Information Systems Graduates' Evaluation of Their IS Curricula by Nancy W. Ashley and Thomas C. Padgett

Nancy W. Ashley nashley@sbf.bus.westga.edu   is an assistant professor and Thomas C. Padgett tpadgett@sbf.bus.westga.edu  is a professor of Business Administration at the Richards College of Business, State University of West Georgia.

INTRODUCTION

A major rationale for conducting the study reported on in this article is the fact that the rapid pace of change in the computer industry affects the curricula of  information system programs that prepare graduates for this volatile environment. Changes in three areas have been significant. The technologies themselves, the hardware and software, have become cheaper, faster, and smaller. The business environment has become more competitive, requiring that IS professionals must find uses of IS that will solve business problems in a cost-effective way. Lastly, the role of IS in organizations has shifted from from centralized mainframe processing to the support of distributed end users. [Lee] This means that not only have the technical skills required of IS personnel been changing, but there is an increasing need for communication skills and an understanding of the business environment.

Designers of undergraduate programs are trying to meet these skill needs, but not all are keeping up. As a consequence, businesses have lost money as a result of mistakes made by inadequately educated IS workers. There have been a couple of business responses to this situation. Some have increased their training budgets to make up for the gaps in academic programs [Maglitta]. Some have decreased their hiring of IS graduates [Maglitta], relying instead on IS professionals without IS degrees who have been put through the training programs [Khan]. Outsourcing is increasing [Maglitta].

Those trying to design IS curricula to produce workers who are adequately prepared face many challenges. Accreditation standards require that students take 50% of their course work outside of business schools. This leaves little room in which to acquire business and technical course work. Further, it is not clear which technical skills should fill the limited slots in the curriculum.

The conclusions of several studies of   the needs of businesses differ as to which skills are most important for the IS professional to possess. Several studies indicate that interpersonal and communication skills are more important than technical skills [Padgett, Leitheiser]. However, Richards and Pelley [Richards] place technical skills at the top of the list. Even if we accept that both interpersonal skills and technical skills are important [Nelson], there is still a decision to make about which technical skills are most important. One study gives skills with CASE tools high priority [Trauth]), while two others place CASE tools quite low on the list [Richards, Litecky]. Two studies suggest object-oriented languages are an important skill [Traitj, Leitheiser], two other studies find them of surprisingly little importance [Richards, Litecky]. Decision support systems and fourth generation languages are given top priority by Trauth, Farwell, and Lee [Trauth], while Richards and Pelley [Richards] assign these skills lower priority. It is intriguing to note that even in a recent study, COBOL programming  was still rated as a top priority skill [Richards]. This lack of consensus suggests that more study is needed.

The rapid changes in the IS work environment and the conflict between studies in identifying important skills make further study in this area vital. Although previous studies have been individually excellent, the difference in methodologies between the studies makes it difficult to make comparisons between their results. The study reported here is a longitudinal study, with the same methodology used on the same sample, six years apart. The research project poses several questions. What courses have IS graduates taken, which have been more valuable, which have been less valuable, and finally, have the answers to these questions changed over time? It is expected that this will be of interest to academics in the virtually continuous process of redesigning IS curricula, as well as to managers concerned about the academic preparation of their IS employees.

METHODOLOGY

The current survey replicated a study which was done in 1990. The same sample population and an updated version of the same survey instrument were used. In 1990 a survey sponsored by the Atlanta Chapter of the Association of Systems Management (ASM) was made of IS graduates from a random sampling of colleges and universities throughout the United States. Twenty-two schools agreed to participate; approximately 5000 questionnaires were mailed; and 1095 responses were received. The results of the survey were reported in Computer Personnel [Beise]. In 1994 a similar questionnaire, updated with new course topics, was sent to graduates of programs from two colleges. Questionnaires were mailed to 255 graduates, and 114 responses were received. The 1994 survey revealed sufficient differences from the 1990 survey to warrant a larger survey in 1996. In 1996, the survey was revised and updated to include some additional tools and topics, such as hypertext markup language, JAVA, Visual Basic, etc. A few courses were added to the lists in the general and business core as well, most notably mathematics. However, much of the questionnaire remained the same, so comparisons could be made to the earlier study. Where courses were added to the 1996 survey or dropped, it is indicated in the following tables with n/a (not applicable).

The 22 schools from the 1990 survey were contacted to request their participation in this survey. Questionnaires were sent to 3,758 graduates, and 472 were returned, for a response rate of 13%. After unusable questionnaires were deleted from the database, a total of 397 cases in the 1996 database remained.

In order to be able to compare the 1990 data to the 1996 data the responses of graduates of the same schools that were included in the 1996 were survey selected from the 1990 database. This resulted in there being 323 cases in the 1990 database. Respondents from the sample populations in 1990 and 1996 appear to be demographically similar. Both the 1996 and 1990 cases are 44% female and 56% male. Minorities account for 16% of the 1990 respondents and 17% of the 1996 respondents.

In the survey, IS graduates were asked to rate a list of courses commonly found in information systems programs. Courses were rated on a scale of 1 to 5 based on their perceived value in preparing a student for work in the IS area. The left side of the scale was labeled "of no value" and the right side was labeled "very valuable." The respondents were first asked to rate general, non-business courses, then courses common to all business majors and, last, the courses specific to the information systems area.

RESULTS and DISCUSSION

General, Non-Business Courses

The ratings in the 1996 and 1990 surveys of the general, non-business courses were very similar. The average rating for the general courses was slightly higher in 1996, but there are no significant differences between the years for any topic. English was the most valuable course both years. This stability between surveys confirms the need for course work to develop communication skills. Mathematics was regrettably omitted from the 1990 survey, but in 1996 had the second highest rating. As an IS curriculum is revised, it is easy to question the value of mathematics, a course which presents a stumbling block to some students. The importance given mathematics here suggests that rather than removing the hurdle, students with difficulty in this area should be encouraged to choose another field. Table 1 shows the results.

Business Courses

Business courses were rated nearly the same in 1996 as they were in 1990, with a few differences. The two top rated general business courses in both 1990 and 1996, introduction to information systems and business communication, switched places in ranking. Introduction to information systems moved to first place in 1996, and its rating, already high in 1990, was significantly higher in 1996. The increased value placed on this introductory course which integrates the various aspects of IS and relates them to the business organization may indicate employees' increased need to understand their organization's strategic mission and utilize various IS components to help meet that mission. Organizational theory dropped significantly in 1996, both in ratings and rankings. Results for all business courses can be seen in Table 2.

Information Systems Courses

Ratings of information systems courses are fairly consistent over the years, however, there are a few interesting differences. These ratings are shown in Table 3. The seven topics with the highest ratings in 1996 included five of the seven topics with highest ratings in 1990. Systems analysis and design had the highest ratings in both surveys. The small, but significant, decrease in the value of systems analysis and design in 1996 may indicate a trend away from the structured approaches taught in most academic courses. This decrease was unexpected. Windows was a new topic in the 1996 survey and had the second highest rating. This clearly indicates the importance of understanding the personal computer environment in which many end users operate. Database design, project management, file structure and management, and data structures retained their high ratings, suggesting that large mainframe database development projects are still a large part of many IS employees' work.

There are some differences, however, that may be the result of some fundamental changes in the information system graduate's environment. There was a significant increase in the ratings for database management. In an effort to find a reason for this, an analysis was conducted to relate the importance of database management to the type of position held. As expected, in both surveys database administrators rated this topic highly, 4.40 in 1990 and 4.45 in 1996. However, in the 1996 survey the average rating by all other respondents (everyone who was not a database administrator) for the topic of database management rose considerably, from 3.78 to 3.99. Perhaps as end users gain access to corporate databases, database management may be becoming more important for a broader spectrum of information systems workers.

There was a similar increase in the ratings for data communications. The ratings from those who work in data communications went up dramatically, from 3.00 in 1990 to 4.69 in 1996. As the percentage of the respondents working in data communications is larger than in 1990, but still rather small (3.5% of the respondents), this difference alone would not account for the large increase in the overall ratings. However, as was true in the previous case, in the 1996 survey the average rating by all other respondents (everyone who did not work in data communications) for the topic, data communications, also went up significantly, from 3.0 to 3.68. This is consistent with the increased usage of the local area networks to make corporate data accessible to more and more individuals within the organization, as well as usage of wide area networks that are a strategic link in today's multinational businesses.

The co-op or internship was a new topic for 1996, and had the eighth highest rating, confirming the importance of gaining real world experience found in a previous study [Richards].

The significant increase in ratings for word processing, hardware architecture, spreadsheets, and decision support systems could indicate an increased involvement with the end-user with respect to things like procuring and installing computers and software, training users, solving user problems, etc. This would be consistent with the increase found in this survey in the percentage of respondents working in end user support, which rose from 5.57% in 1990 to 8.56% in 1996.

Another group of topics showing substantial change are the programming languages. The C language showed a significant increase, perhaps because C is used to interface with many of the modern database and spreadsheet packages. The object-oriented programming languages, C++ and Smalltalk, rated higher than the any of the traditional languages except COBOL. This confirms the expectations that object-oriented languages are growing in importance. Ada, while still not highly rated, did show a significant increase. Assembly language and BASIC were rated significantly lower than in 1990, but they were not highly rated then either. The low ratings of assembly language would go along with the finding of significantly fewer systems programmers than in 1990, as systems programmers are the primary users of assembly languages.

COBOL, which had a fairly high rating in 1990, was down substantially in the 1996 survey. While we have been expecting the decline of COBOL for years, it is an unexpected result for this survey because a greater percentage of the respondents than in the 1990 survey identified themselves as third generation language (mostly COBOL) programmers. Analysis showed that the third generation programmers rated COBOL highly, both in 1990 and 1996. However, in 1990, the non-programmers also rated COBOL highly, 4.07. In 1996, the non-programmers only rated COBOL at 3.41. This suggests that the majority of IS personnel see COBOL as declining in importance. It has been expected for some time that the old, third generation languages would be replaced by the more modern languages, but COBOL has remained important because of the enormous amount of COBOL code that is in existence. The survey suggests that the perceived importance of COBOL may be in decline.

Importance of Various Skills In Work

Respondents were asked to rate, on a scale of 1-5, whether they strongly agreed or disagreed that various computing skills and environments were important to their work. In both the 1990 and 1996 surveys they were asked about the importance of third generation language skills (i.e. COBOL, Pascal) and fourth generation development skills such as spreadsheets and PC database management systems. In the the 1996 survey they were also asked about the importance of mainframe database processing skills, understanding of local area networks (LAN's), and understanding of the Internet.

Responses are shown in below in Table 4.

The findings both confirm expectations and surprise. Of the five skills, respondents found fourth generation development skills the most important. This is a change from 1990, when they found third generation programming skills more important than fourth generation skills. This confirms assertions frequently found in the literature that fourth generation development skills are increasingly important as end-users increase their use of corporate data.

It was not expected that respondents would find understanding the Internet least important. Excitement in the news media over the possibilities presented by the Internet and students' enthusiasm for Web page design and Internet surfing may cause us to over-value the new information highway. It seems that these respondents, working in the trenches of IS, value the ability to manage their own data environments with LAN's, databases, and programming languages more highly than the ability to understand the Internet outside their corporate firewalls. However, the survey reveals little, as the ratings of each of the skills averaged only somewhat more important than neutral.

Level of Emphasis on Different Kinds of Course Work, and Preparedness for Job Entry

When asked to what extent their degree program prepared them for their job entry, the average ratings in 1990 and 1996 were very comparable, 3.86 and 3.87 respectively. However, when asked about the emphasis placed on certain subject areas, there were some differences. Subjects were asked to rate whether the emphasis on various kinds of courses was too high, too low, or about right. This was rated on a scale of one to three, where one was too low and three was too high. 1996 found the respondents fairly content with the emphasis on programming. This was an improvement from 1990, when more indicated that the emphasis on programming was too high. This may suggest that the amount of programming in the curricula of these schools has been lowered since 1990 to match the lowered importance of programming skills in the workplace.

Respondents found serious and unexpected gaps in their education in other areas, however. Consider fourth generation application programming, a skill that had increased in importance between 1990 and 1996. While in 1990 they had indicated that the emphasis in the curriculum in this area had been about right, in 1996, with these skills increasing in importance, most indicated the emphasis was too low. A similar situation exists with telecommunications. In 1996, it was rated the second most important skill. But it was also indicated that there had not been enough emphasis on it in respondents' curricula. IS workers recognize the gaps in their curricula; academics need to recognize them as well. These rankings are shown below in Table 4.

SUMMARY AND CONCLUSIONS

The consistent ratings for many items in the 1990 and 1996 surveys adds to one's confidence in the validity of  the study; thus justifying taking differences between the two surveys seriously. The high ratings of English and business communications are consistent with the need expressed in the literature for IS professionals who can communicate. Mathematics was also recognized as providing a useful foundation for IS professionals. This suggests that this topic's place in the curriculum should not be diminished. That IS professionals recognize their need to understand the business environment is demonstrated by the high ratings of the business background courses, all of which rated higher than the non-business courses except for English and mathematics.

Among the information systems courses, there were several differences between the 1990 and 1996 surveys. Some of these differences support predictions from the literature, but others were unexpected. Among the expected differences were higher ratings in the 1996 survey for database management, data communications, word processing, hardware architecture, spreadsheets, decision support systems, artificial intelligence, and the C language, while COBOL, assembly and BASIC were rated lower. Object-oriented programming languages rated higher than non-object oriented languages. In 1996, fourth generation languages were considered more important than third generation languages, a reversal from 1990. Possible conclusions to be drawn from these results are that the information systems graduate is now more involved in dealing with end users, and that computers are increasing in importance in support of business decision making.

These conclusions support the predictions from the literature over the last several years. However, there were unexpected results from the survey as well. The slightly decreased rating of the systems analysis and design course was not expected. This may indicate that the more flexible approaches to analysis and design made possible by fourth generation software and required by rapidly changing business environments are beginning to make the structured approach taught in most academic courses less relevant. It may be that it is time for a change in focus in systems analysis and design.. However, the slight decrease found in this study should be taken as only a possible shift in emphasis; not as a total discrediting of the course, because the systems analysis and design course is still rated very highly. The finding that the Internet was rated the least important skill should remind academics to remain focused on the real work that IS professionals are likely to do and not get swept away by media and student enthusiasms. The indication that there was not enough emphasis in curricula on two important skills, fourth generation programming and telecommunications, should be of concern to academics. Well-prepared graduates -- preparing them and finding them -- are the objective of academics designing curricula and recruiters evaluating transcripts; so both groups may find the results of this survey relevant.

REFERENCES

Beise, C. M., T. C. Padgett, and F. J. Ganoe. "What Information Systems Graduates Are Really Doing: An Update," Computer Personnel, 13:2, July 1991, pp. 4-11.

Khan, M.B. and S. Kkalis. "MIS Professionals: Education and Performance," Information and Management, 19, 1990, pp. 249-255.

Lee, D.M.S., E.M. Trauth, and D. Farwell. "Critical Skills and Knowledge Requirements of IS Professionals: A Joint Academic/Industry Investigation," MIS Quarterly, 19:3, 1995, pp. 313-340.

Leitheiser, R.L. "MIS Skills for the 1990s: A Survey of MIS Managers' Perceptions," Journal of Management Information Systems, 9:1, Summer 1992, pp. 69-91.

Litecky, C. and K. Arnett. "Job Skills and Business Applications in Mid-range Computing Sites," Journal of Systems Management, 45:9, September 1994, pp. 36-39.

Maglitta, J. "IS Schools: Need Improvement," Computerworld, 30:8, February 19, 1996, pp. 78-83.

Nelson, R. "Educational Needs as Perceived by IS and End-user Personnel: A Survey of Knowledge and Skill Requirements," MIS Quarterly, 15:4, December 1991, pp. 503-525.

Padgett, T. C., C. Beise, and F. Ganoe. "Job Preparation of IS Graduates: Are They Rready for the Real World?" Journal of Systems Management, 42:8, August 1991, p. 17.

Richards, M. and L. Pelley. "The Ten Most Valuable Components of an Information Systems Education," Information and Management, 27:1, July 1994, pp. 59-68.

Trauth, E., D. Farwell, and D. Lee. "The IS Expectation Gap: Industry Expectations Versus Academic Preparation," MIS Quarterly, 17:3, September 1993, pp. 293-307.