characteristics thought to be desirable in a synthetic
primary (contact) flight trainer than had the trainers used previously.
To evaluate this device in the role of a contact flight trainer, a series of
experiments was undertaken at the University of Illinois Airport.
The initial experiment
The initial experiment was conducted during the fall semester,
1947-1948, and involved forty-eight subjects. 5 The criterion used to
measure the transfer of training from the School Link to the flight
situation was the flight time required to solo. The design of the ex-
periment included two experimental groups and a control group. The
experimental groups first learned a number of maneuvers in the Link.
Proficiency in these maneuvers was assumed to be necessary before
solo in the aircraft was possible. However, the subjects were not re-
quired to perform the maneuvers to any set level of proficiency, but to
learn them as well as possible in the time allotted. After either two or
four hours in the trainer, they were assigned to one of four flight
Each flight instructor received twelve students; six from the con-
trol group and six from the experimental groups. His experimental
students were further divided into two groups; three students had had
two hours, and three had had four hours of Link-training experience.
The flight instructor was unaware of the experience in the trainer that
had been given any of his students. His task was to solo all of his
Students as quickly as possible while maintaining the conventional
standards of proficiency in the maneuvers required before solo. The
"'A. C. Williams, Jr. and Ralph E. Flexman, Evaluation of the School Link
as an Aid in Primary Flight Instruction. Urbana: University of Illinois, 1949.
THE UNIVERSITY OF ILLINOIS 7
hypothesis was that, if the required pre-solo maneuvers could be
learned in the trainer, the four-hour group would solo with the least
amount of flight time, the two-hour group next, while the control group
would require the greatest length of time.
The results of this experiment showed that there was no real dif-
ference between the groups in respect to the number of hours of dual
instruction necessary before solo. The results were considered incon-
clusive as to the value of the trainer in this situation because the cri-
terion used was heavily weighted in the performance of landings and
take-offs which was not taught in the Link. They did indicate, however,
that the time required to solo was chiefly dependent upon the instruc-
tors' standards and the student's ability to make take-offs and landings,
and not upon performance of the pre-solo maneuvers. Since time to
solo was the only criterion considered, there was no satisfactory meas-
ure of the transfer of training involved on the maneuvers that were
practiced in the trainer.
The second experiment
A second evaluation of the device was conducted during the spring
semester, 1948. The procedure in this experiment was to compare an
experimental group to a control group on the amount of practice re-
quired to achieve proficiency in a series of maneuvers performed in
the aircraft. The criterion of learning for any given maneuver was the
performance of three consecutive errorless trials. Each group was
comprised of twenty-four students, none of whom had had previous
flight training. In the trainer the experimental group first learned to
proficiency an exercise composed of several related maneuvers. After
learning the exercise in the trainer, the student practiced it in the
airplane until he again reached the criterion. The control group prac-
ticed the maneuvers in the airplane only and were required to achieve
the same level of proficiency as the experimental group. The students
were assigned in groups of twelve to each of four instructors. Six of
each instructor's students were from the control group, and six were
from the experimental group.
This second experiment was different from the first in several other
ways. The flight instructor served also as the Link instructor. The in-
structional methods used in the airplane and the trainer were not those
ordinarily used in flight training. Each maneuver was demonstrated to
the student once; then he attempted to perform it. The student was
told, at the end of a trial, what error or errors he had made and what
he must do on the next trial to correct the error or errors. If no errors
8 SCHOOL LINK EVALUATION AND INSTRUCTION
were made, the student was instructed to repeat the trial. However, if
the student persisted in his errors for three trials, the maneuver was
again demonstrated, and each demonstration after the initial one was
recorded as a trial. This procedure was continued until the criterion
The results of this experiment showed that the Link-trained group
had learned the maneuvers to criterion with 28 per cent fewer trials
and 22 per cent fewer errors than had the control group. An over-all
transfer of training of approximately 25 per cent occurred, and one
trial in the Link was estimated to be the equivalent of .29 of a trial
in the air.
Further analysis of the data revealed that the transfer of training
was not the same on all exercises. An examination of these differences
indicated that on certain exercises the amount of transfer probably
could be increased by increasing the accuracy with which the School
Link simulated the flight characteristics of the aircraft used. In ad-
dition, it seemed likely that, if the flight syllabus and instructional
techniques were modified to take advantage of the training received in
the School Link, an even greater saving in flight time might be accom-
plished, and, if the restrictions imposed by experimental control were
relieved, the effectiveness of Link training would be enhanced. This
would permit flexibility in using the trainer to meet the needs of indi-
The third experiment
To investigate part of the preceding hypothesis, a third experiment
was conducted during June and July, 1949. This experiment was con-
cerned primarily with ways of using the trainer and with improving
instructional techniques. Its purpose was to train students to private
pilot proficiency in as few hours as possible by using any method
deemed effective by the experimenter. The School Link used in this
exploratory project was modified in accordance with the findings of
the second experiment.
The design of this study involved six students who had had no
previous flight training. The group was under the control of one in-
structor at all times. 6 The criterion used was the time required to
achieve proficiency on basic maneuvers as indicated by passing a CAA
Private Pilot Flight Test.
The results of this project showed that the average time in the air
before solo was two hours and fifty-two minutes, while the average
'The instructor was Mr. Ralph E. Flexman, Institute of Aviation, University
THE UNIVERSITY OF ILLINOIS
Table 1. Distribution of Students
Number of Students
Number of Students
Link Control Total
Link Control Total
Students recruited from
Institute of Aviation. .
9 8 17
8 7 15
Students recruited from
other sources .
14 16 30
13 14 27
23 24 47
21 21 42
flight time required to reach proficiency was nine hours and forty-
seven minutes. The average Link time prior to passing the flight test
was ten hours and forty minutes. The average of the combined Link
and air time required to pass the test was twenty hours and twenty-
seven minutes. 7
These results indicated that it was possible for the new student to
achieve private pilot proficiency in substantially less time than is nor-
mally required. The design of the study, however, made it impossible
to determine the relative contribution of the methods used. Because of
this, the present experiment was undertaken.
THE PRESENT EXPERIMENT
The fourth experiment was designed to test the hypothesis that
a student could be trained to pass the Private Pilot Flight Test after
ten hours of flight training if a revised training syllabus were used. It
was also designed to determine the contribution of a synthetic trainer,
the School Link, to such a program.
In order to evaluate the contribution of the synthetic trainer, two
groups of students were trained under the revised training syllabus.
The Link group was given practice in the synthetic trainer while the
control group was not. At the end of the ten hours of flight time both
groups were given flight tests. The performance of the students on
these tests was the measure used to compare the two groups. Distribu-
tion of the students by group is shown in Table 1.
7 See Appendix A.
10 SCHOOL LINK EVALUATION AND INSTRUCTION
Table 2. Distribution of Students by Instructor and Group
Number of Students
Number of Students
In order to approximate more closely the use of the revised syllabus
and of the synthetic trainer in a "practical" training situation, eight
flight instructors were used. Although students were assigned to these
instructors at random, each instructor was to train an equal number
of Link students and control students. It was not possible to meet this
requirement completely, due to difficulties in scheduling. The distri-
bution of students by instructor and group is shown in Table 2.
All students were trained to "intellectual" proficiency in a group
of maneuvers before attempting to perform them. "Intellectual"
proficiency was defined as the ability to verbalize the pertinent cues
and cue responses necessary for successful acomplishment of the
maneuvers. When the subjects in the control group reached this level,
they attempted the maneuvers in the aircraft. When the Link subjects
attained this proficiency, they practiced the maneuvers in the synthetic
trainer until an acceptable level of performance was reached. Following
this, they attempted the maneuvers in the aircraft. Both groups of
students received an equal amount of training in the aircraft. The
distribution of each student's air time, Link time, and discussion time
is shown in appendices B and C.
Upon the completion of the ten hours of flying, each student was
given two Private Pilot Flight Tests. The first flight test was ad-
ministered by CAA Air Safety Agents, and the students were as-
signed to them at random. The second flight examination, given the
following day, was conducted by Private Pilot Plight Examiners.
THE UNIVERSITY OF ILLINOIS 11
Again, the students were assigned at random. Neither the CAA Ex-
aminers nor the Private Pilot Examiners were informed of the type
of training received by any student. They knew, however, that the
students had had only ten hours of flying time and that, in an attempt
to mask the identity of the experimental subjects, several private pilots
were added as "ringers" to the groups being tested. Unknown to the
CAA Agents and Examiners, the ringer group was composed of seven
private pilots recruited from other flight classes and their average
flight time was approximately 100 hours.
In addition to the measures of proficiency mentioned previously,
records of each student's flight time before he was allowed to solo, as
well as of the instructor's over-all grade of the student's proficiency at
the end of the ten hours, were kept.
Subjects participating in this experiment were selected from two
sources within the University. The first source was the Private Pilot
Flight Course at the University's Institute of Aviation. This elective
course, for which credit is given toward graduation, is offered to all
University students. The second source was other academic depart-
ments. Students from the latter source were not enrolled in the Private
Pilot Flight Course. They received only the ten hours of flight instruc-
tion and did not participate in the formal ground school program which
is part of the Private Pilot Flight Course. A person who had had
previous flight training could not become a subject for the experiment,
nor could a person who was unable to pass successfully a third-class
CAA physical examination.
A total of forty-seven subjects were selected, seventeen from the
first source and thirty from the second. The distribution of students
by source is given in Table 1. All students selected from the second
source were required to pay a fee of $25. The fee was assessed to
help assure the students' continuing in the experiment, as well as to
help defray the cost of the program. Prior to the completion of the
experiment, five students withdrew for personal reasons, and their data
were not included in the results. The subjects selected from the first
source received additional formal instruction in ground school; hence,
the number of these subjects in the Link and the control group was
equated as nearly as possible.
Eight instructors, six men and two women, were used. These in-
structors were selected on the basis of availability and experience level.
12 SCHOOL LINK EVALUATION AND INSTRUCTION
The experience level of the instructors ranged from no previous in-
struction time to over 2,500 hours of instruction time. Only one had
previously given instruction in the School Link. The instructors were
given approximately five hours of indoctrination and training. They
were briefed on the purpose of the experiment, the methods of in-
struction to be employed, the controls to be observed, and the use of
the School Link. Instruction in the use of the School Link included
demonstrations of its flight characteristics and its limitations and of
starting and stopping procedures, as well as information concerning the
coordinated use of the blackboard in teaching ground contact maneuvers
and the tolerances to be observed in grading the various maneuvers.
After the initial indoctrination period, each instructor received
a mimeographed pamphlet, "Purpose of the Experiment, Instructor's
Guide, and Some Principles of Learning Applicable to the Flight
Situation," and copies of the Link and the Control syllabi. 8 The syllabi
were presented as guides, and strict adherence to them was not re-
quired. The students' individual differences determined the manner in
which each instructor applied the information. However, the im-
portance of equalizing instruction for the control and the Link groups
was carefully explained to the instructors. The desirable instructional
techniques were outlined, but the actual instruction was not supervised
in either the air or the Link situation. The supervision of the instruc-
tors was concerned only with maintaining the basic design of the
The two synthetic flight trainers used in this experiment were
modified models of the preflight trainer known popularly as the School
Link. They were furnished by Link Aviation, Inc., of P>inghampton,
The trainer, made of aluminum, resembles an all-metal, open cock-
pit, midwing monoplane of the light aircraft type. Its wing span is
9 feet 6 inches, and it has the same measurement from nose to tail.
Mounted on a pedestal, the trainer is free to pitch, yaw, and bank, and
to turn 360° in response to control manipulation in the cockpit. This
manipulation is also reflected in the movement of the appropriate con-
The trainer possesses the following flight characteristics: dynamic
Stability, adverse yaw, nose-heaviness during turns, change in pitch
attitude with change in power-setting, overbanking tendency, torque
"Sec appendices I), E, and F.
THE UNIVERSITY OF ILLINOIS 13
effect with power and airspeed change, spinning (accidental and/or
intentional), elevator trim effect, power-on and power-off stall, and
positive stall recovery. In addition, the effects of turbulent air can be
simulated. The cockpit configuration is that of the usual light aircraft
and includes operative light aircraft instruments, the usual flight con-
trols, plus an elevator trim control, and a functioning headset and
A cloth cyclorama 15 feet in diameter, 11 feet in height, and 270°
in circumference was placed around each trainer. The white cloth was
unmarked except for a black line representing the horizon, reference
marks for normal climb and glide attitudes, and indications of cardinal
compass directions. A projector in the nose of the trainer made it
possible for the instructor to flash a small silhouette on the screen so
that he could check the aircraft's attitude or give the student a refer-
ence point in the initial stages of learning.
For certain maneuvers the student was required to wear blue gog-
gles, and the projector light was covered by a small piece of amber
plexiglass. This combination of blue goggles and amber plexiglass pre-
vented the student from seeing the projected silhouette. Thus the in-
structor could use it as a scoring device.
A portable schoolroom blackboard was used as a device for teach-
ing ground reference maneuvers. The blackboard was equipped with
rollers so that it could be moved to simulate drift effects and the
effects of forward and side slips.
A small wooden replica of a typical light airplane was available
for the demonstration and practicing of the various maneuvers and
attitudes. It did not have movable surfaces.
A wooden mock-up and an actual photograph of the University
Airport were used for student orientation and for teaching traffic
patterns. Both of these training aids were of a size which permitted
the student to move a small paper silhouette along the runways and
taxiways in a manner which simulated the actual taxiing and traffic
The airplane used for the actual flight training was an Aeronca
7 AC, a tandem-type, high-wing, lightplane powered by a 65-horsepower
Continental four-cylinder engine.
The CAA Flight Test administered either by a CAA Examiner or
by a Private Pilot Examiner is the final measure of the student's pro-
ficiency as a private pilot. The student must receive an over-all grade
of seventy or better on this test to be eligible for the private pilot rating.
14 SCHOOL LINK EVALUATION AND INSTRUCTION
Table 3. Coefficients of Correlation Among the Several Criteria
Measures of Over-all Grade
Link (n = 21)
Control (n = 2T
Link and Control
(n = 42)
Pilot Instruc- Pilot Instruc- Pilot Instruc-
Exam- tor Exam- tor
CAA Examiner. . . .
. . .240
This test consists of seventeen items or maneuvers. Each item is scored
on a scale from one to five; five being the failing score. Failure on any
one of the items constitutes failure to pass the flight test. 9
The grade given by the examiner at the conclusion of the flight test
is based not upon a simple summation or equivalent weighting of the
individual items or maneuvers, but rather upon the examiner's judg-
ment of the student's over-all performance during the test. The ex-
aminer may assign the student a grade between zero and one hundred.
The student's proficiency may be evaluated also by his instructor and
given a numerical value between zero and one hundred, with seventy or
higher indicating proficiency adequate to pass the CAA Flight Test.
A subgoal for the student in the early stages of training is time
to solo. This constitutes a "time to learn" or "time to criterion" measure
of proficiency, although there is undoubtedly variation between and
among instructors as to the level of proficiency believed necessary
Thus, the proficiency measures or criteria available for study of
training procedures within the private pilot training program may be
first, the CAA Flight Test, second, the instructor's grade, and third,
the amount of flight time before solo. For an over-all evaluation of
any training program, any one or a combination of these might be used.
The choice of any one or all would seem to be dependent upon the
immediate question to be answered concerning the training program
and upon available information as to the reliability of the criteria. In
'These items are shown in Appendix J.
THE UNIVERSITY OF ILLINOIS 15
Table 4. Correlation Between Solo Time and Over-all Grades
Graded by : Private
Group ~^\^ Examiner Examiner Instructor N
Control — . 374
Link and control — .451
general, however, the CAA Flight Test seems preferable since, prac-
tically speaking, it is the one which the student must pass to receive the
private pilot rating. Further, this flight test offers an opportunity to
analyze the effects of the training program in some detail in terms of
In the present study, the CAA Flight Test was administered to each
of the experimental groups by two groups of examiners. The CAA
Examiners checked the students first, and then the Private Pilot Ex-
aminers from other airports checked them. The flight test provided
data for three analyses: a dichotomous breakdown of "pass" or "fail";
the over-all grades received by both the Link and the control groups
on the check flights with either the CAA Examiners or the Private
Pilot Examiners; and the number of items passed or failed by the
students within each of the groups for each of the check rides. The
data were also analyzed with respect to the instructors' over-all grades
as well as the time before solo for each of the two experimental groups.
The several coefficients of correlation between the various measures
or criteria were computed in order to obtain an estimate of the extent
of agreement among the measures. These coefficients of correlation
are presented in Tables 3 and 4.
Table 3 shows that there was fair agreement among the various
observers in their grading of the students' over-all proficiency. The
lack of high correlation may be due, in part, to lack of reliability of
observations. Some notion as to the reliability of the check ride itself
as a measuring instrument may be obtained from the correlation be-
tween the CAA Examiners' grades and the Private Pilot Examiners'
grades. For the combined Link and control groups, this coefficient of
correlation was .52.
The coefficients of correlation between over-all grades and flight
time before solo were, in general, lower than those between over-all
grades. An exception to this, however, is the correlation between in-
16 SCHOOL LINK EVALUATION AND INSTRUCTION
Table 5. Frequencies and Percentages of "Pass" or "Fail"
on the Private Pilot Flight Test
CAA Examiners Private Pilot Examiners
Link Control Link Control
Nam- Per Nam- Per Num- Per Num- Per
her Cent her Cent her Cent her Cent
structor over-all grade and solo time, which suggests that the instruc-
tor's grading of a student may be affected to some extent by time to
It should be pointed out again that the CAA and Private Pilot
examiners were unaware of the type of training, either Link or non-
Link, received by any student being checked. Further, seven pilots
already holding private pilot licenses were tested along with the two
experimental groups. The examiners knew only that some students
were "experimental" and had received but ten hours of air training
time, while others were certified private pilots.
The statistical analysis of the data was oriented toward testing the
null hypothesis with respect to the two subgroups concerned, that is,
that there were no differences in performance between the Link and
the control groups as measured by any one of the criteria used. The
first criterion considered was the dichotomy of "pass" or "fail" on the
flight lest as administered by either the CAA Examiners or the Private
Pilot Examiners. The results of this analysis are shown in Table 5.
The hypothesis tested in this case was that there was no significant
difference between the proportions passing the test in each of the two
groups, cither Link or control.
The analysis showed that a higher percentage of students in the
Link group had passed the flight test for both types of examiners. A
test for the significance of the difference between proportions passing
THE UNIVERSITY OF ILLINOIS 17
Table 6. Means, Standard Deviations, and Tests of Significance
Between Over-all Grades of Link and Control Groups
for the Three Criteria Employed
X SD F 10 P u r n P
Link 68.14 5.23 3.25 .02 3.91 .001
Control 58.71 9.41
Link and control ... 63 . 43 8 . 96
Private Pilot Examiners
Link 67.71 4.21 3.11 .02 2.25 .05 .01
Control 63.42 7.42
Link and control. . . 65.57 6.40
Link 68.28 7.70 1.70 .20 1.24 .30 .20
Control 64.76 10.05
Link and control. . . 66.52 9.13
the test in each of the groups showed that, although more Link-trained
students than control students passed each of the flight tests, the dif-
ference was not highly significant. The results of the CAA Examiner's
test showed that 43 per cent of the Link group passed while 19 per cent
of the control group passed. This difference in proportions was found
to be significant at the .10 level of confidence. The difference in propor-
tions for the Private Pilot Examiner's test was found to be significant
at the .15 level of confidence. It is interesting to note that, of the
small group of private pilots (n = 7) who took the test at this time,
only 43 per cent passed.
Over-all grades on the CAA Flight Test, as well as the instructors'
over-all grades, were analyzed. Two hypotheses were tested for these
measures. These hypotheses were, first, there are no differences be-
tween the grougs in terms of the variability of grades received, and
second, there are no differences between the groups in terrns of mean
grade received. The results of these analyses are presented in Table 6.
Table 6 shows that in the evaluations by both the CAA and the
Private Pilot examiners the grades of the Link group showed signifi-
cantly less dispersion about the mean grade. In other words, the per-
10 F ratio between the variances of the Link and control groups.
"Palmer O. Johnson, Statistical Methods in Research. New York: Prentice-
Hall, Inc., 1949, pp. 74-75.
SCHOOL LINK EVALUATION AND INSTRUCTION
OVER-ALL GRADES ON CAA FLIGHT TEST
Percentage of Link Groups Receiving Score N — 21
R^^M Percentage of Control Group Receiving Score N = 21
OVER-ALL GRADE — CAA PRIVATE PILOT FLIGHT TEST
THE UNIVERSITY OF ILLINOIS 19
formance of the Link group was more homogeneous with respect to
over-all grades than was that of the control group. This trend was re-
peated in the instructors' grading, although the difference did not
approach the significance level of the other two sets of grades.
The lack of homogeneity of variance for the two groups poses a
problem in testing the null hypothesis with respect to mean differences.
The test of this hypothesis was made, however, utilizing the method
of W. G. Cochran and G. M. Cox as outlined by Palmer O. Johnson
in which a weighted criterion "t" is obtained. The complete computa-
tion was not necessary in this case, since N x = N 2 and the interpreta-
tion of "t" was made by entering the "t" table with df = Ni — 1.
The test of this second hypothesis — that is, that there are no
differences between the groups in terms of mean grades received —
showed that the CAA grades discriminated between the groups at a
high level of statistical significance (P = .001). The Private Pilot
Examiner grades discriminated at a lower level (P = <.05>.01),
while the flight instructor grades, although showing the same trend,
failed to show a high level of discrimination (P = <.30>.20).
The percentage of students receiving the various over-all grades
is shown in Figure I. The raw data are presented in appendices G,
H, and I.
Since over-all grades, as well as grades on the separate items in
the test, may reflect an over-all evaluation of the student by the ex-
aminer, it was decided to examine the groups in terms of the number
of items failed. That the over-all grade given by the examiner is par-
tially independent of the number of items failed is indicated by a
coefficient of correlation of — .479 between items failed on the CAA
test and the over-all grades for that test. Thus, there is no direct corre-
spondence between the over-all grade and the number of items failed,
which indicates a differential weighting of the items by the examiners.
The items were rated "pass," a score of one through four, or "fail,"
a score of five. The item scores for each of the students in the two
groups were tabulated to determine the number of students failing no
items, one item, and so forth. 12 With twenty-one cases in each group,
it was necessary to combine the data as shown in Tables 7 and 8 to
obtain enough cases in any one cell to make an analysis meaningful.
Chi square values were then computed for the two four-fold tables.
From Tables 7 and 8 it can be seen that the two groups of examiners
discriminated between the groups of students on the basis of the num-
ber of items failed at a level of statistical confidence of .03. No data
These scores are presented in appendices J and K.
20 SCHOOL LINK EVALUATION AND INSTRUCTION
were available as to the instructors' ratings in terms of items passed
A related measure of pilot proficiency which is sometimes used is
the amount of flight time required before solo. This measure was avail-
able for the two groups in the experiment and was examined for dif-
ferences between the groups. This analysis is summarized in Table 9.
Table 9 shows the Link group to have had less average flight time
than the control group had had before being allowed to solo. However,
Table 7. CAA Examiner's Ratings
Number of Items Failed
Failed to 3 items
Failed 4 to 9 items
Table 8. Private Pilot Examiner's Ratings
Number of Items Failed
Failed to 3 items
Failed 4 to 7 items
Table 9. Flight Time Before Solo
Means, Variabilities, and Tests of Significance
. ... 4.36
THE UNIVERSITY OF ILLINOIS 21
the mean difference between the two groups does not show a high
degree of statistical significance.
Some additional pertinent, though subjective, evaluations of the
performance of the group can be made. In the Link group an additional
six individuals, or 28.6 per cent, received scores from 66 through 69,
a score of 70 or better being passing. In the control group, no indi-
vidual received a score in this "near-passing" range. Thus, a total of
71.6 per cent of the Link group received scores of 66 or above, while
only 19 per cent of the control group received scores in this range. This
suggests that the Link group would have profited more than the control
group, in terms of the number of individuals passing, from an ad-
ditional two to three hours of training.
The findings of this experiment become more significant when
viewed in the light of the following facts: The records of 685 primary
flight students trained at the University of Illinois between the spring
semester of 1947 and the summer session of 1949 were examined; these
records showed that, among the students whose instructors recom-
mended that they take the Private Pilot Flight Test, only 67.9 per cent
passed the test at the end of 35.9 hours of flight training. The other
32.1 per cent were unable to pass the test at this time, either because
they were not recommended by their instructor, or because they failed
the test after recommendation.
Summary of analysis
After all the data relevant to the experiment were compiled, an
analysis of the data was made. The following points form a summari-
zation of this analysis:
1. A comparison of the performances of the two groups, the Link
and the control, showed that the Link-trained group was more pro-
ficient with respect to the several criteria analyzed.
2. The criterion of "pass" or "fail" on the flight test differentiated
between the groups but not at a high degree of statistical significance.
3. The criterion of over-all grade on the flight test showed that
the Link group received a significantly higher mean score than the
4. With respect to the criterion of number of items failed, the
analysis showed that a significantly smaller number of students in the
Link group than in the control group failed four or more items.
Students trained in the synthetic trainer exhibited greater pro-
ficiency in performing the maneuvers in the CAA Private Pilot Flight
22 SCHOOL LINK EVALUATION AND INSTRUCTION
Test than did the students who did not receive this training, but not
all of the students were trained to a level of proficiency that enabled
them to pass the test at the end of ten hours of flying. However, the
results indicate that the time required to train students to pass this test
may be substantially reduced from the thirty-five hours now required.
The analysis of the data showed that ten hours of flight training
were not sufficient to train all students to private pilot proficiency.
However, from these data and from observations during the course
of this experiment, it is believed that, with further revisions in this
type of training program, the goal of training students to private pilot
proficiency in ten hours may be attained. This belief is predicated on
the following observations:
1. At the completion of the experiment, it was evident that addi-
tional supervision of the instructors giving the Link training would
have been desirable. The inadequacy of supervision resulted in some
instructors' omitting certain parts of the Link training syllabus, par-
ticularly those maneuvers requiring the coordinated use of the black-
board and the trainer.
2. It is postulated that, if the instructors who were used in this
experiment had received more training on the syllabus and in methods
of using the trainer, the general performance of the Link group would
have been improved.
3. The time factor in the scheduling of the flight checks prevented
the full utilization of the Link trainer in the latter part of the course.
The Link periods that were omitted would have been devoted to prac-
ticing the complete flight test in the trainer.
4. Certain modifications in the mechanisms of the trainers used
in this experiment resulted in excessive malfunctionings and break-
downs. This caused frequent disruptions of the training schedule and
prohibited optimal use of the training time. Further, it seemed evident
that these malfunctions caused a decrease in confidence on the part
of some students and instructors in the value of the School Link and
a i onsequent decrease in motivation.
During the course of the experiment, problems arose which indi-
cated the need for further research. Some of these follow:
1. There are many possible ways of using synthetic trainers as aids
THE UNIVERSITY OF ILLINOIS 23
in the learning of specific and general flight tasks. On the basis of our
past experience with this trainer, certain temporary and experimental
methods of utilizing the trainer seemed desirable and were incorpo-
rated into the Link training syllabus. However, the fact that these
may not be the optimal methods is indicated by an item analysis of the
scores made by both groups of students in performing the maneuvers
of the flight test. These scores show that the Link group did better on
some maneuvers, while on others there is little or no difference between
the groups. It may be assumed that any one, or a combination, of the
following reasons could account for the low transfer of training that
occurred in the learning of these maneuvers:
a. There was insufficient trainer practice given in certain ma-
b . Inadequate methods were used in teaching these maneuvers.
c . The transfer of training on certain maneuvers was limited by the
Each of these possibilities should be investigated in relation to the
maneuvers that showed low transfer from practice in the trainer to the
performance of the maneuver in the aircraft.
2. The results of "intellectualizing" a maneuver, as referred to
here, include the ability to verbalize the important cues, their source,
the correct responses to the cues, their sequence, and the additional
cues arising as an immediate result of these responses. If this is done,
the student should know when to respond and how to respond. He
should know whether his response was appropriate, that is, whether
it was too soon or too late, too fast or too slow, too much or too little,
coordinated or uncoordinated, and so forth. Student performance in this
and previous experiments indicated that intellectualizing a maneuver
before attempting it in the air reduced the time necessary to learn it
in the air and, also, that additional time could be saved in the air if
the intellectual knowledge of the maneuver had been previously asso-
ciated with the psycho-motor response via practice in the synthetic
trainer. This indicates that the trainer will be more efficient as a syn-
thetic flight-instruction device if the cues and required responses in
the correct performance of a maneuver in the trainer are similar to
those in actual flight. Again, the contribution of these factors to the
efficient use of synthetic flight trainers is a matter which will require
3. To date, there is no evidence as to the level of proficiency a
student should attain in the trainer to give a maximum transfer of this
24 SCHOOL LINK EVALUATION AND INSTRUCTION
training to the performance of the same maneuver in the aircraft.
The importance of this information in designing an efficient syllabus
warrants additional investigation.
4. The problem of determining the optimal time interval between
synthetic training and flight training still exists. Experimental findings
are needed in this area for practical scheduling of synthetic training
in a flight syllabus.
5. Other factors which are worthy of experimental investigation are
the motivation of the students while they operate the synthetic trainer,
the effect of their knowledge of results, and the effect of freedom from
distracting and extraneous stimuli.
6. It is expected that when certain functional faults of the School
Link are eliminated — redesign is now in process by the manufacturer
— the trainer's utility as an aid in the teaching of primary flying will
be increased. As a result of this experiment the following recommenda-
tions for modification were made:
a. The functioning of the rudder should be changed so that
"weathercocking" will occur if the rudder is used without ailer-
ons when making a turn.
b. A device should be installed by means of which the instructor
can determine the adequacy of the student's coordination.
c. A 360° cyclorama for the School Link should be designed which
will provide a smooth and fixed horizon line equidistant in all
directions from the trainer.
d. The trainer should be redesigned to function consistently with
students weighing up to 200 pounds.
e. The trainer should be improved to withstand prolonged periods
of operation with a minimum of maintenance.
THE UNIVERSITY OF ILLINOIS
C IO f^ ^ i*N *-
IT; IT; IT) Tt lO lO C<"
) i— 1
o r 3
lO OO (N fO N (N O -
^ ^ ^ lO i/) lO CN
■^ rO lO fO (N (N
to o io o o c
^ cti u,
lO (^ ^ •* (N C
t^ © © ~ ^ ^~
■r-t cn cn cn cn cn
O O O O O C
"5 fO fO f^ f^ f
© © © ^ © c
to © io © © c
fN O ^ " >0 "
t^' © On © © c
© to © © © ©
i5 ^ J -
IT) CO CM *— I © C*
r o •* <
On On © <^h ^ C
,— 1 ^H ^H ,—
© to © © © ©
m <h ro O O C
t^ i>- f^ oo do t^
© © © © © c
(N CN i^ h O (^
(N (N (N ^ r^ rr
SCHOOL LINK EVALUATION AND INSTRUCTION
■r- on co
On lo NO
-H N NO
CO CS o
LO NO NO
•— 1 NO O
* * *
IU 11 D
<V QJ <U
£ P g
o o o
u a o
c c c
i — I
X On o
lo On CM
^-H ^— 1
a> a; aj
a> a; a;
a a a
S E S
o o o
u o o
C G C
i— i i— i i— i
.3 rt u,
** NO ON
,° •£ <
X t^ »H
lo r^ On
NO On X
co co *-H
no r- X
THE UNIVERSITY OF ILLINOIS
Analysis of Instruction Periods for Control Group
Time to Solo
. . Discus-
Data not available from log books.
28 SCHOOL LINK EVALUATION AND INSTRUCTION
Some Principles of Learning Applied to the Flight Situation
The instructors were informed that in this experiment an attempt
would be made to reach the highest level of flying proficiency in the
allotted flight time, and that to attain this goal use would be made of
some of the applicable conditions which had been found to promote
efficient learning and performance.
In order to aid the instructors in conducting their part of the ex-
periment a set of generalized principles was drawn up for their use.
These principles, as they were presented to the instructors, follow:
A. Motivation. The instructor's first task is to maintain a high de-
gree of interest on the part of each subject and to secure cooperation from
him. One approach to securing cooperation is to point out to the subject
the rewards and advantages that will be available to him according to
the time he requires to develop proficiency. These rewards, depending
upon the hours saved, may include many items such as the following:
1. Additional solo cross-country flights.
2. A 1,000-mile cross-country flight in a four-place airplane.
3. A dual night cross-country flight.
4. A dual practice flight on "lost procedures."
5. Additional proficiency in the regular maneuvers required for a
private pilot rating.
6. Work on such advanced maneuvers as deadstick landings, short
field landings and take-offs, and crosswind wheel landings.
7. Check-out and solo in types of aircraft other than the type used
for primary training.
8. Check-out and night solo in the Aeronca.
9. Link experience in instrument and radio range flying.
10. Radio communications practice in the airplane, leading to a
restricted radio operator's permit.
11. An opportunity to visit the control tower, weather station, air
traffic control center, and a representative airline operation at the
Chicago Municipal Airport.
These opportunities apply only to those students enrolled for the
full thirty-live-hour course. The ten-hour group should be motivated
by an explanation of the pioneering nature of the experiment and by
the presentation of the challenge of reaching private pilot proficiency
in less than one-third the time normally required.
B. Knowledge of results. It is generally accepted that perform-
ance will improve more rapidly if there is a knowledge of the results.
THE UNIVERSITY OF ILLINOIS 29
For the students receiving Link training-, the instructor should give
readily understood and comparable scores on such items as straight and
level flight, torque correction, stalls, climbing and gliding turns, and
slow flight. This can be done by counting the number of times the
projected image on the screen is outside the arbitrarily set standards.
For the control group, knowledge of results will be the instructor's
evaluation of the student's performance given immediately after com-
pletion of each maneuver.
C. Anticipatory set. Experiments have shown that efficiency of
performance can be increased if the individual knows in advance exactly
what he is expected to do when a particular event occurs; that is, if a
set is developed, he will respond in a certain way when a given stimulus
occurs. An example of this in the flying situation is the experienced
pilot who is quite prepared to put the stick forward in a spin recovery
long before the strategic second arrives. It is possible for the Link sub-
jects to get practice in these sets via practice in the trainer. The control
group should also have the importance of these pointed out. It is ex-
pected that they will learn them to some extent by verbalizing, in
minute detail, the maneuvers to be practiced in the flight lesson.
D. Part versus whole learning. When the task to be learned is
simple and relatively short, the most efficient learning method is prob-
ably that of learning the task as a whole. When the task is complex
and relatively long, the most efficient learning may take place if the
task is divided into smaller units and these units are learned first and
then combined. It is possible to divide any maneuver into its component
parts and to practice these parts as such in the Link and in the air.
For example, in a turn the task may be broken down into an entry,
maintaining altitude and bank in the turn, a roll out of the turn to
straight and level flight, and directional control.
E. "Supervised practice" method. By the "supervised practice"
method the student performs the task as he receives verbal instructions
from the instructor. This method differs from the much-used "follow
me through" technique by which the student is introduced to the
maneuver by the instructor who performs the maneuver while the stu-
dent attends by putting his hands and feet on the controls. When this
technique is used, the student fails to get the kinesthetic cues, or the
"feel," that would result from his own manipulation of the controls.
After following through, the student attempts to perform the maneuver
himself. At the completion of the trial, the instructor tells him of his
errors. This results in a trial and error type of practice which is a
relatively inefficient method of learning.
30 SCHOOL LINK EVALUATION AND INSTRUCTION
The "talk through" method provides a situation in which the stu-
dent is able to associate the correct response with the various stimuli
or cues which bring about the correct reaction of the aircraft. For
example, during the performance of a turn the instructor will point
out the cue, such as nose down, and tell the student which maneuver
will correct the condition. Hence the student learns what the im-
portant cues are and associates them immediately with the correct
F. Overlearning. Additional practice of a task after the task has
been learned to criterion will usually increase the subject's retention
of the task and his facility in performing it under different conditions.
This principle does not lend itself to economical use of the actual
flight period, but it can be utilized in the School Link because of its
relatively low operating cost. The principle can be applied to the con-
trol group by additional verbalizing of the maneuvers.
G. Spaced practice versus massed practice. There is some evi-
dence from experiments in the use of synthetic flight trainers that
spaced practice in learning a flight maneuver is more economical than
continued or massed practice. In addition, most of the evidence from
other types of experiments demonstrates the same theory. Conse-
quently, the syllabus for this experiment was planned to provide a
series of short practice sessions on individual maneuvers.
H. Intellectual knowledge of maneuvers. The two groups must
be thoroughly briefed on the flight maneuvers, and both written and
oral information should be used. Briefing should consist of a detailed
explanation of each technique or maneuver to be presented or practiced
in the next flight. The flight lesson should be given to the student when
he can verbalize accurately and completely these techniques, procedures,
and maneuvers. After this verbalization or "intellectualization," the
control group should receive the lesson in the aircraft and the Link
group in the synthetic trainer. The Link students should be instructed
in the trainer until, in the instructor's opinion, they have reached a
satisfactory level of proficiency. When this proficiency is attained in
the desired number of maneuvers, the student may attempt them in
Immediately following each air flight, both the Link and the control
groups should have a discussion period in which the flight is discussed
in detail. The short flight periods, which should be employed especially
during the pre-solo stage, should aid both the instructor and the student
in remembering nearly every feature of the lesson and thus enhance
l lie value of the post-flight discussion periods.
THE UNIVERSITY OF ILLINOIS 31
Methods Used in Teaching Individual Maneuvers
in the School Link
The following is a description of the methods employed in teaching
the various maneuvers in the School Link.
Prefiight. The instructor required the Link students to preflight
the School Link, using the same procedure followed in prerlighting the
Aeronca. When a particular part which was regularly inspected on the
airplane, such as the landing gear assembly, was not on the trainer, the
student verbalized the check procedure as he would in checking that
component on the aircraft. This was done prior to each Link flight
until each student had reached a satisfactory level of proficiency.
Starting. In starting the School Link the students employed the
same procedure used in starting the Aeronca. The instructor, after
calling out the appropriate commands, simulated the swinging of the
propeller. This procedure was continued until the student reached a
standard level of proficiency.
Taxiing. The Link students practiced taxiing procedures in the
trainer by leaving the lateral and longitudinal locks in place and
releasing only the pedestal lock. The trainer then responded to rudder
pressure only. By calling out winds from various directions, the in-
structor drilled the Link student in the proper use of the ailerons and
elevator to correct for wind from any desired relative direction.
Run-up check. The instructor taught run-up checks by having
the student go through the same procedure in the Link as required in
the Aeronca. The physical limitations, such as the single "on-off"
magneto switch in the Link instead of the "on-left-right-off" switch
in the Aeronca, were partially overcome by having the student verbalize
the whole procedure, item by item, while accomplishing the check.
Steep turns. The instructor taught steep turns by first instruct-
ing the student in the proper procedure to use in rolling into and out
of the turn. This involved learning the correct use of the throttle as
well as noting the large amount of back pressure necessary in the
steeper turns. Also included in this phase was instruction in decreas-
ing the throttle and applying forward pressure on the controls when
rolling out of the turns. When the student reached satisfactory pro-
ficiency in rolling into and out of turns, he was instructed in the
complete 720° turn. In this practice the student utilized the trainer
altimeter as his altitude reference. Because of the physical limitations
32 SCHOOL LINK EVALUATION AND INSTRUCTION
of the trainer, the actual angle of bank in a steep turn could not be
simulated, but the student could practice certain procedures required
in such a turn.
After proficiency in maintaining orientation and altitude had been
attained, the student was told to disregard the trainer altimeter and to
make his attitude corrections from the cues given him by the instructor.
For example, if the instructor told him he was either high or low, the
student was to use the controls to effect the proper attitude response.
Stalls, spins, climbs, glides, climbing turns, gliding turns, me-
dium turns, shallow turns, and straight and level flight. All of these
maneuvers were taught in the Link in essentially the same manner
employed in the airplane. The instructor checked coordination by
watching the student's movements of the controls and the resulting
control surface motion. By standing beside the trainer, the instructor
also observed the instrument readings. He used the cues from the
trainer attitude and the instruments to instruct the student in the
maneuvers just as he would in the aircraft.
Take-off. In order to facilitate instruction in take-offs, a rough
diagram of the runway was drawn on a blackboard which was placed
directly behind the trainer. After take-off, the student was instructed
to look behind him and note the position of the runway to determine
whether or not he appeared to be maintaining a straight path off the
end of it. Because the Link was not designed to simulate the proper
airspeed readings on a take-off, the student was told to disregard the
readings of the instrument and take his cues of airspeed from the
The instructor simulated drift by moving the board to one side. If
the student noticed drift, he was to institute the proper corrections.
The instructor would then slide the board back to the proper position
so the student could get the conception of drift correction on take-off.
The student learned actual take-off procedure by easing the trainer
up to the three-point attitude while the lateral straps were locked and
then giving it full power to start the take-off run. The instructor
supplied extra torque and weathercocking effects of crosswinds by
pushing against the side of the trainer. When the trainer reached
flying speed, a matter of about thirteen seconds depending on the
simulated wind conditions, the instructor would unlock the lateral
straps, thus completely freeing the trainer and simulating leaving the
ground. The student would then proceed with the take-off, making the
power and attitude adjustments and looking behind at the runway to
< heck for drift.
THE UNIVERSITY OF ILLINOIS 33
Taxi and traffic patterns. As an aid in giving instruction in
taxiing to the take-off position, making the take-off, and flying the
traffic pattern, a constant scale drawing of the airport, its runways,
and taxi strips was made on the portable blackboard which was placed
close to the trainer.
As the student went through the procedure of taxiing from the
ramp to the take-off point, the instructor indicated the course the
student was following by tracing the course on the board. When
the take-off point was reached, the student went through the simulated
cockpit check, made his clearing turn, and proceeded with the take-off
as previously described.
The instructor kept the chalk line moving in the proper direction,
varying the trace to show drift and to indicate the correction of it as the
student corrected the heading of the trainer. The student was able to
watch his progress throughout the traffic pattern as the instructor
moved the chalk to keep pace with the position of the trainer in relation
to the runway.
For practice in entering the traffic pattern, a small wind-tee was
drawn in the proper position on the field layout on the blackboard; and
the student was started at a scaled distance of one to five miles from
the field. The distance depended upon the particular problem. Then the
student determined the wind direction by checking the tee, planned his
pattern, and proceeded to maneuver the trainer so that he arrived at the
proper entry point at the proper altitude. The instructor could check
the student's altitude in the pattern, either by looking in the cockpit
when the trainer was facing the proper direction or by having the
student call out the indicated altitude. If the student had difficulty in
any part of the pattern, such as in planning the entry, using carburetor
heat, placing the downwind leg or base leg, or in planning the turn onto
the final approach, that particular part could be singled out and given
concentrated practice. The practice continued until the student had
mastered the particular part sufficiently to incorporate it correctly in
the over-all pattern.
Pylon eights. The instructor prepared for teaching eights by
drawing on the blackboard a grid pattern which resembled section
lines of the local terrain. He explained the scale of the rectangles and
squares so that the student could pick his pylons at the recommended
one-half to three-quarters of a mile apart. The student then maneu-
vered the trainer into position so that he was flying directly crosswind
at the proper altitude for eights.
The instructor continuously plotted the student's progress in the
34 SCHOOL LINK EVALUATION AND INSTRUCTION
same manner as he used for the other ground-contact maneuvers. The
student would look upwind from the trainer's position and pick out and
identify his pylons. Intersections of the grid lines were most fre-
quently used because they corresponded more closely to what was
available to the student in the actual flight. The student maneuvered
the trainer so that he would pass on the downwind side of the pylons
on a line parallel to the two pylons and about one-quarter of a mile
from them. When he was opposite the second pylon, he would start his
turn and proceed with the pattern. The student observed the path of
the chalk in relation to the path he desired, and, if any deviation
occurred, he would institute the proper correction which would then
be reflected by the corresponding change the instructor would make
in the direction of the trace.
Forced landings. The blackboard was first laid off in rectangular
patterns resembling the local terrain, and certain portions were desig-
nated as being of suitable surface to permit a forced landing. The
instructor moved the chalk line across this area at a relative rate of
speed and in a direction corresponding with the speed and direction
of the trainer. At some point in the area, he would declare an emer-
gency landing. The student, taking into consideration his altitude,
direction of wind, availability of fields, and other pertinent factors,
would plan his approach. The instructor would continue to move the
chalk with a speed and in a direction corresponding to those of the
trainer while estimating and indicating the effect of the wind. The
student would maneuver the trainer in appropriate patterns in order
to arrive at the key position for his field at the proper altitude as indi-
cated by the trainer's altimeter. If he reached the key position at an
improper altitude, he was to recognize the fact and start the proper
correction procedure immediately. For example, if he reached his key
position at an altitude that was too high, he would immediately start
to alter his base leg to take him away from the field, allowing his
extra altitude to be dissipated before turning on the final approach.
Spot landings. The airport runway system was first drawn on
the blackboard and then the spots for landing were drawn at appro-
priate places. The instructor would have the student start the ma-
neuver on the downwind leg at the proper position and altitude. To aid
the student in estimating the placement of the base leg, the wind
velocity for the problem was indicated before each attempted spot
landing. The student would go through the proper procedure of using
carburetor heat, closing the throttle, establishing the glide, planning the
turns, clearing the engine, and making the proper corrections for drift.
THE UNIVERSITY OF ILLINOIS 35
By moving the chalk on the blackboard, the instructor would apply
the effect of wind and keep track of the student's position.
Practicing spot landings in this manner was aimed at teaching the
student to recognize the need for correction and what correction should
be made. It was also brought to his attention that he could maintain a
consistent pattern by allowing only one component of the maneuver to
vary at a time, that is, the placement of the base leg. The altimeter in
the trainer gave the only altitude cue available for spot landings. The
student attempted to maneuver the trainer so that when the chalk mark
arrived at the landing spots on the "blackboard runway" the altimeter
would register zero altitude. To aid him in making spot landings,
several methods of regulating the glide path were taught in the trainer.
These included making slight changes in the airspeed above and below
the normal gliding speed and in making slips.
Landings. Landings were also simulated with the help of the
blackboard. A rough perspective diagram of the runway was drawn
on the board. By rotating the board about its horizontal axis, a reason-
able simulation of the appearance of the runway during the final
approach and landing could be produced. Thus the student was taught
to respond to visual cues similar to those he would receive during an
The method used was based on the fact that the closer the plane is to
the ground, the more nearly horizontal the runway appears from the
pilot's viewpoint. The instructor, starting the board at about a 45°
angle, would gradually reduce the angle to simulate the plane's ap-
proach to the runway. When the blackboard was horizontal the trainer
appeared to be at ground level. The rate of angular change of the
board was varied by the instructor to simulate a variety of conditions.
By bringing the board down to a horizontal attitude before the student
attained the three-point attitude with the trainer, a simulation of flying
into the ground was obtained.
When a simulated bounce was desired, the angle of the board was
quickly increased a few degrees to give the appearance of leaving the
runway. As the student initiated the correction to ease the trainer back
to the runway, the board was brought back to the horizontal position.
By using other similar methods, nearly all conditions that the student
could be expected to encounter on the final approach and landing were
simulated. Because only the last one hundred feet of descent were
simulated, taking about thirteen seconds in the Aeronca, a large
number of landings could be practiced in a short period of time.
36 SCHOOL LINK EVALUATION AND INSTRUCTION
Each instructor was given a copy of a guide which contained the
following rules to promote uniformity in carrying out the experiment:
1. To the very best of his ability, the instructor should give the
same instruction to both experimental and control students.
2. Accurate entries should be made in the logbooks for all students
on the following items: Aeronca flight time, Link flight time, discus-
sion time on the ground, and the number of landings made in the
Aeronca (by either instructor or student) during the pre-solo stage.
3. Both groups of students should be able to verbalize a maneuver
correctly and in minute detail before attempting the maneuver in the
Link or in the Aeronca. Part of the requirements for all students
should be to draw on the blackboard any maneuver which can be so
transcribed. The instructor may supplement these requirements by
asking pertinent questions on the maneuvers.
4. During an instruction flight in the Aeronca, the instructor should
refrain from touching the controls except in extreme cases. He should
rely more on talking the student through an adequate performance.
5. Although the time required for a student to solo is not a major
measure in this experiment, it is, nevertheless, of some significance;
therefore, part of the instructor's task is to solo his students as soon as
it is possible with safety. The usual criteria for solo include practice
on all pre-solo maneuvers to reach a safe level of proficiency, adequate
performance while flying the traffic pattern, three consecutive safe
landings, and the ability to demonstrate positive recovery from bounce
landings and from leveling off too high during a landing.
6. In general, the total solo time allowed the student prior to the
flight test should be not more than thirty minutes. Additional solo time
may be advantageous to those students lacking in self-confidence and
may be used at the discretion of the instructor.
7. The syllabus in this experiment should be considered as a guide
and not as a schedule to be followed rigidly. However, the maneuvers
arc to be given to all students in the same sequence and the student
is to progress at his own best rate.
X. The criteria used to determine when the student has had optimal
practice in the Link on a certain maneuver or exercise remain largely
a matter of the instructor's judgment. Some indications an instructor
mighl look for when making this judgment of the student include,
THE UNIVERSITY OF ILLINOIS 37
first, consistent correct reactions in the performance of a maneuver;
second, ability to anticipate all details or events of a maneuver; third,
corrections for bank, altitude, and so forth, made rapidly enough that
the required correction is a small one; fourth, knowledge of where to
look for pertinent cues; and finally, adaptation of as much of a
maneuver as possible to habitual type responses.
One method of determining this latter criterion is to distract the
student's attention while he is performing a maneuver and note his
reactions on the controls. By using the projector light on the Link,
scores for proficiency in the following maneuvers can be obtained:
straight and level flight, climbs, glides, stalls, turns, climbing turns,
gliding turns, and slow flight.
9. The time normally required for discussion can be reduced
through the use of prepared student lesson plans. The student will be
given the lesson plan prior to his next scheduled period. He will be
required to study this material and repeat it to the instructor at their
10. The student should be told before his first flight and prior to
each flight thereafter that if he experiences the first signs of airsick-
ness he is to tell his instructor immediately. The instructor will allow
the student to fly back to the Airport at once, expediting the traffic
pattern. It is important for the instructor to end the period as soon as
possible. A student who is airsick will not be receptive to instruction.
11. Students of both groups may sit in an Aeronca when they have
a few spare minutes at the end of a nonflight period. This time in the
cockpit need not be recorded, but the student should be impressed with
its importance. He should be told that the reason for spending this time
is to make him feel at home in the Aeronca and by looking out of it
from every possible angle to familiarize himself with the three-point
attitude. Familiarity with the three-point attitude is vital for making
early solo possible, since a landing will seldom be dangerous if this
attitude is never exceeded.
12. The instructor should remember that if he speaks to the stu-
dent in other than a friendly manner or a well-modulated voice, regard-
less of whether or not the student is in the Link or the Aeronca, the
student is apt to become tense, and his receptiveness to instruction may
13. When flying the Link or the Aeronca, the student should learn
to divide his attention continually between the performance of his
trainer or plane and his surroundings. There should be a definite
procedure including where to look and what to look for.
38 SCHOOL LINK EVALUATION AND INSTRUCTION
14. Flight time will be recorded from the take-off until the airplane
is returned to the parking area and the switch turned off.
15. The instructor will find that any maneuver or pattern shown
in the private pilot syllabus can be analyzed into very small component
parts. If each component part is looked upon as a separate procedure,
the student can learn much of how to fly before he ever attempts the
maneuver or pattern in the airplane.
16. It is very doubtful that the student can learn much "control
feel" when he is tense. During the first few flights the instructor should
pay particular attention to keeping the student relaxed.
17. Before every flight period the instructor should be sure that
the weather conditions will not hinder the learning of the maneuver
or patterns scheduled for that flight, that the intercommunication
equipment is operating satisfactorily, that both instructor and student
feel up to the flight period, and that the student knows intellectually
what he is going to do, how it is to be done, and why.
18. The instructor should check to see that his student, either in
the Link or the Aeronca, has as many seat cushions as are necessary
for him to have good visibility and as will enable him to maintain a
constant relationship to the reference points in each flight.
THE UNIVERSITY OF ILLINOIS 39
Private Pilot Flight Test
Over-all Grades for Link Group
T ♦ *
40 SCHOOL LINK EVALUATION AND INSTRUCTION
Private Pilot Flight Test
Over-all Grades for Control Group
THE UNIVERSITY OF ILLINOIS 41
Private Pilot Flight Test
Over-all Grades for Experienced Private Pilot Group
Pilot Flying Hours
CAA Examiner Private Pilot Examiner
1 80 68 68
2 80 69 65
3 80 80 66
4 102 65 69
5 95 80 87
6 120 70 80
7 170 68 80
SCHOOL LINK EVALUATION AND INSTRUCTION
Item Analysis of Flight Grades
12 3 4 5 NG Total 12 3 4 5 NG Total
1. Equipment exam
(oral) 1 15 5 21 17 3 1 21
2. Preflight check... 1 19 1 21 18 3 21
3. Taxiing 20 1 21 1 18 2 21
4. Run-up check.... 1 19 1 21 18 3 21
5. Take-off 13 7 1 21 11 7 3 21
6. Climbs and climb-
ing turns 10 9 1 1 21 4 11 5 1 21
7. 720° steep turn... 4 10 7 21 1 2 6 11 1 21
8. Stalls 5 13 3 21 2 5 14 21
(slow flight) 15 3 3 21 6 12 2 1 21
*10. Spins Not graded Not graded
11. Glides and gliding
turns 005 14 11 21 005952 21
12. Pylon eights 2 9 9 1 21 1 9 10 1 21
13. Emergencies 005952 21 0025 14 21
14. Slips 007950 21 003 11 61 21
15. Accuracy landings 004115 1 21 00 195 6 21
16. Landings 007617 21 005637 21
17. Judgment 008922 21 001 11 63 21
18. Smoothness and
coordination 4 10 7 21 2 7 11 1 21
Total 2 1162 127 50 15 357 2 116 119 95 25 357
By per cent 06 002 45 36 14 04 06 32 33 27 7
I'm to recent changes in the CAA requirements for a private pilot certificate, students were
• 'I on Spins.
THE UNIVERSITY OF ILLINOIS
Item Analysis of Flight Grades
Private Pilot Examiners
12 3 4 5 NG Total 1 2 3 4 5 NG Total
1. Equipment exam
(oral) 18 8 10 3
2. Preflight check.. . . 6 13 2
3. Taxiing 6 14 1
4. Run-up check 1 5 14 10
5. Take-off 5 14 2
6. Climbs and climb-
ing turns 11 9 1
7. 720° steep turns.. . 2 5 8 6
8. Stalls 1 7 12 1
(slow flight) 14 6 1
*10. Spins Not graded
11. Glides and gliding
turns 15 5 1
12. Pylon eights 1 6 5 8 1
13. Emergencies 3 3 11 3 1
14. Slips 3 3 11 2 2
15. Accuracy landings 1 3 116
16. Landings 1 2 1 17
17. Judgment 9 8 2 2
18. Smoothness and
coordination 00 3 8 8 2
Total 2 41 143 91 48 32
By per cent 06 11 40 25 13 09
11 2 7 10
9 57 101 75 83 32 357
3 16 28 21 23 9
SCHOOL LINK EVALUATION AND INSTRUCTION
Comparative Analysis of Flight Test Grades
by Number of Items Failed
^ ^ AA Pilot Total
^ ^ AA Pilot Total
exam (oral) ....
2. Preflight check
4. Run-up check.
6. Climbs and
7. 720° steep turns
7 6 13
3 1 4
1 1 2
9 8 17
5 3 8
5 2 7
5 16 21
1 1 2
2 2 4
7 8 15
11 10 20
14 5 19
2 1 3
5 2 7
10 12 22
14 11 25
6 7 13
4 19 23
4 3 7
6 3 9
11 10 21
(slow flight) . . .
11. Glides and
gliding turns. . .
12. Pylon eights. . .
13. Emergencies. . .
15. Accuracy land-
50 48 98
95 S3 177
UNIVERSITY OF ILLINOIS URBANA
3 0112 005630444