JournalofAirTransportationWorldWide Vol.6, No.2 – 2001 LLEEAARRNNIINNGG SSTTYYLLEESS OOFF PPIILLOOTTSS CCUURRRREENNTTLLYY QQUUAALLIIFFIIEEDD IINN UUNNIITTEEDD SSTTAATTEESS AAIIRR FFOORRCCEE AAIIRRCCRRAAFFTT11 CCrraaiiggAA..KKaannsskkee,,SSkkyyVViieewwss,,LL..LL..CC.. MMoooorree,,OOkkllaahhoommaa AABBSSTTRRAACCTT KKoollbb’’ss LLeeaarrnniinngg SSttyyllee IInnvveennttoorryy wwaass uusseedd ttoo iiddeennttiiffyy tthhee pprreeddoommiinnaanntt lleeaarrnniinngg ssttyylleess ooff ppiilloottssccuurrrreennttllyyqquuaalliiffiieeddiinnUUnniitteeddSSttaatteessAAiirrFFoorrcceeaaiirrccrraafftt..TThheerreessuullttssiinnddiiccaatteetthhaatttthheessee ppiilloottss sshhooww aa ssiiggnniiffiiccaanntt pprreeffeerreennccee ffoorr ffaaccttss aanndd tthhiinnggss oovveerr ppeeooppllee aanndd ffeeeelliinnggss.. BByy uunnddeerrssttaannddiinnggtthheepprreeffeerrrreeddlleeaarrnniinnggssttyylleessoofftthheettaarrggeettppooppuullaattiioonn,,ccoouurrsseemmaatteerriiaallccaannbbee ddeevveellooppeeddtthhaattttaakkeeaaddvvaannttaaggeeoofftthheessttrreennggtthhssoofftthheesseelleeaarrnniinnggssttyylleess..TThhiissiinnffoorrmmaattiioonnccaann bbee eessppeecciiaallllyy uusseeffuull iinn tthhee ffuuttuurree ddeessiiggnn ooff ccoocckkppiitt rreessoouurrccee mmaannaaggeemmeenntt ttrraaiinniinngg.. TThhee ttrraaiinniinnggpprrooggrraammccaannbbeeddeevveellooppeeddttooddeemmoonnssttrraatteebbootthhtthhaatttthheerreeaarreeddiiffffeerreennttlleeaarrnniinnggssttyylleess aannddtthhaattiittiissppoossssiibblleettoottaakkeeaaddvvaannttaaggeeoofftthheerreellaattiivveessttrreennggtthhssooffeeaacchhoofftthheesseelleeaarrnniinngg ssttyylleess.. IINNTTRROODDUUCCTTIIOONN TThheeppuurrppoosseeoofftthhiissssttuuddyywwaassttooddeetteerrmmiinneetthheelleeaarrnniinnggssttyylleessooffppiilloottss ccuurrrreennttllyyqquuaalliiffiieeddiinnUUnniitteeddSSttaatteessAAiirrFFoorrcceeaaiirrccrraafftt..HHoowwssttuuddeennttsslleeaarrnn iissiimmppaacctteeddbbyyhhoowwtthheemmaatteerriiaalltthheeyyaarreettoolleeaarrnniisspprreesseenntteedd..SSttuuddiieesshhaavvee sshhoowwnn tthhaatt mmoorree eeffffeeccttiivvee lleeaarrnniinngg iiss aacchhiieevveedd wwhheenn pprrooggrraammss ttaakkee iinnttoo aaccccoouunntt tthhee lleeaarrnniinngg ssttyylleess ooff tthhee ttaarrggeett ppooppuullaattiioonn ((WWoooollddrriiddggee,, 11999955)).. IInnccrreeaassiinnggssttuuddeennttlleeaarrnniinngg,,tthheeddeessiirreeddoouuttccoommeeooffaalllliinnssttrruuccttiioonn,,rreeqquuiirreess ddeevveellooppiinngg aann aabbiilliittyy ttoo rreeccooggnniizzee ssttuuddeennttss’’ lleeaarrnniinngg ssttyylleess aanndd uussee tteecchhnniiqquueesstthhaattiinnccrreeaasseetthheepprroobbaabbiilliittyyooffaacchhiieevviinnggssuucccceessss((AAnnddeerrssoonn&& AAddaammss,,11999922)).. CraigKanske,Ed.D.,isaretiredAirForcepilotandinstructorpilotintheE-3FlightCrew TrainingprogramatTinkerAirForceBase,Oklahoma.Heisalsoanadjunctinstructorinthe AviationEducationDepartmentatOklahomaStateUniversity—Tulsa,Tulsa,Oklahoma;and intheAviationDepartmentofSoutheasternOklahomaStateUniversityattheOklahomaCity Aviation Education Alliance, Tinker Air Force Base, Oklahoma. He earned a Master of ScienceinSystemsManagementfromtheUniversityofSouthernCaliforniaandaDoctorof EducationinAviationandSpaceEducationfromOklahomaStateUniversity. 1Completetextcanbefoundin:Kanske,C.A.(1998/1999).Thelearningstylesofpilots currentlyqualifiedinUnitedStatesAirForceaircraft(Doctoraldissertation,OklahomaState University, Aviation and Space Program 1998). Dissertation Abstracts International, 9918799. ©2001,AviationInstitute,UniversityofNebraskaatOmaha 34 JournalofAirTransportationWorldWide The study of learning styles has its roots in the field of psychology. Therehavebeentwomainpathsforstudyasresearchhasmovedforward from these roots. One path has followed the classic Pavlovian stimulus- response approach, using reinforcement of successful completion at each stepinasequentiallearningprocess.Theotherpathhasfocusedinsteadon thecognitiveprocessesinlearning.Researchersconductingcurrentstudies of learning styles have mainly chosen this second path, focusing on the cognitiveprocessesofthelearner(Sims&Sims,1995). TheMyers-BriggsTypeIndicator,fielddependence/fieldindependence, and brain hemispherocity studies are all examples of measures for cognitive based learning. Another example is the Kolb Learning Style Inventory.Therearemanyotherapproachestomeasuringlearningstyles, with diverse terminology and measurement instruments. The common groundforalloftheseapproaches,however,isthattheyattempttodescribe thelearningstylesoftheindividualbymeasuringtheindividual’sbehavior during the learning process. Through this measurement, each instrument attemptstodescribehowtheindividualtakesinandprocessesinformation. Sims&Sims(1995)provideanaptsummary,statingthat“…regardlessof howthatprocessisdescribed,itisdramaticallydifferentforeachperson” (p.194). Kolb’s (1984) approach to measuring this learning process is through theexperientiallearningmodel.Theexperientiallearningmodelproceeds fromtheassumptionthatalllearningisinfluencedbythepriorexperiences oftheindividuallearner.Becauseofthisassumptionthatpriorexperience influenceseachnewlearningevent,learningcanbeviewedasacontinuous process. How the learner progresses through this process, or uses this process,becomesthefocusfordefiningthatlearner’slearningstyle. Theexperientiallearningmodeldescribesfourphasesofthecontinuous learningexperience.Concreteexperienceisinvolvementwiththelearning event,absorbingthesurroundingsandactivitiesastheyhappen.Reflective observationisreviewingtheexperiencesandattemptingtodeterminewhat isnewanddifferentabouttheexperience,andwhatissimilartoprevious experiences.Abstractconceptualizationistheprocessofintegratingthese experiences and reflections into a modified view of the learner’s environment. Finally active experimentation is the process of testing this newworldview(Kolb,1984). The“perfect”learnerwoulduseallfourmodesoflearningequally,and would shift around the learning model smoothly with each new learning situation. The “normal” learner, on the other hand, develops a preferred mode of learning. Whether this preferred style is adopted as the result of Kanske 35 positive reinforcement in earlier, similar situations (Schmeck, 1988) or risesfromdeeper,personalitybasedroots,theeffectisthatthelearnertends to “specialize” in a specific style of learning. Identifying this preferred styleoflearningisthefocusoflearningstyleresearch. Kolb’s(1985)LearningStyleInventoryusestwelvesentencestemswith fourendingseachtomeasurepreferredlearningstyle.Eachofthesentence endings indicates a preference for one of the four learning modes associated with the experiential learning model. Summing the responses foreachofthetwelvesentencesyieldsasetofnumbersbetween12and48 which represents the degree to which the learner emphasizes each of the four learning modes. These scores provide an indication of the balance betweenthelearningmodesforthelearner. Because the four stages of the experiential learning model represent polar opposites of two learning scales, it is possible to use the individual element scores to derive a number which represents the individual’s position along each of these scales. In Kolb’s (1985) Learning Style Inventorythisisdonebysubtractingthescoreforconcreteexperiencefrom the score for abstract conceptualization and subtracting the score for reflectiveobservationfromthescoreforactiveexperimentation. The range of 12 to 48 on each individual learning mode yields range extremes of plus or minus 36 for the active experimentation minus reflective observation and abstract conceptualization minus concrete experiencescales.Theseformulasprovideanumericalrepresentationofa learner’s relative emphasis for the types of learning represented by each axis. There is no qualitative differentiation between the learning modes; rather this process provides a way to display the results in a linear presentation showing the relative strength of a specific style for an individual. Thequadrantofthegraph,formedbythesetwoscales,whichcontains the combined score for the individual defines that learner’s predominant learningstyle. Withintheexperientiallearningmodelthequadrantformedbyconcrete experience and reflective observation is called divergent learning. The divergerprefersbeingapartofthelearningexperience,andthinkingabout whathashappenedduringthatexperience.Theoppositepreferenceliesin the quadrant formed by abstract conceptualization and active experience and is called convergent learning. The convergent learner takes multiple observationsofmanyeventsandbringsthemtogetherintotheanswertoa specificproblem. 36 JournalofAirTransportationWorldWide The other two quadrants produce assimilative and accommodative learners. The quadrant formed by reflective observation and abstract conceptualization is called assimilative learning. The assimilator is the inductive reasoner who can put together coherent theories based upon observations,integratingmultipleobservationsintoacohesiveexplanation of the events. The active experimentation/concrete experience quadrant producestheaccommodativelearner.Theaccommodatorgetsthingsdone andispartoftheaction. Knowing which learning style the learner prefers provides important information for course design. There is disagreement among researchers over whether it is better to match the preferred learning style to ease the learningprocess,ortomismatchthestyletoforcethelearnerto“stretch” intoanotherstyle.Regardlessofwhichmethodispreferred,however,there isagreementthatthisdecisionmustbedesignedintothecourseasopposed to being the result of ignoring the possibility that differences in learning stylesexist(Sadler-Smith,1996). PILOTLEARNINGSTYLES TheemphasisinpilottraininghasfollowedthePavlovianinsteadofthe cognitivepath.Coursesforpilottrainingarebasedupontasklistswhichthe student must master to successfully complete the training program. The taskispresented,demonstrated,andthenthestudentpracticesuntilthetask ismastered.Appropriatefeedbackisprovidedbytheinstructorduringthe practice session. There is no effort spent on determining the cognitive basedlearningstyleofthestudent. Inthisageoftechnologyandinformationthereisanefforttomovesome of this training into the classroom using computer based training and simulation(Thiesse,Newmeyer,andWidick,1992;Treiber,1994).Itisin this effort to apply new technologies to pilot training that understanding learningstylescanbehelpfulincoursedesign.Oncethelearningstyleof pilotsisunderstood,thedecisiontomatch,ormismatch,thesestylescanbe aconsciousoneinsteadofbeinglefttochance. Currently the predominant learning style of pilots is not well understood.ThreestudiesbyQuilty(1995,1996,1997)haveaddressedthe global versus analytical cognitive bias of pilots with differing levels of experience.StudiesbytheUnitedStatesAirForce(Carretta&Siem,1988) havefocusedonpredictingthechancesofaspecificindividualsuccessfully completing the Undergraduate Pilot Training program. This study used Kolb’s (1985) Learning Style Inventory to identify the predominant Kanske 37 learning styles of pilots currently qualified to fly United States Air Force aircraft. METHODOLOGY Initiallydevelopedin1976,andrevisedin1984(Kolb,1976,1984),the Learning Style Inventory has been used, and validated, in such diverse studies as comparing learning styles between high school and college students (Matthews & Hamby, 1995), a cross-cultural comparison of the learning styles between Western and Oriental learners (Auyeng & Sands, 1996), and comparisons of the learning styles among European management training students (Jackson, 1995). Recent validation studies suchasthatconductedbyWillcoxsonandProsser(1996)haveproventhe continued usefulness of the LSI as a measure of predominant learning styles. Based on this history of use no pre-test evaluation of the Learning StyleInventorywasconductedforthisstudy. Thesampleforthisstudywasgeneratedbythedataretrievalsectionof the Air Force Personnel Center (AFPC). According to S. Heitkamp (personal communication, April 13, 1998) the data storage system has a built in randomization process for extracting names from the master personnelfilebasedonthespecifiedsortcriteria.Usingthisbuiltinsystem alistof600pilotswasdrawnatrandomfromtheAirForceMasterOfficer Personnel File. Copies of the Kolb Learning Style Inventory and an accompanying demographic data questionnaire were mailed to these 600 United States Air Force pilots in March 1998. Two hundred thirty-three surveyswerereturnedbytheendofJune1998,withcompletedandusable instruments.Another63surveyswerereturnedbytheUnitedStatesPostal Serviceasundeliverable.Thereturnof233of537surveysprovideda43.8 percentresponserateforthestudy. Theoriginalsamplesizewaschosenusingmethodologydevelopedby KrejcieandMorgan(1970)toprovidea95percentprobabilityofmatching the population of 14,000 pilots in the United States Air Force. The rank structure of the final sample very closely approximated this population, withslightlyfewercaptainsthanexpected,andslightlymoremajorsthan expected. Table 1 shows the distribution of the sample relative to the populationforrankstructure,gender,ethnicity,andtypeofaircraftflown. Sampledistributionsforgenderandethnicityalsocloselyapproximated populationdistributions.Thesampleheld5percentfemalesand95percent males compared to a population of 2.5 percent female and 97.5 percent male.Sampledistributionsforethnicitywerewithintwopercentagepoints inallcategories.Thedistributionofthesamplebasedupontypeofaircraft 38 JournalofAirTransportationWorldWide flown was within 3 percent of the population distribution for all types of aircraft. Distribution comparisons for female pilots were not meaningful duetothesmallnumberinthesample. Chi-squareanalysisoftheexpecteddistributionofthesampleforrank, gender,andethnicityproducedastatisticallysignificantmatchbetweenthe sample and the population. A Pearson product moment of 28.50146999, with12degreesoffreedomreturnsaprobabilityof<0.005ofachievingthe sampledistributionbyrandomchance. Table1.Samplevs.PopulationDistributions Sample Population percent percent GradeDistribution 2LT 3.4 2.0 1LT 5.1 5.0 CPT 48.1 54.0 MAJ 27.9 24.0 LTC 15.5 15.0 GenderDistribution Female 5.2 2.4 Male 94.8 97.6 EthnicDistribution AmericanIndian/Alaskan 1.3 0.3 Asian/PacificIslander 0.4 1.0 Black(non-Hispanic) 1.7 2.0 Hispanic 1.7 1.0 Other 2.6 1.0 White(non-Hispanic) 92.3 94.7 AircraftDistribution Fighter/Attack/Reconnaissance(FAR) 27.5 33.7 Tanker/Transport/Bomber(TTB) 57.1 60.1 Helicopter(HELO) 5.2 5.3 Other/None 0.0 0.5 Both 9.9 0.0 FINDINGS Kolb’s (1984) norming process for the Learning Style Inventory producedmedianscoresof5.9foractiveexperimentationminusreflective observation and 3.8 for abstract conceptualization minus concrete experience.Thesampleresultof5.93fortheactiveexperimentationminus reflectiveobservationaxisshowsthereisnogreateremphasisplacedupon activeexperimentationorreflectiveobservationbypilotsthanisshownin Kanske 39 the general population. The sample result of 8.39 for abstract conceptualizationminusconcreteexperience,however,issignificant.Two- tailed t-test probability is less than .0001 for achieving this result at random. Pilots show a significantly stronger tendency to emphasize abstractconceptualizationoverconcreteexperience. Basedonthisconcreteexperienceandabstractconceptualizationdata,it can be said that the average pilot in the United States Air Force significantlyemphasizesthingsandthoughtoverpeopleandfeelings.The reflectiveobservationandactiveexperimentationdatareflectsapreference for active participation over observation; however, this preference is not statisticallydifferentfromthepreferenceshownbythepopulationatlarge whencomparedtothenormingsample(Kolb,1984). Descriptivestatisticsforthesample,showingthescoresforeachofthe sixlearningstylemeasurements,areshowninTable2. Table2.SurveyResults—DescriptiveStatistics Standard Mean Median Deviation ConcreteExperience 23.85 20.00 10.28 ReflectiveObservation 28.98 28.00 6.47 AbstractConceptualization 32.24 32.00 7.23 ActiveExperimentation 34.91 37.00 9.10 AbstractConceptualizationminusConcreteExperience 8.39 11.00 14.86 ActiveExperimentationminusReflectiveObservation 5.93 9.00 12.41 LEARNINGSTYLES Plotting active experimentation minus reflective observation and abstract conceptualization minus concrete experience as the “X” and “Y” axesofagridformsamatrixwhichcanbeusedtodefinethequadrantsof the experiential learning cycle. Kolb (1984) used this graphic representationtoplotthefourlearningstyles(Figure1).Themedianscores for active experimentation minus reflective observation (5.9) and abstract conceptualizationminusconcreteexperience(3.8)definetheintersection. Themeanvaluesfromthesamplewere5.93foractiveexperimentation minusreflectiveobservationand8.39forabstractconceptualizationminus concreteexperience.Whenplottedonthestylegridthesemeanvaluesfall on the boundary between Converger and Assimilator. Medians for the sample were 9.0 for active experimentation minus reflective observation and 11.0 for abstract conceptualization minus concrete experience. The 40 JournalofAirTransportationWorldWide Figure1.StyleGrid plot for these median values falls within the Converger learning style. Whenindividualresponsesareplottedonthisgrid15.8percent(n=37)are accommodators, 23.6 percent (n = 55) are assimilators, 44.2 percent (n = 103)areconvergers,and16.3percent(n=38)aredivergers. Thisdistributionoflearningstylesissignificant(p<.0001)relativetoa hypotheticaldistributionof25percentineachstyleaswouldbeshownina randomsampleofthepopulationatlarge.Thissignificancecorrespondsto the predictive nature of the Learning Style Inventory (Kolb, 1984). In the case of pilots currently qualified in United States Air Force aircraft, the predominant learning style is convergence. A secondary learning style is assimilation. Divergent and accommodative learning styles are each used bysignificantlysmallgroupsofpilotswithinthestudygroup. Theanalysisofpilots’learningstyleswasbaseduponaveragedatafor the entire sample. Demographic data was collected for military rank, gender,ethnicity,typeofaircraftflown,andnumberofflyinghours.Inall categories except ethnicity the convergent learning style was the predominant selection at a statistically significant level. Analysis of learningstylesbyethnicitywasnotaccomplishedduetothesmallnumbers ofnon-whiteethnicgroupswithinthesample. Kanske 41 CONCLUSIONS The predominant learning style of pilots currently qualified in United StatesAirForceaircraftistheconvergentlearningstyle.Thislearningstyle fitswithinthepredictivenatureofKolb’s(1985)LearningStyleInventory asdemonstratingtraitswhicharevaluableinspecialistandtechnicalfields. Thereareimplicationsforfuturecoursedesigntomatchthislearningstyle. Learners with the convergent style prefer to know how it works as opposed to who says it works. These learners want to do it themselves ratherthanbeingshownhowtodoit,buttheywouldratherbeshownthatit works than take an expert’s word that it works. This will be especially important in the design of computer based training modules which introducenewequipmentandtechnologytopilotsduringtrainingcourses. How the system fits together, and why it works, are more important to convergersthanjustbeingtoldthatthesystemworks. The secondary learning style for pilots currently qualified in United StatesAirForceaircraftistheassimilativestyle.Thisstyleisincludedasa secondary learning style because of the relationship convergent and assimilative learning styles have relative to concrete experience and abstract conceptualization. Both styles show a preference for abstract conceptualization over concrete experience. The choice between a preference for active experimentation over reflective observation is the difference in learning preference which separates the converger from the assimilator. Although more individuals fell into the converger learning stylethantheassimilatorlearningstyle,thesamplemean,veryclosetothe dividing point between these two styles, was used in considering the assimilativestyleasasecondarylearningstyleforthepilotsinthisstudy. There are similar considerations for designing future courses for the accommodativelearnersintheclass.Theysharetheconverger’sdesireto know how something works rather than who says that it works. Their preferenceforreflectiveobservation,however,canleadthemtolookforall the available alternatives and overlook the fact that they have a workable solution already. Building into the training program justifications for limiting the scope of information will be important for the assimilative learner. Taken together, the predominant converger learning style and the secondary assimilative learning style support the effectiveness of the current training program. The abstract conceptualization focus shared by these two learning styles works well with the demonstration/performance modeofteachingbecauseofthefocusonhowthingsworkasopposedto who says these things work. By seeing how things are done, and 42 JournalofAirTransportationWorldWide understandingtheimplications,theabstractconceptualizercanworkfrom theindividualpartstocreateawhole. Thebalancebetweenreflectiveobservationandactiveexperimentation shown by the mean score of the sample also supports the current training program. This balance between careful observation and risk taking, and looking at problems from many angles and putting this information into actionformthebasisforsounddecisionmakinginthetime-criticalnature ofaviation. Inshort,theverythingswhichareusedtopredictsuccessfulcompletion of Undergraduate Pilot Training (Carretta & Siem, 1988) are the factors whichappearinthepredominantlearningstyleofcurrentUnitedStatesAir Force pilots. The sorting process coincides with two elements of the experiential learning model. The first element is socialization, where working in aviation tends to emphasize certain characteristics within the individual due to the requirements of the task. The second element is the process of the individual tending to gravitate towards a field where the requirementsmatchthatindividual’spersonalcharacteristics.Thisprocess of specialization provides a basis for the predictive nature of the learning styleinventoryandtheexperientiallearningmodel(Kolb,1984). Oneofthecurrentareasofemphasisinaviationtrainingiscrewresource management. This training program emphasizes skills in relating to other individuals,bothonthecrewandinpositionswhichinteractwiththecrew focusing on team coordination, attitudes, behaviors, and communications (Driskell&Adams,1992).Addressinglearningstyleswithincrewresource management training courses can provide an additional approach to defining the issues for all crewmembers. Understanding individual differences provides a critical stepping stone toward improvement within theseareas. Pilots with the predominant converger learning style “…would rather deal with technical tasks and problems than with social and interpersonal issues”(Kolb,1985,p.7).Thefocusofcrewresourcemanagementtraining is these very social and interpersonal issues. The characteristics of the convergent learner, as well as the other learning styles, should be incorporated into course design for crew resource management training. Analyzingthedifferentlearningstyles,includingthedifferencesbetween thestylesandthestrengthsandweaknessesofeachstyle,allowsthegroup to make better use of the skills available through its individual members (Sims&Sims,1995). Incorporating learning styles into the design of crew resource management training provides an opportunity for better understanding of