U. &, ASHt AIR FOPXSd SBUCTXON AJflj CIAic-XFICATIOK TVS + + + + + + + + + + + -f + + + + + + + + + 4- -r STANINES + + -f + + + + + -»- + + + + + + + + + + + + H fELiCfi»Kv:v:v:v + + + + + + + + + + + + + + + + + + + - + + + + + + + + + + + + + + + -t- + + + 4-4- + + + + + + + + + + + + + + + + + + + + + - + + + + + + + + + + + + + + + + + + + + + + +_-U + + + + + + + + + + + + + + + ■ +iw VB + + + + + + + + + + + + + + + + + AM +■ m + + + + + + + + + + + + + + + - ■r + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + - + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +»+ + + + H € i AS S IFIC A TOO N + + + + + + + + + + + + + + + + + + + + + H + + + + + + + + + + + + + + + + + + + + + + + + + 4- + + + + + + + + + + + + + + + + H + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + H In + + + + + + + + + + + + + + + + + Jh~ + + + + + + + 4- + 4- + -f + + + H + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + H + + + + + + + + + + + + + + + + + + + + + H + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + H + + + + + + + + + + + + + + + II +■ -I *■ v mf + + + + + + + + + + + + + + H IB ■ J+ m -M + + + + + + + + + + + + + + + +■ + + + + + + + + + + + + + h REPORTED BY THE AVIATION PSYCHOLOGY PROGRAM OFFICE OF THE AIR SURGEON HE ADQU AIBJQBCES During the five year period from 1941 to 1946, a large group of professionally trained psychologists par- ticipated in a program of coordinated research known as the Aviation Psychology Program of the Army Air Forces. By the end of the war, the staff had grown to include over a thousand persons approximately half of whom had received some professional training in psychology beyond the baccalaureate degree. Some of the problems assigned to this group and some of the more important results achieved are presented and illustrated in the pages that follow. SELECTION AAF QUALIFYING EXAMINATION From 1923 to 1939, the Army Air Forces produced 3505 rated pilots/no rated navigators and no rated bombardiers. It had long been generally recognized that aircrew duties place unique demands upon the human body and mind and consequently only the most fit could be expected to succeed. From 1927 to 1941 appli- cants were screened by requiring at least two years of college i training, a rigid physical examination, and a comprehensive in- terview by an experienced Flight Surgeon or Aviation Medical Examiner. When the demand for aircrew rose sharply because of the threatening international situation it became impossible to con- tinue to fill the new quotas. To meet the needs of the expanding Air Forces, the Aviation Psychology Program developed tests of aptitude, Interest, and proficiency which could be used to screen men regardless of their previous training. With the elimination of the college requirement, the possi- bility of entering aircrew training in the Amy Air Forces was open to thousands of young men, who now applied in great numbers to the Aviation Cadet Boards. An initial screening device was required that could be administered to a large group at one time, without in- volving extensive preparations. For this purpose the AAF Quali- fying Examination, a paper and pencil test, was devised. This test, along with all the others in the program, was frequently modified and improved as data on its predictive efficiency were accumu- mulated. The passing score was also modified from time to time to meet the needs of the moment. Well over a million young men took the examination. About 650,000 succeeded in passing. The effectiveness of the Qualifying Examination was demon strated in an experimental group of approximately 1000 appli- cants who were permitted to enter training regardless of what scores they obtained on the selection tests. MEN WHO RECEIVE FAILING SCORES ON THE AAF QUALIFYING EXAMINATION ARE POOR PROSPECTS FOR PILOT TRAINING EXPERIMENTAL GROUP SUCCESS IN COMPLETE FLYING TRAINING CLASSIFICATION AIRCREW CLASSIFICATION BATTERY The different aircrew positions require different skills as well as training. A man who is careful, accurate, and mathe matically Inclined might make an excellent navigator but fail to make the grade as a pilot. The time and equipment involved in training would be wasted and a potentially excellent naviga- tor turned away. Similarly the athletic, well-coordinated and energetic candidate who never got along well with his math in- structor might fall miserably in navigator or bombardier train- ing but would make an ace pilot if properly assigned. It was clear that an instrument was needed that could measure the aptitude of the candidates who had passed the In- itial screening in order to assign them to best advantage. The Aircrew Classification Battery was designed to meet this need. The battery Included paper and pencil tests that could be administered to several hundred men at one time and appa- ratus tests that required individual attention for each candi- date, standardized conditions3 and very carefully trained testers. Some of the functions measured by these tests were numerical ability, mechanical comprehension, general informa- tion about flying, the ability to read dials and tables, mathe- matical reasoning, reading comprehension, muscular coordination and perceptual speed. For the purpose of predicting each'man’s chances of suc- cess for each kind of aircrew training, separate aptitude scores, called stanlnes, ranging from 9 (high) to 1 (low), for bombardier navigator, aad pilot were computed for each candidate from his performance on the Aircrew Classification Battery. The follow- ing three charts show the degree to which each of the stanlnes was able to predict success in their respective types of train- ing. PAPER AND PENCIL TEST APPARATUS TEST SELECTION FOR PILOT TRAINING In this chart can be seen how useful an effective se- lection battery can be. If a candidate had a pilot stanlne of 9, he was a good Investment for pilot training. The chances were only 4 in 100 that he would fail in primary % pilot training. The candidate with a pilot stanine of 1 had the odds against him with 77 chances in 100 to fail early in pilot training. He was not a good investment. The higher the stanine, the more assured was the Investment. THE HIGHER THE PILOT STANINE THE GREATER THE CHANCES OF SUCCESS IN PRIMARY PILOT TRAINING THE BARS INDICATE THE PROPORTIONS ELIMINATED AT EACH PILOT STANINE. ELIMINATION NAS FOR FLYING DEFICIENCY. FEAR AND OWN REQUEST. FLYING EXPERIENCE CREDIT IS INCLUDED IN THE STANINE SCORE. THE DATA ARE FROM CLASSES 43-F THROUGH 45-H. MEN WITH LOW STANINE SCORES ARE NOW DISQUALIFIED FOR TRAINING; MOST OF THE MEN WITH LOW STANINES INCLUDED IN THE CHART ENTERED PRIMARY SCHOOLS EARLY IN 1943. SELECTION FOR NAVIGATOR TRAINING A candidate with a low pilot stanlne might have had a high navigator stanine. This chart shows that a candi- date with a high navigator stanine had a much better chance of completing advanced navigator training than his colleague with a low navigator stanine. THE HIGHER THE NAVIGATOR STANINE THE GREATER THE CHANCE OF SUCCESS IN ADVANCED NAVIGATOR TRAINING THE BARS INDICATE THE PROPORTIONS ELIMINATED AT EACH NAVIGATOR STAN1NE. ELIMINATION WAS FOR FLYING DEFICIENCY, FEAR, M OWN REQUEST. ONLY NEW AVIATION CADETS ARE INCLUDED. THESE DATA ARE FROM CLASSES 43-12, 43-13, 43-14. 43-17, 43-18, AND 44-1 THROUGH 45-13. SELECTION FOR BOMBARDIER TRAINING The last of these three charts reveals that effective prediction for adaptability to bombardier training was ob- tained from the bombardier stanlne. THE HIGHER THE BOMBARDIER STANINE THE GREATER THE CHANCE OF SUCCESS IN BOMBARDIER TRAINING THE BARS INDICATE THE PROPORTIONS ELIMINATED AT EACH BOMBARDIER STAN 1 HE. ELIMINATION WAS FOR UNSATISFACTORY PROCESS, FEAR AND OWN REQUEST. ONLY NEW AVIATION CADETS ARE INCLUDED. THESE DATA ARE FROM CLASSES 43*8 THROUGH 43-18. SELECTION STANDARDS As the data on the predictive efficiency of the Classi- fication Battery were accumulated, it became evident that the odds were against the men with low stanlnes. Such men are now disqualified for training in the aircrew positions for which they show little aptitude. June 1945 standards required a sta- nine of 7 or above to qualify for bombardier, navigator, or pi- lot training. Thus, since there were more candidates for air- crew duty than were needed or could be trained, the Aircrew Classification Battery was not only used to classify men for training in the different aircrew positions, but also to weed out the poorest investments. The disqualified men were not wasted. They went into types of training in which their chances of success were better and the probabilities of a good return on the investment were greater. They’became ground officers, gunners, mechanics and radio operators. A good mechanic was a greater asset to him- self and to the Air Forces than an eliminated pilot. PSYCHOLOGICAL CHARACTERISTICS OF SUCCESSFUL AIRCREW CANDIDATES APTITUDES AND ABILITIES How does it happen that a combined score (stanlne) de- rived from a group of tests administered in eight hours could predict the outcome of many months of complicated training, climaxed by graduation or elimination, a long time after the testing had taken place? To answer this question, we must turn the clock back to July 1941 when the Aviation Psychology Program was first hand- ed the problem of selecting the best possible candidates for aircrew training. It was decided at the outset that the first step the groundwork for efficient predictions, was to find out what had to be predicted. What does a cadet do in pilot train- ing and what kinds of abilities or aptitudes must he have to do it well? Psychologists, working with specialists in pilot training, drew up a tentative list of aptitudes. Then tests were constructed, designed to measure these aptitudes. That was only the beginning. Now xi remained to be determined which of the tests worked. They were administered to large groups of cadets before they received any training. When these men were graduated the test results were compared with their training records. If high scores were related to success in training and low scores to elimination, the test satisfied scientific standards. Such tests were Incorporated into a battery. Poor predictors were discarded. The process of research, test development and construction was continuous. New and better tests were frequently added to the original battery or substi tuted for less effective ones, thus constantly increasing the predictive efficiency of the stanines. These charts present some of the most important require- ments for success as pilots and navigators, respectively, and show how some of the individual tests that measure them can predict success or failure in training. If the tests are good the battery will be better. SUCCESSFUL PILOTS ARE LIKELY TO BE APPROPRIATE IN BACKGROUND QUICK IN PERCEIVING AND REACTING INTERESTED IN & INFORMED ABOUT AVIATION WEI L COORD!NA TED HAVE SUPERIOR ABILITY TO.... DISCRIMINATE RETWEEN VISUAL OBJECTS VISUALIZE MECHANICAL MOVEMENTS EACH BAR REPRESENTS THE PERCENT OF PILOTS ELIMINATED IN CLASS 44E THROUGH ADVANCED TRAINING AT EACH SCORING LEVEL FOR EACH AT THE TESTS REPRESENTED. TOTAL NO. 7164, NO. EL1MINATED-1585. CADETS WITH PREVIOUS FLYING EXPERIENCE ARE NOT INCLUDED. ELIMINATION IS FOR FLYING DEFICIENCY. FEAR OR OWN REQUEST THESE TEST ARE UTILIZED AS BEING LARGELY REPRESENTATIVE OF THE FACTORS FOUND TO Bt IMPORTANT FOR PILOTS THEY DO NOT 1NCLUDETHE WHOLE BATTERY WEIGHTED FOR P1LOTSNOR ARE THEY PURE MEASURES OF THE FACTORS INDICATED. SUCCESSFUL NAVIGATORS ARE LIKELY TO BE GOOD AT ARITHMETIC ACCURATE /N USING TABLES £ INSTRUMENTS INTERESTED IN £ INFORMED ABOUT SCIENCE GOOD REASONERS scopes cvv recHH/CAL VOCABULARY - A/A V/GATOR PERCENT ELIMINATED HAVE SUPERIOR ABILITY TO.... USE MAPS AND PHOTOGRAPHS COMPREHEND VERBAL MATER/AL EACH BAR REPRESENTS THE PERCENT OF NAVIGATORS ELIMINATED IN CLASSES 43-10,11 AT EACH SCORING LEVEL FOR EACH OF THE TESTS REPRESENTED. TOTAL NO-811j NO EL1MINATED=113. ELIMINATION IS FOR UNSATISFACTORY PROGRESS. FEAR OR OWN REQUEST. THESE TEST ARE UTILIZED AS BEING LARGELY REPRESENTATIVE OF THE FACTORS FOUND T6 BE IMPORTANT FOR NAVIGATORS. THEY DO NOT INCLUDE THE WHOLE BATTERY WEIGHTED FOR NAVIGATORS, NOR ARE THEY PURE MEASURES OF THE FACTORS INDICATED. MOTIVATION It seemed reasonable to expect that a cadet who wanted to be a pilot would work harder, longer and with more enthu- siasm at the Job of becoming one than the man who preferred to be a navigator. Therefore, at the time the potential ca- det was tested, he was asked to indicate which aircrew posi- tion he preferred. The chart shows that the expectation was Justified. CADETS WHO WERE MOST SUCCESSFUL IN PRIMARY NAVIGATOR TRAINING PILOT TRAINING PREFERRED TO BE PREFERRED TO 3E P/LOTS NA V/GATORS THERE WAS NO CLEAR RELAT/ON BETWEEN SUCCESS AND PREFERENCE /N BOMBARD/ER TRA/N/NS FIRST PREFERENCE FOR P/LOT TRA/N/NS FIRST PREFERENCE FOR NAVIGATOR TRAINING FIRST PREFERENCE FOR BOMBARD/ER TRAINING THESE DATA ARE FROM PILOT CLASSES 43F, 43G, AND 43H. NA VI GAT 01 CLASSES 43-12 THROUGH 43-15, BOMBARDIER CLASSES 43-15 THROUGH 44-1 Another measure of motivation was the degree of interest felt in each of the aircrew positions Indicated by the cadet on a nine point scale. The chart shows that the more interested men tended to be more successful. These data Indicate that motivation needed to be con- sidered in classification, especially in cases where a cadet qualified for more than one type of training. CADETS WHO WERE MOST SUCCESSFUL IN PILOT AND NAVIGATOR TRAINING HAD A STRONG INTEREST IN THE TYPE OF TRAINING TO WHICH THEY WERE ASSIGNED STRENGTH OF PHOT INTEREST WAS RELATED TO SUCCESS IN PRIMARY PILOT TRAINING STRENGTH OF NAVIGATOR INTEREST WAS RELATED TV SUCCESS IN NAVIGATOR TRAINING THESE WAS WO CLEAR RELATION BETWEEN STRENGTH OF BOMBARDIER INTEREST AND SUCCESS IN BOMBARDIER TRAINING THESE DATA ARE FROM PRIMARY PILOT CLASS 44-C, NAVIGATOR CLASSES 43-12 THROUGH 43-15, BOMBARDIER CLASSES 43-15 THROUGH 44-1. EXPERIENCE A. cadet .who has already learned how to fly will natural- ly have less to learn in pilot training than one who has never been in a plane. However, since only about 12% of the cadets in primary pilot training had previous training, the practical value of flying experience as a selective device was rather limited. It was utilized early in the program by giving bo- nuses to the experienced men, 2 stanine points for a solo cer- tificate and 3 stanine points for a solo certificate plus 30 or more log hours, thus providing an augmented pilot stanine with a somewhat Increased predictive power. But as new and better tests were added to the battery, including tests of interest, Information about flying, and motivation, it was found that the qualities and knowledge associated with previous flying experience were being meas- ured in the tests. The relation between success in pilot training and the stanine increased. Bonuses granted for pre- vious flying experience no longer added to the predictive efficiency of the stanine and were therefore discontinued. THE MORE PREVIOUS FLYING EXPERIENCE A CADET HAD...THE MORE LIKELY HE WAS TO SUCCEED IN PRIMARY PILOT TRAINING COMMERCIAL PILOTS TOTAL NO. PRIVATE PILOTS STOPS NT PILOTS (SOto) student pilots BEEN PASSENGERS IN A PLANE Cuo FORMAL instruction) NEVER BEEN PASSENGER IN A PLANE PREVIOUS PLYING EXPERIENCE WAS NOT RELATED TO SUCCESS IN NAVIGATOR TRAINING PRIVATE PILOTS STUDENT PILOTS (solo) STUDENT PILOTS BREN PASSENGERS IN A PLANE (NO FORMAL INSTRUCTION) NEVER BEEN PASSENGER IN A PLANE THE BREADTH OP THE BARS /S PROPORTIONAL TO THE NUMBER IN EACH CAfeirOiev. PILOT PATA FROM CLASSES 43-F y 43-Gr AMP 45-H. A/Ai//GATOR PATA PROM! CLASSES 43-12 THROUGH 43-/E. BACKGROUND FACTORS It can be seen that after pilots had been screened, a slight relationship remained between success in pilot training and age, education, and marital status. But the relationship was slight and any use of these factors as a basis for selection would have eliminated a great many po tent1ally successful pilots along with the potential fall ures. AMONG_CAPETS WHO HAP BEEN SCREENED FOR APTITUDE THESE MEH HAD A SLIGHTLY BETTER CHANCE OF SUCCEEDING /N PRIMARY PILOT TRAINING' YOUNGER MARRIED. MORE EDUCATED F/SOM CLASS' 44" C ,RECOMMEND ATIONS FOR ASSIGNMENT Recommendations for assignment of cadets to the dif- ferent aircrew positions or to ground duty were made upon the basis of the stanine (aptitude score) obtained, augmented by consideration of the experience and motivation of the cadets. Due largely to local quota conditions, there were some Instances in which men recommended by the Psychological Research Units for bombardier or navigator training were assigned to pilot training. It can be seen that their elim- ination rate was almost twice as high as the men assigned according to recommendation. THE ELIMINATION RATE WAS ALMOST TWICE AS HIGH AMONG CADETS IN PILOT TRAINING WHO WERE RECOMMENDED FOR BOMBARDIER OR NAVIGATOR TRAINING AS AMONG THOSE WHO WERE RECOMMEN DED FOR PILOT TRAINING PILOT BOMBARDIER NAVIGATOR PAT A PROM CLASSES 45 P - +5 G - A3 H THE EXPERIMENTAL GROUP Prom late in 1943 through early 1944 a special exper- imental group of approximately 1000 men, recruited from the regular sources, were permitted to enter pilot training along with the thousands of regular cadets. The experimental group took the Qualifying and Classification tests but all of the men were assigned to pilot training whether or not they re- ceived sufficiently high scores to qualify them for such training. They were recruited from all over the country and their training and assignments were scattered throughout the pilot training schools, mixed unrecognizably with the other cadets. Studying this group enables us to see what would have happened if there had been no selection in effect as well as to make other analyses, several of which are presented in the following charts. men__with low pilot aptitude SCORES (STANINES) ON THE AAF AIRCREW CLASSIFICATION TESTS ARE VERY POOR PROSPECTS FOR PILOT TRAINING EXPERIMENTAL GROUP SUCCESS IN COMPLETE FLYING TRAINING A COMPARISON OF THE ELIMINATION RATES INCOMPLETE PILOT TRAINING ACCORDING TO FOUR DIFFERENT STANDARDS OF SELECTION IN AN EXPERIMENTAL GROUP OF /OOO MEN WHO WERE PERMITTED TO ENTER P/LOT TRAINING IN *943 REGARDLESS OF WHAT SCORES THEY OBTAINED ON THE SELECTION TESTS NO SCREENING AT ALL PRE-WAR SELECTION STANDARDS APTITUDE standards OF NOVEMBER '43 APTITUDE STANDARDS OF NOVEMBER '44 DESCRIPTION OF STANDARDS.' I.... NO SELECTION FOR APTITUDE H.... TWO YEARS OR MORE OF COLLEGE AND PASSING THE ADAPTABILITY RATING FOR MILITARY AERONAUTICS MADE BY FLIGHT SURGEONS. IIE.... A SCORE OF ISO OR OYER ON THE AAF QUALIFYING EXAMINATION AND A PILOT STAN/NE OF 5 OR OVER. IT.... A SCORE OF /SO OR OVER ON THE AAF QUALIFYING EXAMINATION AND A PILOT STANINE OF 7 OR OVER. THE HIGHER THE ENTRANCE STANDARDS THE LOWER THE TRAINING COST BUT... THE GREATER THE REQUIRED SUPPLY OF CANDIDATES ONE MUST HAVE AS A SUPPLY OF APPLICANTS FO/Z training (available fop appucat/on OF SCREENING PROCEDURES) ONE MUST ENTER, INTO TRAINING TO GET IOO GRADUATES FROM PILOT TRAINING USING NO SCREENING PROCEB0RES SCREEN/NS ON AAFQE SCOPE OF 130 ANO MINIMUM PILOT STANINE OF S’ screening on aafqe scope OF /QO AND MINIMUM PILOT STANINE OF 7 WEST POINT STUDIES This chart summarizes data on success in primary pilot training of U. S. Military Cadets and of Aviation Cadets trained in the same classes and schools. These results are of special interest because they make possible a comparison of the procedures at West Point with the increasing selecti- vity which operated in the case of Aviation Cadets. In the case of Military Academy Cadets, the uniform policy has been to give flight training to any cadet who elects it and who passes the physical examination. In the case of Aviation Cadets, the selection standards changed from class to class and became progressively more severe. In class 42-K some men were qualified on the basis of two years of college or the equivalent and some on the AAE Qualifying Examination but none were required to meet any further minimum pilot aptitude score on the AAE Aircrew Classification Tests. All later classes were screened on the AAE Qualifying Examination. In addition, a large ma- jority of the men with pilot stanines of 1 and 2 were being disqualified for pilot training at the time that class 43-J was being tested. During the period while class 44-J was being selected a minimum stanlne of 4, 5, and 6 was in ef- fect for pilot training. Aviation Cadets in class 45-H were required to have a minimum pilot stanine of 6 in order to qualify. IN PRIMARY PILOT TRAINING AVIATION CADETS AS COMPARED WITH WEST POINT CADETS WERE CONSIDERABLY LESS SUCCESSFUL IN THE CLASSES WHEN THERE WERE LITTLE OR NO MINIMUM APTITUDE REQUIREMENTS FOR AVIATION CADETS SU6HTLY MORE SUCCESSFUL IN THE LATER CLASSES WHEN THE MINIMUM APTITUDE REQUIREMENTS FOR AVIATION CADETS WERE RAISED * ELIMINATION IS FOR PLY/NO F£AR AND OWN RFQOSSr The AAF Aircrew Classification Battery was administered to the class of 1946 at the U. S. Military Academy in the fall of 1944, several months before they entered flying training. This is the class which received primary pilot training in class 45-H. The chert shows the elimination rate for West Point cadets at each stanine score. These data show clearly that the pilot stanine is effective in predicting training success for West Point Cadets, as it has been shown to be for Aviation Cadets. For those USMA Cadets who met the Aviation Cadet standard (stanine of 6) which was effective when the Aviation Cadets in 44-J and 45-H were qualified, the elimina- tion rate is only 21.7 percent as compared with 30.2 percent for the class as a whole. IN PREDICTING SUCCESS IN PILOT TRAINING WEST POINT CADETS thjs /$ rue class wh/ch eece/vep pp/AtApy p/lot tpa/h/ng jn class 45 p TRANSITION AND OPERATIONAL TRAINING The ability of the aptitude scores to predict success in early phases of training has been discussed at some length. Now we are ready to pursue the pilot’s career after he had received his wings, through the intervening training periods before he entered combat. TRANSITION TRAINING Most of these new officers entered transition train- ing and a small portion of these entrants were reassigned because of unsatisfactory performance. The stanlne con- tinued to discriminate the better from the poorer pilots even in this highly select group. THE HIGHER THE PILOT STANINE THE GREATER THE CHANCES OF SUCCESS IN BOMBER PILOT TRANSITION TRAINING DATA BASED OH PHOTS JH CLASSES 43-0, P, AND G AT ALL TRANS/T/ON TRA/N/NG SCHOOLS. OPERATIONAL TRAINING Most of the cadets who succeeded in passing transi- tion training went into operational training with the Con- tinental Air Forces. Here again the weeding out process continued and a small proportion of the least capable were removed from flying status by Flying Evaluation Boards. A greater proportion of the low stanine pilots were re-eval- / / uated. OF THE P-38 FIGHTER PILOTS IN OPERATIONAL TRAINING BETWEEN MAY 1944 £ MARCH 1945 35%WERE^E^EVALUAri£ BY FLYING EVALUATION BOARDS 96.7% GRADUATED 5.3% KE-EVALUATED HIGH STANINE PILOTS WERE LESS LIKELY TO BE RE-EVALUATED DATA FROM TME 4r*A/R FORCE. DATA FOR BOTE F/EDRES H/ERE DRAtV/V FROM TRE SAME POPDLAT/OA/ AA/D /MCLUDE RE-EVAWAT/O/V FOR ALL REASONS. ACCIDENTS Aircraft accidents can have many causes but it is im- portant to observe the considerable proportion attributed to pilot error. Some of these "pilot error" accidents were part- ly due to poor design of equipment for example, controls that were too far away to reach comfortably, or too close and too similar to another making it too easy for the pilot who wanted to put down his flaps to retract his landing gear Instead. Some of the dials were so Inadequately illuminated that at least one accident occurred when the pilot, using his flashlight to read a meter while taking off, spoiled his night vision and ran into a hill. Considerable research is being done on the improve- ment of cockpit controls and display to bring them into line with the physiological and psychological capacities of the men who pilot the planes and thus reduce the accident rate. Another way to decrease accidents is to select men who are less likely to have accidents. According to a study con- ducted on 1883 P-38 fighter pilots in operational training in the Fourth Air Force, the high stanine pilots had a lower pilot error accident rate than did the low stanine pilots. 157. OF P-38 PILOTS IN OPER- ATIONAL TRAINING BETWEEN MAY 1944 £ MARCH >945 HAD AT LEAST ONE ACCIDENT OVER HALF OF THESE ACCIDENTS WERE ATTRIBUTED TO PILOT ERROR HIGH STAN1NE P-38 PILOTS WERE LESS LIKELY TO HAVE PILOT ERROR ACCIDENTS data proa1 rue A7* Am ponce. data non rue three n/oones were drawn proa* the same population. experience credit has Nor seen INCLUDED /N THE STAN/NES. FIXED GUNNERY In the latter half of fighter pilot training at the oper- ational level instruction in fixed gunnery was emphasized. The number of hits on the target divided by the number of rounds fired provided a "gunnery score," that enabled the student to tell how well he was doing and also provided a measure of his proficiency. Aerial and ground gunnery scores for 1150 fighter pilots were collected, combined into a single score and then related to the pilot stanlne. The chart shows that high stanine pilots as a group made better scores than low stanlne pilots. HIGH 5TANINE PILOTS WERE LIKELY TOSCOREMOREHITSON AERIAL AND GROUND TARGETS DATA COLLECTED FROM STAT/ONS AND COMM/TMEHTS /H TEE !ST F/GHTEP COMMAND BETWEEN J(/NE /?44 AND MARCH *94E ONLY NEWLY RATED P/LOTS ARE /HCLUDED. PLANE COMMANDER VS. CO-PILOT Training personnel tried to assign the most proficient pilots to be plane commanders and to utilize the others as co-pilots. The Fourth Air Force study on 1218 B-24 airplane commanders and 1232 co-pilots indicates that the men with high pilot aptitude scores (stanines) were likely to become commanders and men with low pilot aptitude scores were likely to become co-pilots. HIGH 5TANINE PILOTS TEND TO BECOME AIRPLANE COMMANDERS RATA INCLUDE /2)8 B-24 AIRPLANE COMMANDERS AND 1232 CO-PILOTS WHO HAD ENTERED OPERATIONAL TRAINING IN THE 4th AlR PORCE DURING THE LAST THREE QUARTERS OP 1944 . EXPERIENCE CRED/T HAS NOT BEEN INCLUDED IN THE PILOT STANINE. COMBAT The ultimate goal in the selection program was to se- lect pilots, navigators and bombardiers and other and other aircrew who would do the best Job in combat. In order to evaluate this work, valid measures of combat proficiency were necessary. Obviously the combat situation does not lend it- self to easy, valid measurement. The accuracy of bombing is a function of altitude, turbulence, visibility, winds, enemy action, the nature of the target, the equipment, the profi- ciency of the other personnel involved, as well as the person being evaluated. Similarly, the number of enemy planes brought down by fighter pilots is related to the number of opportuni- ties a pilot has had, how well protected he was, how profi- cient and numerous the enemy was and so on. From these sparse examples, it can be seen that the evaluation of proficiency in combat conditions is necessarily subject to so many uncontrolled factors as to render its determination very tenuous. In spite of these obvious difficulties, three Aircrew Research Detachments of psychologists went to combat areas in the European and Pacific theaters to collect data that would, among other things, determine whether the classification test scores predict success in the combat situation. Some of the measures of proficiency gathered were bombing errors, advance- ments achieved, victories, ratings of performance, accidents, casualties, and transfer from duty because of unsatisfactory performance. These measures were compared with aptitude scores obtained from the Classification Tests to see whether "goodness" in combat was related to high aptitude scores. As might be ex- pected, due to all the difficulties involved in proficiency measurement in combat, the relations between the test scores and combat measures were small. A few of the results obtained are presented in the follow ing charts. LEAD BOMBARDIERS WITH HIGH BOMBARDIER STANINE TEND TO HAVE LOWER AVERAGE CIRCULAR ERRORS STAN WE 7,8,9 TOP TH/ZP OP TOTAL GZOUP STAN/NE 6 MIDDLE TH/ZPOF TOTAL STAN/NE /. 2,3.4* LOWEST TW*0 Of TOTAL GPOUP THESE PfOOZES ZEPZESEHT A COA4POS/TE OP ALL OP THE PAT A COLLECTEP PZOA4 THE 8m AHp /STM AlZ FOZCES OH THE ZELAT/OH SETA'S EH CJZCOLAZ Ba/HS/HG EZZOZ A HP THE 30 AfOAZ PJEZ STAN/HE. BOMBARDIERS WHO WERE SUPERIOR IN THESE ABILITIES BEFORE THEY ENTERED TRA/MH6 TENDED TO HAVE SMALLER CIRCULAR BOMBING ERRORS /N COMBAT THE ABILITY TO IDENTIFY TARGETS IN PHOTOGRAPHS THE ABILITY TO SELECT QUICKLY THE CORRECT MANUAL MOVEMENTS IN RESPONSE TO THE RELATIVE LOCATION OF TWO COLORED LIGHTS KNOWLEDGE OF AND INTEREST IN AVIATION DATA COLLECTED BY A£R£> AA/D AEEO *2 PBOM TUB B7*4 AA/D /S™ A/* FORCES . THE BARS JTEPRESEAST THE PERCENT SCO*'*6 BELOW TME AiED/AA/ C/RCOCAR ERROR AT EACH SCO*/HE LEVEL. PILOTS REEVALUATED WHILE ON COMBAT DUTY WERE LIKELY TO HAVE LOWER PILOT 5TANINES RE-EVALUATEO PILOTS CONTROL PILOTS WHO HAD FINISHED TOURS OF DUTY PATA COLLECTED BY AERD*2 PROM THE S™. I2rH £ IS™ A !R FORCES AND USSTAP. the RE-EVALUATED P/LOTS WERE GROUNDED Ano/oK RECLASSIFIED BY CME, FEB, OR MD3 FOR PSYCHOLOGICAL REASONS OR BECAUSE OF JLACR OF PROFESSIONAL QUALIFICATIONS. TN£ CONTROL GROUP CONSISTED OF PILOTS VINO NAD FIN I SUED THEIR TOUR OF DOTY AND WERE MATCHED BY TESTING NUMBERS TO THE RE-EVALUATED PILOTS. In conclusion, it Is evident that these procedures were very effective in the work of aircrew selection and classification for the AAF during the war period. Many of the tests and other features of these procedures were adopted by the Royal Air Force, the French Air Forces, knd other allied nations. Later in the war emphasis in the Aviation Psychology Program turned toward problems of the more accurate evaluation of men trained in the various aircrew specialties and to research on other prob- lems of training and operations and of Improving the ef- fectiveness of the use of equipment by a more systematic consideration of human capacities in its design. Plans for the continuation of psychological research during the peacetime period are now being developed and it is hoped that through the systematic use of scientific methods of investigation and development, it will be pos- sible to achieve continuous improvement in operational effectiveness with an integrated research program begin- ning with problems in the design of new weapons and carrying on through selection and classification, train- ing, evaluation and operating procedures.