The Establish Of Basic Cognitive Ability Tests And Their Parallel Tests For Civil Aviators

Flight is a dangerous and complex occupation. Compared to the general occupations,it requires the pilot to have a higher level of cognitive ability,objectively. Therefore, in theprocess of selection of aviators, the group which is not fit for this occupation can beeliminated through psychological selection to achieve the purpose of saving training costsand ensuring flight safety.At present, the work of psychological selection of pilot is done by the Air Forcemostly. The establishment of selection tests of civil aviation is far from enough. Althoughthe Civil Aviation Administration and other departments developed written tests for pilotselection in the1990s, and introducted a full set of test in the cooperation with Lufthansa,these tests were not used in the selection of pilot actually. Therefore, it is very necessary to establish comprehensive and reliable tests for pilot selection.This study attempted to reestablish and screen cognitive ability tests we establishedbefore for the pilot selection in the civil aviation, and to establish three sets of paralleltests and corresponding standards, which was based on the pilot’s basic and requisitecognitive ability.In this process,the candidates whose ability can’t meet the standard forflight would be eliminated, which would save training costs and ensure flight safety.This study was divided into three parts:1The choice, adaption, and optimization of basic cognitive ability test for pilotselection.Objective: On the basis of the literature review and survey in pilots to concludenecessary ability for flight, and to adapt and screen the selected basic ability tests toestablish final tests.Methods:I reviewed the literature, summed up the ability of flight, surveyed maturepilot through questionnaire,and ensured these selected basic capability necessary for apilot. According these basic capability, the initial edition of the basic tests were selected toadapt and optimize. Completely random method was emploied to select participants inselection of a domestic airline. The first group included220subjects, and the secondgroup included240subjects. They completed the adapted11tests, and the second adapted8Aptitude Test. The software of SPSS17.0was used to analyse the degree of normality,difficulty degree and distinction degree.Results:(1) In the11tests the first group completed, the average difficultycoefficient was0.82, and the average distinction degree was0.28. the tests ofcalculating and sum of points and numbers were used to examine the abilitycalculating ability. The difficulty coefficient was0.81, and the distinction degree was0.22in calculating. The difficulty coefficient was0.78,and distinction degree was0.31insum of points and numbers. visual interference and visual search were used toexamine the ability of selection and distribution in attention. The difficulty coefficient was0.83,and the distinction degree was0.29in visual disturbance test. The difficultycoefficient was0.86,and the distinction degree was0.25in visual search. The difficultycoefficient was0.88,and the distinction degree was0.23in visual memory. But somespecific skills were found in test process. So this test could not reflect the participants’trueabilities.(2) The second group completed the adapted8tests. The difficulty coefficientwere between0.76~0.84. And the average difficulty coefficient was0.79. The distinction degree were between0.23~0.37,and the average distinction degree was0.31.Conclusion:(1) According to results of the first group, the tests of calculating,visual search and visual memory were deleted, and the eight following tests wereremained: words understanding, task execution, direction reasoning, spatialrotation,"visual disturbance", graphics reasoning, instrumentation understanding,sum of points and numbers,and attentional convertion.(2) After the modification ofthe eight test,648questions’ distinction degree were more than0.2, accounting for81.5%of the total questions. After the modification of some items in tests, the items in each testwere ultimately determined.2The establishment of parallel tests of basic cognitive ability tests.Objective: To develop three sets of parallel tests, and analyse their reliability,validity and equivalence.Methods: According to the type and difficulty of the ultimately determined items, allof the items in these tests were divided into different levels.Items in each level wererandomly divided into three portions, then. At last they would form three parallel tesst.240randomly selected candidates for re-examination from a domestic airline were dividedinto three groups randomly.Each group completed a parallel set of test. SPSS17.0statistical software was emploied to analyse the data.Results:(1) The average difficulty coefficient of the first set of test was0.79, andthe average degree of distinction was0.35. The average difficulty coefficient of the secondset of test was0.78, and the average degree of distinction was0.34. The average difficultycoefficient of the third set of test was0.78, and the average degree of distinction was0.35.(2)Except for that the split-half reliability of visual interference test in the first setof test was0.69, split-half reliability of all the tests have reached more than0.7.(3) In hefirst set of test, the correlation between each subtest score and the total score was in therange of0.52~0.60. In the second set of tests, the correlation between each subtest scoreand total score was in the range of0.52~0.70. In the third set of test, the correlationbetween each subtest score and total score was in the range of0.44~0.70.(4)Exploratoryfactor analysis was used to analyse the three sets of tests. And five factors could beextracted out as follows: the factor of spatial logic, the factor of extraction from numberand characters information, calculation factor, attention factor, verbal comprehensionfactor. In he first set of test, the five cumulative factors explained83.65%of the variance.In the second set of test, the five cumulative factors explained85.16%of the variance. And in the third set of test, the five cumulative factors explained83.34%of the variance.(5) when the subtest scores of the three tests were compared, the six subtests, directionreasoning, spatial rotation,"visual disturbance", graphics reasoning, instrumentcomprehension, attention convertion, no statistical differences was found(P>0.05).Inthe test of words comprehension and task execution, there was statistical differencebetween the first and the second set of test (P<0.05). In the test of sum of points andnumbers, there was statistical difference between the second and the third set of test. thebetween statistically significant difference (P<0.05). When comparing8subtest in eachset of test, there was no statistical difference between them in difficulty coefficient(P>0.05).Conclusions:(1) The diffifulty, distinction degree, reliability and validity of thethree sets of tests were good enough to achieve the purpose of pilot selection.(2) According the equivalence analysis results of the three tests, the mean, standarddeviation, difficulty distributed roughly similiarly, which met the psychometricrequirements.3Comparative study on basic cognitive ability tests between pilots and retestcandidates.Objective: To test and verify whether the tests could identify differences betweenpilots and ordinary people in specific cognitive ability of flight.Methods:158pilots who were trained in U.S. aviation school and obtained adiploma and U.S. certificaiton issued by Federal Aviation Administration(FAA) wereincluded in this study. They were divided into three groups in the completely random way.Each group completed three sets of parallel tests. The results were compared with theresults in the second part(the candidates who participated in the re-examination). At lastthe software of SPSS17.0was emploied for statistical analysis.Results: In the three sets of tests, there were statistical difference between pilots andcandidates(re-examination) when comparing the tests: direction reasoning, spatialrotation,"visual disturbance", graphical reasoning, instrument comprehension,sum of points and numbers (P<0.05), which indicated that the pilots’ score weresuperior to the candidates (re-examination).Meanwhile, there were no statistical differencein words comprehension and task execution of the first and third set of tests, and inattention convertion of the first set of test between the pilots andcandidates(re-examination)(P>0.05). Conclusion: The established three sets of tests can reflect the ability that pilots havebetter oen than ordinary occupations roughly. Test result is valid, which can provide somebasis for the airline pilot selection.