By: Paul H. Cheney
Skill Requirements Survey
INFORMATION SYSTEMSSKILL REQUIREMENTS: A SURVEY By: Paul H. Cheney Norman R. Lyons
Introduction At this time, it is clear that the dataprocessing industryis in a stateof transitionthat is unlikeany other previously undergone.The advancesin semiconductortechnologyhavemadecomputers cheaperandmorewidely available than ever before. Oneseesa greatdealof discussion of topics like "personalcomputing" in boththe professional andpopularliterature. Theimplicationof much of this literature is that weare on the vergeof a revolutionin computing that will make much of the workcurrently doneby computerprofessionals unnecessary.The newmachinesand languages that are coming into existencewill allowthe user to be his ownsystemsanalyst and programmer. In this way,wewill beable to do awaywith the mistakesand poor quality systemsproducedby the present generation of programmers.This point of viewis greetedwith someskepticismby information systemsprofessionals.1 Theyfeel that if the problems of systemsdesignweresimple enough to be solvedby beginners,theywould have been solved long ago. All this hasled to some uncertaintyas to the type of skills that will berequiredof computer professionals and indeed,someuncertaintyas to the numbers of computerprofessionalsthat will be requiredin the future.
Abstract Thisstudyreportson some of the employment trends andskill requirements in theinformation systems area as seenby the information systems managers of 32of thenation’slargestorganizations. Thedatawasgatheredvia persona/interviews andquestionnaires. We believetheresultsto bequiteinterestingin termsof bothworkforceprojections andtheperceived jobskills that are requiredfor programmers, systems analysts, anddata processingmanagers. Keywords:informationsystemsmanagement, skills, programming, systemsanalyst ACM Categories: 2.42, 3.50
In this study, wereport on someof the employmenttrendsandskill requirements in the information systemsarea as seenby the information systemsmanagers of 32 of the nation’s largest organizations in boththe public andprivate sector. Wefeel that eventhoughthe samplesize is small, theseorganizationsrepresenta goodindicator of current needsin informationprocessing technology,andwithin the limits of their managers’ abilities to projectfuture trends,theygivea goodindicationof whatthe future holdsin a wide variety of data processingorganizations.All 32 organizationshaveutilized computers for fifteen years or moreand the respondents wereexperienced professionals whoseideas represent someof the best thinking in the industry. 1Asanexample, seethearticle by M.I. Bernstein,"Hardware is Easy:It’s SoftwareThat’sHard"in Datamation, November 15, 1978,pp. 32-36.
MIS Quarterly/March 1980 35
Skill Requirements Survey
The Survey This studyconcerned four majorareas:1) informationskills inventoryandpersonnel projections; 2) hardwareconfigurationsand characteristics (i.e., the use of mini-computers);3) software characteristics(L e., the useof database managementsystems); and 4) information systemdepartmentcharacteristics(i.e., centralizationof hardware,data, and people). Theproject was fundedwith a grant fromTexasTechUniversity andit wasconducted over a three monthperiod. Theorganizationsparticipated primarily because theyfelt the informationfromthe studywouldaid in staffing dataprocessing positionsin thefuture. In addition, mostof the respondents felt that a surveyof this type wasneeded to determinethe directions in whichinformationsystemsdepartmentsweremovingwith regardto hardware,software,andtheir internal organizationstructure. Thedata onwhichthis article is basedwasgathered using personal interviews and questionnaires. Theinterviewer had a specific set of questionsthat werecoveredin the interview with the executives.Sincethe interviewer wasalso knowledgeable in the information systemsarea, he couldalso departfromthe format. Severalof the questionswereopenended;others required the interviewee to select a responsefrom a group of alternatives. Table 1 showsa breakdownof the titles of the respondents in the interview. Thetotal number of respondents is greater than 32 becausein a numberof organizations, there wasmorethan one respondentsince no single individual in the company wasknowledgeable in all the areascoveredin the interview. Table2 presentsthe levels of data processing experienceof the respondents.The organizations involvedin the studyrepresented a variety of industries in severalgeographic areas.Tables 3 and 4 presentthe breakdown of the organizations by industry and geographiclocation.
SurveyResults Thefirst part of the surveydealt with the managers’ projectionsof their informationsystems personnel needs. Theywere asked the numberof employees that they hadin eachof severalareas and howmanythey expectedto havein eachof
36 MIS Quarterly/March 1980
these samecategories in 1979and 1982.Their replies are summarized in Table 5. Noticein Table5 that substantialincreaseswere projectedin all staff levels exceptthosefor data entry personnel.Thedecreases in dataentry personneldono.t mean that dataentryis a thing of the past. Theincreasinguseof terminalorientedsystemshasmeantthat dataentry is nolongerbeing doneat a centrallocationunderthe controlof the informationsystems group.Instead,data entry is doneby clerks in the user’s department.This probablydoesmean a net reductionin dataentry staff since thosesameclerks in the user departmentswouldhaveprobablybeenengaged in preparing formsfor processingby the central data processinggroup.To make the table data a little clearer, wehave preparedTable 6 which gives the workforceprojectionsin eachyear as a percentageusing1978as a baseyear. Tables5 and 6 representthe same data in different forms;actual numbersin Table 5 and percentageincreaseswith 1978as the base in Table 6. Thefigures presented in Table6 indicatethat our respondents foreseea growthrate of better than 50%for applications programmers and system analysts over the four year period from1978to 1982.Thenumber of data entry personnelis expectedto dropby half over the next four years, andthe numberof operatorsis expectedto grow by only 18%in the same period. Theseare probably two reflections of the growthin the useof online systems.Feweroperatorsare required to manage online systemsthan conventional card and tape oriented systems,so even thoughthe number of computer applicationsis likely to grow, at least as judgedby the numbers of applications programmers required; the needfor operatorsis not likely to keeppace. Anotherinteresting point demonstrated in Table6 is the projected growthof databasemanagement programmers. Thedemand for this type of programmeris expected to double by 1982 which wouldsuggestthat the organizationsin our sample are goingto increasetheir workin the databaseareasubstantiallyoverthe nextfewyears.In total, the results shownin Tables5 and6 would suggest that statementsabout the demiseof computer programmingas a profession are premature, to say the least. Ourrespondents indicated very strong demand for applications orientedstaff into the early 1980’s.
Skill Requirements Survey
Table 1. Respondents by Job Title Numberof Respondents 8 6 9 22
JobTitle V.P. of Information Systems Director of InformationSystems Data Center Manager Director of Information SystemsDevelopment
Table 2. Data ProcessingExperienceof Respondents Numberof Respondents 3 5 37
Yearsof Data ProcessingExperience 5-10 years 10-15 years over 15 years
Table 3. Respondents by Industry Numberof Respondents 17 4 6 2 3
Industry Manufacturing Service Government Retailing/Wholesaling Banking/Insurance
Table 4. Respondents by Location Numberof Respondents 5 14 6 7
Location East Midwest Southwest West
Table 5. WorkforceProjections
I. Applications Programmers COBOLProgrammers FORTRANor BASIC Database Management I1.Systems Programmers Ill. SystemsAnalysts IV. Operators V. Data Entry
1978 8,250 7,170 290 308 628 4,403 1,408 2,380
1979 10,274 8,883 365 435 721 5,036 1,527 1,945
1982 12,555 10,826 389 642 863 6,697 1,673 1,195
MIS Quarterly/March 1980 37
Skill Requirements Survey
Table 6. PercentageChangesin WorkforceProjections
I. Applications Programmers COBOL FORTRAN or BASIC Database Management II. SystemsProgrammers III. Systems Analyst IV. Operator V. Data Entry
Thesecondpart of our questionnairedealt with the managers’perception of the importanceof certainspecificskills to thevarioustypesof information systemspersonnel. The managerswere givena list of specificskill areasandasked to rank the importance of theseskills for three groups: data center managers,systemsanalysts, and programmers. Theratings wereon a scale of 1 to 5, with 1 meaningnot useful and5 meaning essential. Theresults of this set of questionsare tabulatedin Table7. Weusedthe followingdefinitions for the job titles of data centermanager, systemsanalyst, and programmer: 1. data center managermThe data center manager supervisesboth the systemsdevelopment staff (i.e., analysts and programmers) and the operations staff (i.e., operators, librarians, input/ output control clerks, anddata entry personnel).Heis responsiblefor the
1978 100% 100% 100% 100% 100% 100% 100% 100%
1979 125% 124% 126% 141% 115% 114% 108% 82%
1982 152% 151% 134% 208% 137% 152% 118% 50%
efficient utilization of resources within the informationsystemsareawhile accomplishingdepartmentaland organizational goals. 2. systems analystmThesystems analyst is concernedwith defining the users’ information needs and then building a systemto generatethe requiredinformation.Hewill define the contentandstructure of input forms, outputreports, andfiles. 3. programmer--The programmer is primarily responsiblefor program developmentonce the systemanalyst has definedthe inputs, outputs,andfiles. Twosimilar studies wereconducted in an attempt to determine the skills requiredof dataprocessing professionals. An ACM study suggestedthe importance of six skill clusters:people,models, sys-
Table 7. Skill Ratings, Mean,andStandardDeviation Skill Area 1. Information GatheringTechniques 2. SystemsDesign Topics 3. File Design 4. Planning and Control of Systems Projects
38 MISQuarterly~March1980
Data Center Manager
Systems Analyst
Programmer
3.69 (.966) 3.42 (1.08) 3.00 (1.00) 4.69 (.682)
4.92 (.253) 4.92 (.253) 4.75 (.661) 4.28 (.829)
3.89 (.933) 4.22 (.992) 4.91 (.384) 3.09 (.87O)
Skill
Requirements Survey
Table 7. Skill Ratings, Mean, and StandardDeviation (Continued’ Data Center Manager
Skill Area 5. HumanRelations in Systems Development 6. HumanFactors in Equipment Design and Work Layout 7. Introductory Computerand Information Systems Concepts 8. Applications Programming Languages 9. Job Control Languages 10. Database Management Systems 11. Operating SystemCharacteristics 12. HardwareCharacteristics 13. Minicomputer Characteristics
and Uses
14. Telecommunications Concepts 15. ComputerSecurity Controls and Auditing 16. Software Package Analysis 17. Computer Scheduling 18. Legal Aspects of Computing 19. Improving ComputerCenter Productivity 20. Introductory Statistics 21. Computer Simulation 22. Statistical
Decision Theory
23. Regression Analysis 24. Sampling Theory 25. List Processing 26. Sorting
4.59 (.755) 4.31 (.855) 4.88 (.544) 3.48 (1.17) 2.86 (1.06) 3.81 (1.20) 3.80 (1.30) 4.13 (I.08) 3.65 (I.i2) 4.19 (1.02) 4.88 (.331) 4.53 (.968) 4.53 (.829) 4.44 (.933) 4.66 (.734) 2.70 (.980) 3.35 (1.06) 2.55 (1.25) 2.40 (1.09) 2.36 (1.09) 2.05 (I.05) 2.02 (i.I0)
Systems Analyst 4.69 (.569) 4.39 (.855) 4.97 (. 174) 4.42 (I.01) 3.97 (1.10) 4.42 (.872) 4.03 (1.10) 4.06 (I.09) 3.68 (1.14) 4.33 (1.00) 3.78 (.8O9) 3.34 (I.03) 3.05 (i.02) 2.89 (1.11) 2.92 (i.08) 2.63 (1.06) 2.97 (1.23) 2.42 (1.16) 2.28 (I.02) 2.28 (I.02) 2.62 (1.24) 2.60 (1.29)
Programmer 3.98 (.906) 3.25 (1.15) 4.97 (. 174) 4.88 (.415) 4.88 (.415) 4.50 (.773) 4.56 (.778) 4.16 (1.03) 3.28 (1.07) 4.19 (1.07) 3.25 (.992) 2.59 (1.03) 2.67 (.825) 2.38 (I.02) 2.56 (.864) 2.13 (1.06) 2.30 (1.19) 1.93 (I.04) 1.93 (.989) 1.93 (.989) 2.74 (1.56) 3.98 (1.11)
MIS Quarterly/March
1980 39
Skill Requirements Survey 2 terns, computers,organization, and society. Their recommended curriculum to provide these skills washighly technically oriented. TheManagementInformation SystemsResearchCenter conducteda study to address the ACMrecommendations in an attemptto seekdirection for 3 Theydetermined their MISprogram. that significant skill requirementswerenot being met for programmers, analysts, and data center managers and recommended continuing educationprogramsto help solve this problem. Table 7 presents a formidable array of data. Shortly, wewill try to aggregate this data to a greater degreeand drawsomeconclusionsfrom it. It is instructiveto lookat the areasthat were considered important by the managerswhose responsesmakeup Table7. Notice that in general, they demand a high degreeof skill in many areasfromthe personoccupyingthe position of data center manager.Accordingto the answers given, theyrequirea substantialbreadthof technical knowledgeabout computingequipmentas well as job specific skills like computer scheduling, the legal aspectsof computing, andalso traditional management skills. It wouldseemfrom the responses that a data centermanager is supposedto be generallya technical andmanagerial expert.Perhaps it is expectations like thesethat make the job sucha difficult one.Perhaps this is also whycomputermanagement is regardedas a "dead end" by manymanagers.The skills requiredare so hardto locate that oncesomeone is found whopossessesthem, the tendencyis to keepthat personon the job rather than promote her or him. Alsoin Table7, noticethe skills that are required of a programmer. Thehighest rated skills are those associatedwith job performance,namely the applications programming andcomputersystemsconcepts.Otherskills that wouldbe helpful in management at somepoint are regarded as less importantin the background of a programmer.If theseskill requirements are an accurate reflection of the hiring practicesof management, 2ACM CurriculumCommittee on Computer Educationfor Management,"CurriculumRecommendations for GraduateProfessional Programsin Information Systems,"Communications of the ACM,May1972, Volume15, Number5. 3Henry,R. M., Dickson, G. W., and LaSalle J. F. "Human Resources For MIS: A Reportof Research,"MISRC Working PaperSeries (74-01), University of Minnesota,1973.
41~ MISQuarterly~March1980
andwesuspectthat they are, this maywell be a reason for the poor performanceof manyprogrammerswhenthey are movedinto managementpositions. Theskills whichwerevaluedin the programming job are not the sameoneswhich are required in the management job, and this meansthat the programmer is promotedbeyond his level of competence. Anycompany that hasa policy of promotingprogrammers to management positions ought to see that they receive some additional training in the management areasso that they can adequatelyhandletheir newresponsibilities whenpromoted. Table10 indicates the lack of correlationbetween the skills required by data center managers and the skills required by programmers, (i.e., .271). It is interestingto notein Table7 that thequantitative management areas, as indicated by skill areas20 through24, werenot regardedas highly importantfor either managers, systemsanalysts, or programmers.Thesetopics would form the basefor anykind of sophisticatedMISfor useby top level management, andthe fact that they are not regardedas importantwouldsuggestthat the informationsystemseffort in thesecompanies is oriented towardconventional data processing andlower level operationalcontrol systems.It appears fromthis datathat weare still a longway off from a stage of MISdevelopment wherethe computer is usedextensivelyfor tactical planning by middlemanagers or for strategic planningby top level management. In orderto make thedetail of Table7 a little easier to grasp,wehavedividedthe skill areasinto five majorcategoriesandcomputed ratings for these categories. This has beendonein Table 8. Table8 points up evenmoreclearly someof the comments wemadeabout Table 7. It appears that systemsanalysts and programmers are expectedto be fairly specializedin their ownareas. Thedata center manager,on the other hand,is expectedto be widelyliterate acrossmanyinformationsystemsareas. Becausethe job expectations for systemsanalysts and programmers are quite different fromthe job expectations for data centermanagers, it appears that recruiting future computermanagersfrom the ranks of programmersand analysts could be a risky process. Toget a clearerideaof thetypesof skills that were consideredmostimportant by the managers we
Skill Requirements Survey
Table 8. AverageSkill Ratings by Category
Category 1. SystemsDesign (Questions1-6) 2. Hardwareand Software (Questions7-14) 3. Information SystemsManagement (Questions15-19) 4. Quantitative Management (Questions20-24) 5. ComputerProcessing Methods (Questions25-26)
Data Center Manager 3.95
Systems Analyst 4.66
Programmer 3.89
3.85
4.24
4.43
4.61
3.20
2.69
2.67
2.52
2.04
1.91
2.36
2.80
surveyed,Table9 presentsthe skill areasranked in order of importance as determined by our survey. When twoor moreskill areasare listed at the samerank in the table, that means that these areashadtied ranks.Notethat the topareafor all three types of jobs wasIntroductory Computer and Information Concepts.This wouldsuggest that managers considera base level of competence in computersystemsto be important for anyoneworkingin the informationsystemsarea. After that, theranksfall prettymuch into theareas you might expectbasedon our earlier observations. Therankswouldsuggestthat the peoplein eachof thesethree job areasare expected to be fairly specialized,andthat for a programmer or systemsanalyst to moveto a management position, moretraining wouldbe necessary. Toshowthe overlap,or lackof it, among the skill areasfor the three job categorieswehaverun a Spearman RankCorrelation betweenthe skill areasdesiredfor eachof the threejob types. The results are presentedin Table 10. This table wouldsuggestonly slight correlation between the skill areasdesired for programmers and those desired for managers,and a somewhat better correlation betweenmanagers and systemsanalysts. Thedifferencesbetween the skills required for the programmer and the data center manager are primarily in the areaof informationsystems management.Data center managersmust be skilled at planning,organizing,anddirectingpersonnel whereasprogrammersneed not have theseskills to as greata degree (i.e., seeTable8).
Table 10. SpearmanRankCorrelation Between Skill Area Rankingsfor Different Jobs Jobs Correlation Data Center Managerswith SystemsAnalysts .528 Data Center Managerswith Programmers .271 SystemsAnalysts with Programmers .818
Conclusions Admittedlyour samplesize of 32 wasnot large andweare awareof the limitations this presents in termsof our findings. Wedo, however, believe that our sample is stratified. It is stratifiedby industry andgeographicallocation with regardto large computer installations. In addition,the companieswereselectedat random froma list of the 500largestpublicandprivateorganizations in the UnitedStates. Webelieve that the distribution showncouldbetruly representativeof the large mainframe installations aroundthe country. A number of interesting results havecomeout of this survey.First of all, it seems that the demand for information systemspersonnelis going to growat a veryrapidrate (i.e., 52% for application programmersby 1982).
MISQuarterly~March1980 41
Skill Requirements Survey
Table 9. Skill AreasRankedin Order of Importance Rank Data Center Managers Rank 1
2 3 4 5
6 7
8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Systems Analyst
Introductory Computer 1 Introductory Computer and Information and Information System Concepts System Concepts ComputerSecurity Con- 2 SystemsDesign Topics trois andAuditing InformationGathering Techniques Planningand Control of SystemsProjects 3 File Design Improving ComputerCen- 4 Human Relations in ter Productivity SystemsDevelopment HumanRelations in Applications Program5 Systems Development ming Languages Software Package Database Management 6 Analysis Systems ComputerScheduling 7 Human Factors in Legal Aspectsof EquipmentDesign and Computing Work Layout HumanFactors in 8 Telecommunications EquipmentDesign and Concepts Layout 9 Planningand Control of Telecommunications Systems Development Concepts Project Hardware Characteristics 10 Hardware Characteristics Database Management 11 Operating Systems Systems Characteristics Operating Systems 12 Job Control Languages Characteristics 13 ComputerSecurity ConInformationGathering trois andAuditing Techniques 14 MinicomputerCharacMinicomputerCharacteristics andUses teristics andUses 15 Software Package SystemsDesign Topics Analysis Applications Program16 ComputerScheduling ming Languages 17 ComputerSimulation ComputerSimulation 18 ImprovingComputerCenter Productivity File Design Job Control Languages 19 Legal Aspectsof Computing IntroductoryStatistics 20 IntroductoryStatistics Statistical Decision Theory 21 List Processing RegressionAnalysis 22 Sorting Sampling Theory 23 Statistical Decision Theory List Processing 24 RegressionAnalysis Sorting SamplingTheory
42 MIS Quarterly/March 1980
Rank 1
Programmer
Introductory Computer andInformation Concepts 2 File Design 3 Applications Programming Languages Job Control Languages 4 Operating System Characteristics 5 Database Management Systems 6 SystemsDesign Topics 7 Telecommunications Concepts 8 i Hardware Characteristics 9 Human Relations in Systems Deve opment 10 Sorting 11 InformationGathering Techniques 12 MinicomputerCharacteristics and Uses 13 HumanFactors in EquipmentDesign and WorkLayout ComputerSecurity Controis andAuditing 14 Planningand Control of SystemsProjects 15 List Processing 16 ComputerScheduling 17 Software Package Analysis 18 Improving Computer CenterProductivity 19 Legal Aspectsof Computing ComputerSimulation 20 IntroductoryStatistics 21 Statistical Decision Theory 22 RegressionAnalysis SamplingTheory
Skill Requirements Survey
An additional point should be madeabout this conclusionconcerningthe growthin the needfor informationsystems personnel.It is probablytoo conservative. Our samplewastaken from large sophisticatedorganizationswith a longtradition of computer use.This fails to takeinto accounta very large segmentof computerusers, namely thosesmallercompanies whoare setting up their first computerized data processingsystemon a mini or micro-computer.Thesecompaniesare also generatingconsiderabledemand for appli4 cations programmers. Ourstudyindicatesthat the growthin demand for information systemspersonnelin the next few years will be largely in the traditional areas. COBOL basedsystemsstill seemto be dominant in mostinformation systemdepartments.Databasetechnologyis also becoming increasingly important.Theoneareaof decentralizationthat did showupin our results wasthe dataentry area whereuser departmentswill be bearing an increasingshareof the loadfor dataentry. It is too early to say whetherthis also means that eventually theseuserdepartments are likely to wantto developtheir ownapplicationsas well andgenerate a push for decentralized computing.Our guessis that they will not. Oursurveyof the skill areasvaluedby the informationsystemsmanagers showsthat there is a tendencyto compartmentalize the job areasand keeppeoplein their ownareasof specialty. The skill areasdesiredfor informationsystemsmanagersas shownin the surveyare a very demanding combination of technicalandmanagerial skills that probablygo beyondthe requirements placed on mostother types of middle managers.This ’~Cheney,P. andFuerst,W."Sharinga Computer Programmer: Onesolutionto Software Problems in SmallFirms,"Journalof Small Business Management, October 1979, Volume17, Number 4, pp. 43-48.
wouldmake the job difficult to fill, as mostof us havenotedin practice, and meansthat oncea suitable candidateis found,he tendsto be stuck in the job because replacements are hardto find. Theskill areasdemanded for managers are sufficiently different fromthosefor programmers and systemsanalystswhichsuggeststhat there is a definite andprobablygrowingneedfor information systemsprograms in collegesanduniversities with a managerialas well as a technical emphasis. In conclusion,this surveypoints upa strongand continuing demand for information systemspersonnelthat is not likely to be affected by new developments in technology.It is far to early to predict the demiseof the large computer system and the information systemsprofessional.
Aboutthe Authors Paul H. Cheneyreceived his doctorate in ManagementInformation Systems from the University of Minnesota. His current research includes the areas of auditing and controlling computer based accounting systems, decision support systems, and implementationstrategies for newinformation systems. Heis presently an associate professor in the Industrial Administration Department at Iowa State University. NormanR. Lyonsreceived his doctorate in Industrial Administration from CarnegieMellon University. Heis currently an associate professor at the Naval Post-graduate School in Monterey,California. He has published in ManagementScience, Communications of the ACM,Journal of Accountancy, and others. His majorresearchinterest is in the area of database management systems.
MIS Quarterly~March1980 43
INFORMATION SYSTEMSSKILL REQUIREMENTS: A SURVEY By: Paul H. Cheney Norman R. Lyons
Introduction At this time, it is clear that the dataprocessing industryis in a stateof transitionthat is unlikeany other previously undergone.The advancesin semiconductortechnologyhavemadecomputers cheaperandmorewidely available than ever before. Oneseesa greatdealof discussion of topics like "personalcomputing" in boththe professional andpopularliterature. Theimplicationof much of this literature is that weare on the vergeof a revolutionin computing that will make much of the workcurrently doneby computerprofessionals unnecessary.The newmachinesand languages that are coming into existencewill allowthe user to be his ownsystemsanalyst and programmer. In this way,wewill beable to do awaywith the mistakesand poor quality systemsproducedby the present generation of programmers.This point of viewis greetedwith someskepticismby information systemsprofessionals.1 Theyfeel that if the problems of systemsdesignweresimple enough to be solvedby beginners,theywould have been solved long ago. All this hasled to some uncertaintyas to the type of skills that will berequiredof computer professionals and indeed,someuncertaintyas to the numbers of computerprofessionalsthat will be requiredin the future.
Abstract Thisstudyreportson some of the employment trends andskill requirements in theinformation systems area as seenby the information systems managers of 32of thenation’slargestorganizations. Thedatawasgatheredvia persona/interviews andquestionnaires. We believetheresultsto bequiteinterestingin termsof bothworkforceprojections andtheperceived jobskills that are requiredfor programmers, systems analysts, anddata processingmanagers. Keywords:informationsystemsmanagement, skills, programming, systemsanalyst ACM Categories: 2.42, 3.50
In this study, wereport on someof the employmenttrendsandskill requirements in the information systemsarea as seenby the information systemsmanagers of 32 of the nation’s largest organizations in boththe public andprivate sector. Wefeel that eventhoughthe samplesize is small, theseorganizationsrepresenta goodindicator of current needsin informationprocessing technology,andwithin the limits of their managers’ abilities to projectfuture trends,theygivea goodindicationof whatthe future holdsin a wide variety of data processingorganizations.All 32 organizationshaveutilized computers for fifteen years or moreand the respondents wereexperienced professionals whoseideas represent someof the best thinking in the industry. 1Asanexample, seethearticle by M.I. Bernstein,"Hardware is Easy:It’s SoftwareThat’sHard"in Datamation, November 15, 1978,pp. 32-36.
MIS Quarterly/March 1980 35
Skill Requirements Survey
The Survey This studyconcerned four majorareas:1) informationskills inventoryandpersonnel projections; 2) hardwareconfigurationsand characteristics (i.e., the use of mini-computers);3) software characteristics(L e., the useof database managementsystems); and 4) information systemdepartmentcharacteristics(i.e., centralizationof hardware,data, and people). Theproject was fundedwith a grant fromTexasTechUniversity andit wasconducted over a three monthperiod. Theorganizationsparticipated primarily because theyfelt the informationfromthe studywouldaid in staffing dataprocessing positionsin thefuture. In addition, mostof the respondents felt that a surveyof this type wasneeded to determinethe directions in whichinformationsystemsdepartmentsweremovingwith regardto hardware,software,andtheir internal organizationstructure. Thedata onwhichthis article is basedwasgathered using personal interviews and questionnaires. Theinterviewer had a specific set of questionsthat werecoveredin the interview with the executives.Sincethe interviewer wasalso knowledgeable in the information systemsarea, he couldalso departfromthe format. Severalof the questionswereopenended;others required the interviewee to select a responsefrom a group of alternatives. Table 1 showsa breakdownof the titles of the respondents in the interview. Thetotal number of respondents is greater than 32 becausein a numberof organizations, there wasmorethan one respondentsince no single individual in the company wasknowledgeable in all the areascoveredin the interview. Table2 presentsthe levels of data processing experienceof the respondents.The organizations involvedin the studyrepresented a variety of industries in severalgeographic areas.Tables 3 and 4 presentthe breakdown of the organizations by industry and geographiclocation.
SurveyResults Thefirst part of the surveydealt with the managers’ projectionsof their informationsystems personnel needs. Theywere asked the numberof employees that they hadin eachof severalareas and howmanythey expectedto havein eachof
36 MIS Quarterly/March 1980
these samecategories in 1979and 1982.Their replies are summarized in Table 5. Noticein Table5 that substantialincreaseswere projectedin all staff levels exceptthosefor data entry personnel.Thedecreases in dataentry personneldono.t mean that dataentryis a thing of the past. Theincreasinguseof terminalorientedsystemshasmeantthat dataentry is nolongerbeing doneat a centrallocationunderthe controlof the informationsystems group.Instead,data entry is doneby clerks in the user’s department.This probablydoesmean a net reductionin dataentry staff since thosesameclerks in the user departmentswouldhaveprobablybeenengaged in preparing formsfor processingby the central data processinggroup.To make the table data a little clearer, wehave preparedTable 6 which gives the workforceprojectionsin eachyear as a percentageusing1978as a baseyear. Tables5 and 6 representthe same data in different forms;actual numbersin Table 5 and percentageincreaseswith 1978as the base in Table 6. Thefigures presented in Table6 indicatethat our respondents foreseea growthrate of better than 50%for applications programmers and system analysts over the four year period from1978to 1982.Thenumber of data entry personnelis expectedto dropby half over the next four years, andthe numberof operatorsis expectedto grow by only 18%in the same period. Theseare probably two reflections of the growthin the useof online systems.Feweroperatorsare required to manage online systemsthan conventional card and tape oriented systems,so even thoughthe number of computer applicationsis likely to grow, at least as judgedby the numbers of applications programmers required; the needfor operatorsis not likely to keeppace. Anotherinteresting point demonstrated in Table6 is the projected growthof databasemanagement programmers. Thedemand for this type of programmeris expected to double by 1982 which wouldsuggestthat the organizationsin our sample are goingto increasetheir workin the databaseareasubstantiallyoverthe nextfewyears.In total, the results shownin Tables5 and6 would suggest that statementsabout the demiseof computer programmingas a profession are premature, to say the least. Ourrespondents indicated very strong demand for applications orientedstaff into the early 1980’s.
Skill Requirements Survey
Table 1. Respondents by Job Title Numberof Respondents 8 6 9 22
JobTitle V.P. of Information Systems Director of InformationSystems Data Center Manager Director of Information SystemsDevelopment
Table 2. Data ProcessingExperienceof Respondents Numberof Respondents 3 5 37
Yearsof Data ProcessingExperience 5-10 years 10-15 years over 15 years
Table 3. Respondents by Industry Numberof Respondents 17 4 6 2 3
Industry Manufacturing Service Government Retailing/Wholesaling Banking/Insurance
Table 4. Respondents by Location Numberof Respondents 5 14 6 7
Location East Midwest Southwest West
Table 5. WorkforceProjections
I. Applications Programmers COBOLProgrammers FORTRANor BASIC Database Management I1.Systems Programmers Ill. SystemsAnalysts IV. Operators V. Data Entry
1978 8,250 7,170 290 308 628 4,403 1,408 2,380
1979 10,274 8,883 365 435 721 5,036 1,527 1,945
1982 12,555 10,826 389 642 863 6,697 1,673 1,195
MIS Quarterly/March 1980 37
Skill Requirements Survey
Table 6. PercentageChangesin WorkforceProjections
I. Applications Programmers COBOL FORTRAN or BASIC Database Management II. SystemsProgrammers III. Systems Analyst IV. Operator V. Data Entry
Thesecondpart of our questionnairedealt with the managers’perception of the importanceof certainspecificskills to thevarioustypesof information systemspersonnel. The managerswere givena list of specificskill areasandasked to rank the importance of theseskills for three groups: data center managers,systemsanalysts, and programmers. Theratings wereon a scale of 1 to 5, with 1 meaningnot useful and5 meaning essential. Theresults of this set of questionsare tabulatedin Table7. Weusedthe followingdefinitions for the job titles of data centermanager, systemsanalyst, and programmer: 1. data center managermThe data center manager supervisesboth the systemsdevelopment staff (i.e., analysts and programmers) and the operations staff (i.e., operators, librarians, input/ output control clerks, anddata entry personnel).Heis responsiblefor the
1978 100% 100% 100% 100% 100% 100% 100% 100%
1979 125% 124% 126% 141% 115% 114% 108% 82%
1982 152% 151% 134% 208% 137% 152% 118% 50%
efficient utilization of resources within the informationsystemsareawhile accomplishingdepartmentaland organizational goals. 2. systems analystmThesystems analyst is concernedwith defining the users’ information needs and then building a systemto generatethe requiredinformation.Hewill define the contentandstructure of input forms, outputreports, andfiles. 3. programmer--The programmer is primarily responsiblefor program developmentonce the systemanalyst has definedthe inputs, outputs,andfiles. Twosimilar studies wereconducted in an attempt to determine the skills requiredof dataprocessing professionals. An ACM study suggestedthe importance of six skill clusters:people,models, sys-
Table 7. Skill Ratings, Mean,andStandardDeviation Skill Area 1. Information GatheringTechniques 2. SystemsDesign Topics 3. File Design 4. Planning and Control of Systems Projects
38 MISQuarterly~March1980
Data Center Manager
Systems Analyst
Programmer
3.69 (.966) 3.42 (1.08) 3.00 (1.00) 4.69 (.682)
4.92 (.253) 4.92 (.253) 4.75 (.661) 4.28 (.829)
3.89 (.933) 4.22 (.992) 4.91 (.384) 3.09 (.87O)
Skill
Requirements Survey
Table 7. Skill Ratings, Mean, and StandardDeviation (Continued’ Data Center Manager
Skill Area 5. HumanRelations in Systems Development 6. HumanFactors in Equipment Design and Work Layout 7. Introductory Computerand Information Systems Concepts 8. Applications Programming Languages 9. Job Control Languages 10. Database Management Systems 11. Operating SystemCharacteristics 12. HardwareCharacteristics 13. Minicomputer Characteristics
and Uses
14. Telecommunications Concepts 15. ComputerSecurity Controls and Auditing 16. Software Package Analysis 17. Computer Scheduling 18. Legal Aspects of Computing 19. Improving ComputerCenter Productivity 20. Introductory Statistics 21. Computer Simulation 22. Statistical
Decision Theory
23. Regression Analysis 24. Sampling Theory 25. List Processing 26. Sorting
4.59 (.755) 4.31 (.855) 4.88 (.544) 3.48 (1.17) 2.86 (1.06) 3.81 (1.20) 3.80 (1.30) 4.13 (I.08) 3.65 (I.i2) 4.19 (1.02) 4.88 (.331) 4.53 (.968) 4.53 (.829) 4.44 (.933) 4.66 (.734) 2.70 (.980) 3.35 (1.06) 2.55 (1.25) 2.40 (1.09) 2.36 (1.09) 2.05 (I.05) 2.02 (i.I0)
Systems Analyst 4.69 (.569) 4.39 (.855) 4.97 (. 174) 4.42 (I.01) 3.97 (1.10) 4.42 (.872) 4.03 (1.10) 4.06 (I.09) 3.68 (1.14) 4.33 (1.00) 3.78 (.8O9) 3.34 (I.03) 3.05 (i.02) 2.89 (1.11) 2.92 (i.08) 2.63 (1.06) 2.97 (1.23) 2.42 (1.16) 2.28 (I.02) 2.28 (I.02) 2.62 (1.24) 2.60 (1.29)
Programmer 3.98 (.906) 3.25 (1.15) 4.97 (. 174) 4.88 (.415) 4.88 (.415) 4.50 (.773) 4.56 (.778) 4.16 (1.03) 3.28 (1.07) 4.19 (1.07) 3.25 (.992) 2.59 (1.03) 2.67 (.825) 2.38 (I.02) 2.56 (.864) 2.13 (1.06) 2.30 (1.19) 1.93 (I.04) 1.93 (.989) 1.93 (.989) 2.74 (1.56) 3.98 (1.11)
MIS Quarterly/March
1980 39
Skill Requirements Survey 2 terns, computers,organization, and society. Their recommended curriculum to provide these skills washighly technically oriented. TheManagementInformation SystemsResearchCenter conducteda study to address the ACMrecommendations in an attemptto seekdirection for 3 Theydetermined their MISprogram. that significant skill requirementswerenot being met for programmers, analysts, and data center managers and recommended continuing educationprogramsto help solve this problem. Table 7 presents a formidable array of data. Shortly, wewill try to aggregate this data to a greater degreeand drawsomeconclusionsfrom it. It is instructiveto lookat the areasthat were considered important by the managerswhose responsesmakeup Table7. Notice that in general, they demand a high degreeof skill in many areasfromthe personoccupyingthe position of data center manager.Accordingto the answers given, theyrequirea substantialbreadthof technical knowledgeabout computingequipmentas well as job specific skills like computer scheduling, the legal aspectsof computing, andalso traditional management skills. It wouldseemfrom the responses that a data centermanager is supposedto be generallya technical andmanagerial expert.Perhaps it is expectations like thesethat make the job sucha difficult one.Perhaps this is also whycomputermanagement is regardedas a "dead end" by manymanagers.The skills requiredare so hardto locate that oncesomeone is found whopossessesthem, the tendencyis to keepthat personon the job rather than promote her or him. Alsoin Table7, noticethe skills that are required of a programmer. Thehighest rated skills are those associatedwith job performance,namely the applications programming andcomputersystemsconcepts.Otherskills that wouldbe helpful in management at somepoint are regarded as less importantin the background of a programmer.If theseskill requirements are an accurate reflection of the hiring practicesof management, 2ACM CurriculumCommittee on Computer Educationfor Management,"CurriculumRecommendations for GraduateProfessional Programsin Information Systems,"Communications of the ACM,May1972, Volume15, Number5. 3Henry,R. M., Dickson, G. W., and LaSalle J. F. "Human Resources For MIS: A Reportof Research,"MISRC Working PaperSeries (74-01), University of Minnesota,1973.
41~ MISQuarterly~March1980
andwesuspectthat they are, this maywell be a reason for the poor performanceof manyprogrammerswhenthey are movedinto managementpositions. Theskills whichwerevaluedin the programming job are not the sameoneswhich are required in the management job, and this meansthat the programmer is promotedbeyond his level of competence. Anycompany that hasa policy of promotingprogrammers to management positions ought to see that they receive some additional training in the management areasso that they can adequatelyhandletheir newresponsibilities whenpromoted. Table10 indicates the lack of correlationbetween the skills required by data center managers and the skills required by programmers, (i.e., .271). It is interestingto notein Table7 that thequantitative management areas, as indicated by skill areas20 through24, werenot regardedas highly importantfor either managers, systemsanalysts, or programmers.Thesetopics would form the basefor anykind of sophisticatedMISfor useby top level management, andthe fact that they are not regardedas importantwouldsuggestthat the informationsystemseffort in thesecompanies is oriented towardconventional data processing andlower level operationalcontrol systems.It appears fromthis datathat weare still a longway off from a stage of MISdevelopment wherethe computer is usedextensivelyfor tactical planning by middlemanagers or for strategic planningby top level management. In orderto make thedetail of Table7 a little easier to grasp,wehavedividedthe skill areasinto five majorcategoriesandcomputed ratings for these categories. This has beendonein Table 8. Table8 points up evenmoreclearly someof the comments wemadeabout Table 7. It appears that systemsanalysts and programmers are expectedto be fairly specializedin their ownareas. Thedata center manager,on the other hand,is expectedto be widelyliterate acrossmanyinformationsystemsareas. Becausethe job expectations for systemsanalysts and programmers are quite different fromthe job expectations for data centermanagers, it appears that recruiting future computermanagersfrom the ranks of programmersand analysts could be a risky process. Toget a clearerideaof thetypesof skills that were consideredmostimportant by the managers we
Skill Requirements Survey
Table 8. AverageSkill Ratings by Category
Category 1. SystemsDesign (Questions1-6) 2. Hardwareand Software (Questions7-14) 3. Information SystemsManagement (Questions15-19) 4. Quantitative Management (Questions20-24) 5. ComputerProcessing Methods (Questions25-26)
Data Center Manager 3.95
Systems Analyst 4.66
Programmer 3.89
3.85
4.24
4.43
4.61
3.20
2.69
2.67
2.52
2.04
1.91
2.36
2.80
surveyed,Table9 presentsthe skill areasranked in order of importance as determined by our survey. When twoor moreskill areasare listed at the samerank in the table, that means that these areashadtied ranks.Notethat the topareafor all three types of jobs wasIntroductory Computer and Information Concepts.This wouldsuggest that managers considera base level of competence in computersystemsto be important for anyoneworkingin the informationsystemsarea. After that, theranksfall prettymuch into theareas you might expectbasedon our earlier observations. Therankswouldsuggestthat the peoplein eachof thesethree job areasare expected to be fairly specialized,andthat for a programmer or systemsanalyst to moveto a management position, moretraining wouldbe necessary. Toshowthe overlap,or lackof it, among the skill areasfor the three job categorieswehaverun a Spearman RankCorrelation betweenthe skill areasdesiredfor eachof the threejob types. The results are presentedin Table 10. This table wouldsuggestonly slight correlation between the skill areasdesired for programmers and those desired for managers,and a somewhat better correlation betweenmanagers and systemsanalysts. Thedifferencesbetween the skills required for the programmer and the data center manager are primarily in the areaof informationsystems management.Data center managersmust be skilled at planning,organizing,anddirectingpersonnel whereasprogrammersneed not have theseskills to as greata degree (i.e., seeTable8).
Table 10. SpearmanRankCorrelation Between Skill Area Rankingsfor Different Jobs Jobs Correlation Data Center Managerswith SystemsAnalysts .528 Data Center Managerswith Programmers .271 SystemsAnalysts with Programmers .818
Conclusions Admittedlyour samplesize of 32 wasnot large andweare awareof the limitations this presents in termsof our findings. Wedo, however, believe that our sample is stratified. It is stratifiedby industry andgeographicallocation with regardto large computer installations. In addition,the companieswereselectedat random froma list of the 500largestpublicandprivateorganizations in the UnitedStates. Webelieve that the distribution showncouldbetruly representativeof the large mainframe installations aroundthe country. A number of interesting results havecomeout of this survey.First of all, it seems that the demand for information systemspersonnelis going to growat a veryrapidrate (i.e., 52% for application programmersby 1982).
MISQuarterly~March1980 41
Skill Requirements Survey
Table 9. Skill AreasRankedin Order of Importance Rank Data Center Managers Rank 1
2 3 4 5
6 7
8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Systems Analyst
Introductory Computer 1 Introductory Computer and Information and Information System Concepts System Concepts ComputerSecurity Con- 2 SystemsDesign Topics trois andAuditing InformationGathering Techniques Planningand Control of SystemsProjects 3 File Design Improving ComputerCen- 4 Human Relations in ter Productivity SystemsDevelopment HumanRelations in Applications Program5 Systems Development ming Languages Software Package Database Management 6 Analysis Systems ComputerScheduling 7 Human Factors in Legal Aspectsof EquipmentDesign and Computing Work Layout HumanFactors in 8 Telecommunications EquipmentDesign and Concepts Layout 9 Planningand Control of Telecommunications Systems Development Concepts Project Hardware Characteristics 10 Hardware Characteristics Database Management 11 Operating Systems Systems Characteristics Operating Systems 12 Job Control Languages Characteristics 13 ComputerSecurity ConInformationGathering trois andAuditing Techniques 14 MinicomputerCharacMinicomputerCharacteristics andUses teristics andUses 15 Software Package SystemsDesign Topics Analysis Applications Program16 ComputerScheduling ming Languages 17 ComputerSimulation ComputerSimulation 18 ImprovingComputerCenter Productivity File Design Job Control Languages 19 Legal Aspectsof Computing IntroductoryStatistics 20 IntroductoryStatistics Statistical Decision Theory 21 List Processing RegressionAnalysis 22 Sorting Sampling Theory 23 Statistical Decision Theory List Processing 24 RegressionAnalysis Sorting SamplingTheory
42 MIS Quarterly/March 1980
Rank 1
Programmer
Introductory Computer andInformation Concepts 2 File Design 3 Applications Programming Languages Job Control Languages 4 Operating System Characteristics 5 Database Management Systems 6 SystemsDesign Topics 7 Telecommunications Concepts 8 i Hardware Characteristics 9 Human Relations in Systems Deve opment 10 Sorting 11 InformationGathering Techniques 12 MinicomputerCharacteristics and Uses 13 HumanFactors in EquipmentDesign and WorkLayout ComputerSecurity Controis andAuditing 14 Planningand Control of SystemsProjects 15 List Processing 16 ComputerScheduling 17 Software Package Analysis 18 Improving Computer CenterProductivity 19 Legal Aspectsof Computing ComputerSimulation 20 IntroductoryStatistics 21 Statistical Decision Theory 22 RegressionAnalysis SamplingTheory
Skill Requirements Survey
An additional point should be madeabout this conclusionconcerningthe growthin the needfor informationsystems personnel.It is probablytoo conservative. Our samplewastaken from large sophisticatedorganizationswith a longtradition of computer use.This fails to takeinto accounta very large segmentof computerusers, namely thosesmallercompanies whoare setting up their first computerized data processingsystemon a mini or micro-computer.Thesecompaniesare also generatingconsiderabledemand for appli4 cations programmers. Ourstudyindicatesthat the growthin demand for information systemspersonnelin the next few years will be largely in the traditional areas. COBOL basedsystemsstill seemto be dominant in mostinformation systemdepartments.Databasetechnologyis also becoming increasingly important.Theoneareaof decentralizationthat did showupin our results wasthe dataentry area whereuser departmentswill be bearing an increasingshareof the loadfor dataentry. It is too early to say whetherthis also means that eventually theseuserdepartments are likely to wantto developtheir ownapplicationsas well andgenerate a push for decentralized computing.Our guessis that they will not. Oursurveyof the skill areasvaluedby the informationsystemsmanagers showsthat there is a tendencyto compartmentalize the job areasand keeppeoplein their ownareasof specialty. The skill areasdesiredfor informationsystemsmanagersas shownin the surveyare a very demanding combination of technicalandmanagerial skills that probablygo beyondthe requirements placed on mostother types of middle managers.This ’~Cheney,P. andFuerst,W."Sharinga Computer Programmer: Onesolutionto Software Problems in SmallFirms,"Journalof Small Business Management, October 1979, Volume17, Number 4, pp. 43-48.
wouldmake the job difficult to fill, as mostof us havenotedin practice, and meansthat oncea suitable candidateis found,he tendsto be stuck in the job because replacements are hardto find. Theskill areasdemanded for managers are sufficiently different fromthosefor programmers and systemsanalystswhichsuggeststhat there is a definite andprobablygrowingneedfor information systemsprograms in collegesanduniversities with a managerialas well as a technical emphasis. In conclusion,this surveypoints upa strongand continuing demand for information systemspersonnelthat is not likely to be affected by new developments in technology.It is far to early to predict the demiseof the large computer system and the information systemsprofessional.
Aboutthe Authors Paul H. Cheneyreceived his doctorate in ManagementInformation Systems from the University of Minnesota. His current research includes the areas of auditing and controlling computer based accounting systems, decision support systems, and implementationstrategies for newinformation systems. Heis presently an associate professor in the Industrial Administration Department at Iowa State University. NormanR. Lyonsreceived his doctorate in Industrial Administration from CarnegieMellon University. Heis currently an associate professor at the Naval Post-graduate School in Monterey,California. He has published in ManagementScience, Communications of the ACM,Journal of Accountancy, and others. His majorresearchinterest is in the area of database management systems.
MIS Quarterly~March1980 43
