ELSEVIER Decision Support Systems 14 (1995) 147-156 What does it take for successful executive information systems? R. Kelly Rainer, Jr. a,, Hugh J. Watson b a Department of Management, College of Business Building, Auburn Uni~'ersity, Auburn, AL 36849, USA b Department of Management, University of Georgia, Athens, GA 30602, USA Abstract Executive information systems (EISs) are high-risk systems as evidenced by the large percentage of companies that have reported EIS failures. To minimize this risk and exploit the potential of EISs, it is important to study the keys to successful EIS development and operation. Forty-eight persons representing three groups of EIS stakeholders (EIS executive users, EIS providers, and EIS vendors and consultants), noted 23 factors important to successful EIS development and 46 factors important to successful operational EIS. The findings denote factors that had not previously appeared in the literature, rank order the factors across all three stakeholder groups, and note differences in stakeholder rankings. The findings integrate and extend the EIS literature by contributing two stable sets of factors that are suitable for further research. Keywords: Executive information systems; System development; Management of information systems 1. Introduction A large n u m b e r of organizations have developed executive information systems (EISs). These systems are high-risk, as shown by recent surveys indicating that approximately sixty percent of the firms responding had experienced an EIS failure [32,33]. These failures reflect the myriad of managerial and technical issues that must be resolved for successful EIS development and operation. In spite of the problems, many organizations are making senior executives hands-on computer users for the first time. To increase the likelihood of effectively providing computer support for executives, it is important to know the factors that contribute to successful EIS development and operation. However, no single, comprehensive listing of factors has emerged in the literature. * Corresponding author. This study develops two listings: one that consists of factors contributing to successful EIS development and a second that consists of factors that contribute to successful EIS operation. Two listings are generated because the literature has used the term, implementation, to refer to aspects of information systems development and operation (see for example, [13,14,15,24,30]). An EIS is a computer-based information system that provides executives with easy access to internal and external information that is relevant to their critical success factors. EISs exhibit several pertinent characteristics (see for example, [32]): - are tailored to individual executive users; - extract, filter, compress, and track critical data; - provide online status access, trend analysis, exception reporting, and "drill-down" (drill-down allows the user to access supporting detail or data that underlie summarized data); 0167-9236/95/$09.50 © 1995 Elsevier Science B.V. All rights reserved SSDI 0 1 6 7 - 9 2 3 6 ( 9 4 ) 0 0 0 0 8 - G 148 R.K. Rainer, Jr., H.J. Watson/Decision Support Systems 14 (1995) 147-156 access and integrate a broad range of internal and external data; are user-friendly and require minimal or no training to use; - are used directly by executives without intermediaries; present graphical, tabular, a n d / o r textual information; - provide support for electronic communications. Insuring that an EIS provides these characteristics is a daunting process, involving continuous input from three EIS stakeholder groups (called constituencies): EIS executive users, EIS providers (i.e., persons responsible for building and maintaining the EIS), and often, EIS vendors and consultants. The multiple constituency approach has been used to investigate other types of information systems [1,6,9,34]. The three EIS stakeholder groups are useful in this study because the groups have differing agendas which potentially cause each to have different ideas on factors affecting successful EIS development and operation. Many authors have identified key variables purported to affect EIS success [2,3,5,7,8,10,11, 12,16,17,18,19,20,21,22,23,25,26,27,28,29,31]. Previous writings are helpful, but do have limitations. Various articles are based on a single firm's experiences. Even when multiple firms are studied, data have not been collected systematically enough to allow comparisons across organizations. No attempts have been made to distinguish between the initial development and the on-going operation phases. Previous research has not used a multiple constituency approach to study different types of EIS stakeholders. No comprehensive single listing of key variables has appeared in the literature. Finally, no attempts have been made to rank order or prioritize the keys to success. - Table 1 Demographic data on firms in the study Company Industry A Natural Gas B Government C Utility D Manufacturing E Healthcare F Communications G Manufacturing To overcome these limitations, this study utilized a diverse sample of interviewees, firms, and EISs. The use of multiple constituencies, different industries, and different EIS software, provides a rich, contextual basis for identifying, comparing, and contrasting the factors affecting successful EIS development and operation. Therefore, this research was designed to accomplish four goals: (1) elicit and rank order the keys to successful EIS development; (2) elicit and rank order the keys to successful on-going EIS operation; (3) compare the keys elicited in this study with those in the literature; (4) compare the keys across the three constituencies. 2. The study The authors interviewed forty-eight individuals who had been involved with the development and operation of one or more EISs. Each interviewee answered the same two questions with no initial probing: (1) What factors do you believe are most important in the successful development of an EIS? (2) What factors do you believe are most important in the successful on-going operation of an EIS? Interviewee comments were probed to gain contextual and anecdotal richness. Interviewees were not asked about the "success" of their EISs. That is, the study was designed to identify independent variables, not to evaluate a dependent variable. The total number of times a factor was mentioned served as a surrogate for the importance of that factor. This approach resulted from the belief that the interviewees noted factors they considered important because they received no prompting. Assets 3.1 billion 8.2 billion 3.7 billion 5.5 billion 2.0 billion 31.6 billion 2.1 billion EIS Software CADET Command Center-Pilot Command Center-Pilot Custom-built Custom-built Custom-built Commander EIS-Comshare R.K. Rainer, Jr., H,J. Watson/Decision Support Systems 14 (1995) 147-156 2.1. The sample The interviewees represent seven firms. Table 1 shows that the firms are relatively large and encompass a variety of industries. Each firm had an EIS that had been operational for more than one year at the time of this study. Table 1 also shows that the firms employed a variety of EIS software, including vendor-supplied and custombuilt software. The interviewees include the three EIS stakeholder groups and consisted of 18 executive EIS users, 18 EIS providers, and 12 EIS vendors and consultants. 3. Findings 3.1. Factors contribut&g to successful EIS development The interviewees made a total of 214 comments regarding 23 factors they considered ira- 149 portant for successful EIS development. Table 2 shows the number of comments made by each constituency for each factor and the total number of comments for each factor. The factors can be considered to be ranked overall in order of importance. Table 2 also shows the relative importance of each factor to each constituency. The factors from earlier studies are in bold and comprise 14 of the 23 factors contributing to successful E1S development (see Table 2). The Kendall coefficient of concordance for the factor rankings across the three constituencies is 0.9865 (p = 0.0042). This finding indicates that the constituencies rank all development factors similarly. (The null hypothesis for Kendall's coefficient of concordance is that there is no agreement in the factor rankings across the constituencies. Therefore, a p-value less than 0.05 indicates that the constituencies show significant agreement in their factor rankings.) Table 2 shows that the executive users mentioned an average of 3.67 factors, the providers Table 2 Successful EIS development factors Factors Exec Prov Executive sponsor [7,8,17,18,28] Define information requirements [2,8,10,17] Top m a n a g e m e n t support Manage data [2,7,18,28,29] Cost considerations Manage system spread and evolution [7,28] Manage user expectations [18] Deliver first version quickly [22] Manage organizational resistance [7,8,18,28] Link EIS and business objectives [7,8,17,18,22,28] Evolutionary development approach [8] EIS development, support team [8,17] Appropriate technology [7,8,22,28,29] Decide between vendor, custom software Start limited, not small [2] Develop as user system Use consultant Provide benefits statement Provider liaison with executives Appropriate I / S staff [7,8,28] Operating sponsor [7,8,17,28] E1S support executive staff Providers sell concept, then technology Total n u m b e r of c o m m e n t s by constituency Average n u m b e r of c o m m e n t s per respondent 11 11 11 6 4 5 1 3 2 3 2 1 1 3 12 12 7 7 6 5 6 2 6 2 5 4 2 2 1 1 * p < [).05 ** p < 1}.01 1 1 66 3.7 1 2 3 1 87 4.8 V/C Total X2 9 6 4 2 4 2 5 6 2 4 1 1 2 32 29 22 15 14 12 12 11 10 9 8 6 5 5 5 4 3 3 3 3 1 1 1 214 0.63 0.84 2.80 1.71 (/.67 0.59 6.(}7* 4.38 2.86 2.42 2.54 2.54 3.48 2.16 9.3{}** 1.70 9.60** 3.44 1.54 5.33 1.70 3.06 3.06 4 2 3 2 1 1 61 5.1 150 R.K. Rainer, Jr., H.J. Watson/Decision Support Systems 14 (1995) 147-156 mentioned an average of 4.83 factors, and the vendors/consultants an average of 5.08 factors. These findings suggest that the executive users have fewer responsibilities and duties or perhaps just a narrower perspective during the development process than the other two constituencies. The more important factors contributing to successful EIS development are discussed in three groups in the following section: the executive involvement and leadership factors, s t a f f / executive factors, and developmental issues. 3.2. Executive involvement and leadership factors Two factors pertain to executive involvement and leadership in the EIS development process: top management support and executive sponsor. Top management support, while not explicitly stated in the literature, can be construed as an extension of the factor, executive sponsor. Interviewees noted that one executive sponsor may not be enough to insure the successful development of an EIS because h e / s h e could retire, be fired, or move to another firm. Interviewees stated that broad, general support from top management is needed. One key responsibility of the executive sponsor is to gain this support. 3.3. Staff~executive factors The staff/executive factors pertain to the relationship between the EIS staff and the executives. These factors include determine information requirements, link EIS and business objectives, manage organizational resistance, and manage EIS user expectations. One executive commented on the process of determining executive information requirements, "This process (determining information requirements) is the most important part of the development p r o c e s s . . . i t is, in essence, a complete corporate planning meeting. The process would have been beneficial even if we had not developed our EIS." A consultant stated that the process of determining executive information requirements relates directly to linking the EIS to business objec- tives. " T h e quickest way to get an executive to realize the value of an EIS is to make sure the system directly addresses a business problem he or she has. If the EIS does not relate to a specific business problem, it is much more difficult to get their information requirements. What we did not want to have was a solution in search of a problem." Respondents supplied several examples of managing organizational resistance. A provider explained, "Resistance in our company took the form of foot-dragging on supplying data. It took a couple of phone calls from the executive sponsor to straighten the problem out." In another firm, " W e had one executive (Pete) who categorically had no intention of using the EIS. One morning, Pete's boss called him early and asked him about a problem down in Pete's organization. Pete did not yet know about the problem. He asked his boss how he knew about it and was told that the problem was clearly highlighted on the firm's EIS. Pete became a committed user that day." Interviewees point out that managing user expectations means that providers and v e n d o r s / consultants must deliver what they promise, when they promise it. An executive clarified, " I f they (the providers) had promised us too much and had failed to deliver, they would have been hung by their tongues." As one vendor stated, " W e always felt that we had to make good on our deliverables or we would not get a second chance. After all, executives have heard the same song and dance from their MIS people about new information systems before." 3.4. Developmental factors The developmental factors relate to the technical and behavioral issues of EIS development. R.K. Rainer, Jr., H.J. Watson/Decision Support Systems 14 (1995) 147-156 These factors include managing data, cost considerations, appropriate technology, decide between vendor and custom software, managing EIS spread and evolution, deliver the first EIS application quickly, evolutionary development approach, and EIS development and support team. Managing data involves a large effort. One provider noted, " D a t a comes from many sources into our EIS, both internal and external. The data must have consistent formats, but, more importantly, they must be accessible to us. We had to work very hard to get the data we needed to support our executive's information requirements." Conventional wisdom holds that cost considerations play a minor role in EIS development because executives typically mandate the process. However, interviewees note that cost considerations are an important consideration in EIS development. Some respondents related cost to justification of their EIS. Unlike costs, they noted that the benefits of E1Ss are difficult to quantify and that low-cost EISs are easier to justify than high-cost EISs. One executive illustrated, " N o n e of us (executives) felt politically comfortable supporting an expensive information system (the E1S) for so few people. If it had failed with a huge price tag, everyone felt that its champion might take a fall." Appropriate technology and decide between vendor and custom software were both related to cost considerations. One provider stated, " O u r firm adopted a L A N solution for our EIS in spite of the fact that we knew that a cooperative processing, mainframe-workstation solution would be more effective in the long run. We did it that way because of cost, and cost only. We also developed our own software because vendor software was much more expensive." The interviewees rank managing EIS spread and evolution highly. In particular, executives appear to be taking an active role in deciding who will have access to the EIS (spread). In one firm, access to the EIS is considered to be an "execu- 151 tive p e r k " and the system is limited to the firm's top fifteen executives. One provider in that firm commented, " W e all know who the 'big 15' are - they have the keys to the executive washroom, the executive dining room, and the EIS." Regarding additional capabilities (evolution), one provider noted, "Every one of our executive users has asked this question more than once. If my EIS can do this, why can't it also do this? Our users always want more." Delivering the first EIS application quickly helps to keep executive interest in the EIS project. The interviewees note that executives are best served by enabling them to have hands-on use of the first version of the EIS quickly. The evolutionary development approach and the EIS development and support team are closely related. As one provider said, " O u r EIS stayed in the evolutionary stage for several months while the executives settled on their information requirements. Our EIS team experienced a lot of frustration during this process, but they stayed close to the executives and made our EIS a success." Insights may be gained from factors that were not highly rated, such as appropriate I / S staff and operating sponsor. A possible explanation is that EIS development is considered a function of I / S departments and an 1 / S manager may assume the duties of the operating sponsor. Respondents may have taken these two factors for granted as part of the I / S d e p a r t m e n t ' s normal functions. 3.5. Significant differences across the constituencies Table 2 indicates the chi-square values for goodness-of-fit for each factor across the three constituencies. A p-value less than 0.05 implies that the constituencies differ significantly in relative n u m b e r of comments for a factor. The constituencies commented significantly differently on only three factors. Manage user 152 R.K. Rainer, Jr., H.J. Watson/Decision Support Systems 14 (1995) 147-156 expectations and use a consultant are expected. However, vendors/consultants rate start limited not small significantly higher than the other two constituencies. This factor implies that the EIS should initially address a specific, solvable problem for which the data are readily available, but Table 3 Successful operational EIS factors Factors Timely information [8,12,16,17,31] Improve efficiency [7,19,28,31] Accurate information [8,12,16,17,31] Relevant information [8,12,16,17,31] Ease of use [2,8,10,11,17,28,31] Status access [2,4,10,11] Improve communications [8,17,28] Minimal or no training [2,8,28] Adaptable interface [2,8] Adapt to changing information requirements Exception reporting [2,4,10] Convenient information Standard definitions in the enterprise [8,16] Access external data [2,4,8,28] Drill down [4,10,11] Multiple modes of presentation [8,28] Access soft, human data [8,28] Access internal data [4,8,28] Accountability for providers Color [2,8,16,17,28] Graphics [2,8,10,11,16,17,28] Fast response time [2,8,17,28] Easy to obtain hard copy Trend analysis [4,10,11] Improve operational control [7,19,28,31] Concise information [12] Electronic rolodex [28] Educate execs about IT [28] Decision support and analysis Low maintenance New, unique information Promote continuity in enterprise System reliability [2] Public screen Spreadsheets [28] Visual system maps Retrace feature Comprehensive information Access away from the office [28] Multiple methods to find information Higher quality decisions [12,19] Word processing [2,28] Integrate all types of data [8,10,28] Tickler file [28] Improve mental model of enterprise [28] Artificial intelligence Total number of comments by constituency Average number of comments per respondent * p < 0.05 ** p < 0.01 *** p < 0.001 Exec Prov 17 16 15 10 9 13 8 8 6 4 7 11 6 4 3 2 4 3 2 3 2 2 3 2 2 14 8 8 4 10 9 6 5 7 8 6 2 4 6 5 1 2 3 4 2 3 2 2 2 1 3 2 1 2 2 1 1 2 2 1 1 1 1 1 2 2 1 1 1 V/C 11 8 6 11 5 2 6 2 1 2 1 l 4 1 1 2 1 1 2 5 1 1 3 2 1 1 1 2 1 42 32 29 25 24 24 20 15 14 14 14 13 10 10 9 7 6 6 6 6 6 6 6 5 5 5 4 4 4 4 3 3 2 2 2 2 2 2 2 2 2 1 1 1 150 8.3 91 7.6 X2 1 2.54 8.00 6.42 14.05 0.56 8.89 1.19 2.75 3.50 3.36 3.50 17.34 4.88 4.88 1.87 4.74 3.30 2.29 3.30 2.16 2.16 0.25 2.16 0.07 0.97 16.74 3.27 1.46 0.36 5.18 3.44 1.54 3.48 0.69 0.69 3.48 3.48 1.39 1.39 1.39 5.26 1.70 1.70 1 3.06 1 1 427 3.06 3.06 1 1 186 10.3 Total * * *** ** *** *** R.K. Rainer, Jr., H.J. Watson/Decision Support Systems 14 (1995) 147-156 should be powerful enough to handle a broad spectrum of requirements. V e n d o r s / c o n s u l t a n t s are typically involved in the EIS development process on a contractual basis and may be more aware of setting specific deliverables. 3.6. Factors contributing to successful operational EIS The interviewees made a total of 427 comments regarding 46 factors they considered important for a successful operational EIS. Table 3 shows the n u m b e r of comments m a d e by each constituency for each factor and the total number of comments for that factor. The factors can be considered to be ranked overall in order of importance. Table 3 also demonstrates the relative importance of the factors to each constituency. The factors from earlier studies are in bold and comprise 32 of the 46 factors contributing to successful operational EIS (see Table 3). The Kendall coefficient of concordance is 0.8509 (p = 0.0052), indicating that the constituencies rank all operational factors similarly. Table 3 indicates that the executive users, as expected, mentioned an average of 10.33 factors, the EIS providers an average of 8.33 factors, and the v e n d o r s / c o n s u l t a n t s an average of 7.58 factors. The more important factors contributing to successful operational EISs are discussed in four groups in the following section: information quality factors, impact on executive work factors, EIS function factors, and EIS characteristics factors. 3. 7. Information quality factors Information quality factors pertain to the information delivered by the EIS to executive users. Executive interviewees rated timely information, accurate information, relevant information, and convenient information very highly. Executive comments illustrated the importance of these four factors. " I have to have timely information to react quickly to problems. Before I had my EIS, I depended on my sneaker n e t w o r k . . , you know, guys running up and down stairs with computer p r i n t o u t s . . , for information I needed on a flash basis. My EIS is faster." " I used to get my information from my staff 153 mostly and I trusted it. If my EIS had ever not provided me with accurate information, I would have shut it off." "1 must have information that applies to the problems I am facing at the moment. If I can't get it from my EIS, then the system wouldn't be giving me the relevant information that I need." " I can find information easier using my EIS. I don't have to look for hard copy in file cabinets. This kind of convenient information is what I want." 3.8. Impact on executive work factors The impact on executive work factors relate to the effects that EISs have on executive work. These factors include improve efficiency and im- prove communications. Many executives were impressed that their EISs improved their efficiency. As one executive noted, " I can do more work faster than I ever dreamed I could before I got my EIS. The system enables me to utilize my time much better." Improve communications was often linked to electronic mail. An executive commented, " M y EIS lets me easily access electronic mail. I can communicate with my people all over the world very efficiently. I do not have to play telephone tag and I don't worry about different time zones." 3. 9. EIS function factors EIS function factors pertain to EIS functions that produce usable information for a user. These factors include status access, exception reporting, standardize definitions across the firm, access external data, and drill down. Status access and exception reporting are closely related. An executive linked the two factors when he said, " T h e first things I look at (on my EIS) every morning are my status screens. I want to know how everything is going. I want to know if the 154 R.K. Rainer, Jr., H.J. Watson/Decision Support Systems 14 (1995) 147-156 baby is sick. Variances are flagged in color for me, so I can spot them easily. T h e n I can start asking questions." EISs have the capability to standardize data definitions across the firm. Respondents noted that the use of an EIS required that all terms used in the system m e a n the same thing. Further, they noted that the process of defining the data names was beneficial to everyone. One provider said, " W e are all on the same page now with respect to our d a t a . . , even our executives. This is something new in our organization.., and very welcome for us." One executive of an energy company commented on external data access, " W e have color weather maps on our EIS. If it's cold, we know we're selling gas. We also have hurricane tracking maps on the EIS. If a hurricane enters the Gulf (of Mexico), we can track it and see if we have to evacuate our offshore drilling rigs." An executive stated that drill down was helpful as he noted, " I want to look at status and exceptions. If I can see that, then I can ask appropriate questions and give appropriate directions. T h e n I want to see the detail so that I can get some of the flavor backing up the summaries." still can't respond to electronic mail very well unless I poke the keys with a pencil eraser." Minimal or no training is closely related to ease of use. As one executive explained, " I had neither the time nor the inclination to be trained on our EIS. I got the system and ten minutes later I was up and running. That's the way to go." An adaptable interface is another factor associated with ease of use. Adaptable interface means that the executives may use keyboards, mice, touchscreens, infrared devices, or any combination of these. Several respondents commented that keyboards could be employed with nontypists if only a few keystrokes were needed to access screens. Interviewees noted that executive information requirements change often and the EIS must be flexible enough to adapt to changing information requirements. As one provider noted, " O u r president was the chairman of the United Way campaign last year. In this capacity, he received an 850 page report each week on the progress of the campaign. Obviously, the report was too much for him to read, so I went to the I / S people at United Way, got them to give me the data electronically, and I put it on our EIS. Our president could look at the data in color and see it graphed by way of United Way divisions. After his year was over, we took it off the system." 3.10. EIS characteristics factors Unexpectedly, two groups of factors were not highly rated. The first group included access soft EIS characteristics factors relate to attributes of EISs that enable users to easily access and assimilate information from the system. These factors include ease of use, minimal or no training, adaptable interface, and adapt to changing data, access internal data, color, graphics, fast response time, and trend analysis. These factors information requirements. The majority of executives are not sophisticated computer users, therefore ease of use elicited many anecdotal comments. One executive said, " M y finger is about the size of two computer keys. I find that I cannot type at all. My EIS people quickly rigged me up with a mouse and I began to use the EIS much more regularly. I were undoubtedly taken for granted by the respondents. Probing as to why these factors were not highly mentioned, revealed that interviewees thought that all EISs had these capabilities and functions. The second group of factors included improve mental model of the firm, higher quality decisions, educate executives about IT, and improve operational control. Probing revealed consistent comments from many executives, " M y EIS doesn't give me anything more than I was getting already. I ' m just getting it faster with less effort. I think my EIS is increasing R.K. Rainer,Jr., H.J. Watson/Decision Support Systems 14 (1995) 147-156 my efficiency, but I still do my job the same way as before I got the computer. I view the system as just another tool at my disposal. I got here (present job) by knowing my business inside out and by making good decisions. My EIS helps only peripherally in these areas." 3.11. Significant differences across the constituencies Table 3 indicates the chi-square values for goodness-of-fit for each factor across the three constituencies. The constituencies commented significantly differently on six factors: improve efficiency, accurate information, relevant information, status access, convenient information, and concise information. Interestingly, executives note the first five factors significantly more than one or both of the other constituencies. 4. Discussion Clearly, the opinions of the executives are the most important. However, the study results indicate that the constituencies occasionally differ. In these cases, the providers and vendors and consultants must pay attention to the executives. If the providers, vendors, and consultants listen to the executives, why do some 60% of EISs fail? [32]. The study results indicate that the executives must take an active role in EIS development, an area in which they are, for the most part, unfamiliar. They must champion the system and devote time to clarify their information requirements. For an EIS to operationally successful, it must meet executive needs. First, executives want high quality information that they can obtain more quickly and with less effort than they could prior to having their EISs, thus improving their efficiency. Second, the executives want to access information quickly and easily, requiring EISs to be easy to use and require minimal training. Third, the executives want status access and exception reporting functions to enable them to assimilate information efficiently and effectively. Finally, they want the EIS to improve their communications. EIS providers, vendors, and consultants should pay particular attention to these factors to ensure that EISs meet executive needs. 155 5. Conclusion This study makes several contributions to the EIS literature. First, the study has compiled two thorough factor listings that are suitable for further research: the factors that contribute to successful EIS development and on-going operation. These two listings have richness and face validity resulting from the study of dissimilar organizations, dissimilar EIS software, dissimilar EIS hardware, and dissimilar personnel with EIS experience who represent different constituencies. Second, the study has confirmed and integrated the factors found in the disjoint, largely practitioner-oriented EIS literature into the two listings. In addition, new factors were added to the listings that had not appeared in the literature. Third, the study rank ordered the factors in each phase. In doing so, insights were gained not only from what respondents noted, but also from what respondents did not note. Fourth, the study demonstrates the value of separating the development and on-going operational phases of EIS. The factors contributing to success in each phase are clearly different. Even though some factors could be construed to be important in both phases (e.g., top management support, minimal training, etc.), the interviewees evidenced no confusion regarding the placement of factors in the two phases. Fifth, the study noted that the three constituencies were in agreement on the majority of factors in both phases. The study also gained insight from the factors on which the constituencies did not agree. A logical direction for further research would be to use these factor lists in a large-scale survey. Such a survey would increase generalizability while noting if the lists were complete. In addition, a factor analysis would indicate underlying dimensions, if any, in the factors. References [1] M. Alavi, An Assessment of the Concept of Decision Support Systems As Viewed By Senior Level Executives, MIS Quarterly, v. 6, no. 4, pp. 1-8 (December 1982). [2] W.C. Burkan, Making EIS Work, DSS 88 Transactions, Providence, R.I., The Institute of Management Sciences, pp. 121-136. 156 R.K. Rainer, Jr., H.J. Watson/Decision Support Systems 14 (1995) 147-156 [3] V.T. Dock, Executive Computer Use Is Doomed Without Five Key Properties, Data Management, pg. 27 (December 1985). [4] D. Friend, The Three Pillars of EIS, Information Center v. 3, no. 8, pp. 32-38 (August 1987). [5] D. Friend, Executive Information Systems: Successes, Failures, Insights, and Misconceptions, DSS 86 Transactions, Providence, R.I., The Institute of Management Sciences, pp. 35-40. [6] S. Hamilton and N.L. Chervany, Evaluating Information System Effectiveness-Part II: Comparing Evaluator Viewpoints, MIS Quarterly, pp. 79-86 (December 1981). [7] S.S. Hoffman, Executive Support Systems: MIS Faces A Dilemma, InformationWeek, pp. 36-41 (January 11, 1988). [8] G. Houdeshel and H.J. Watson, The Management Information and Decision Support (MIDS) System at Lockheed-Georgia, MIS Quarterly, v. 11, no. 1, pp. 127-140 (March 1987). [9] J.T. Hogue and H.J. Watson, Management's Role in the Approval and Administration of Decision Support Systems, Information and Management v. 8, no. 4, pp. 205-212 (April, 1985). [10] M.L. Jordan, Executive Information Systems Make Life Easy For the Lucky Few, Computerworld, pp. 51-58 (February 29, 1988). [11] K. Kukula, An Interface Even an Executive Could Love, Solutions: Special Issue, Unisys Corporation, One Unisys Place, Detroit, Michigan; 48232, #3008529-000032; 1989. [12] E. Lindholm, The View From the Top, Information Center, v. 4, no. 2, pg. 18 (February, 1988). [13] L. Long, Management Information Systems, (PrenticeHall, New Jersey, 1989). [14] H.C. Lucas, Unsuccessful Implementation: The Case of a Computer-Based Entry System, Decision Sciences, v. 9, pp. 68-79, (1978). [15] H.C. Lucas, Empirical Evidence For a Descriptive Model of Implementation, MIS Quarterly, v. 2, no. 2, pp. 27-41, (June 1978). [16] J. Main, At last, software CEOs can use. Fortune, March 13, 1989, 77-83. [17] B.C. McNurlin, Executive information systems. EDP Analyzer 25, 4 (April, 1987). [18] K. Melymuka, High-impact EIS. CIO 2, 5 (February 1989), 44. [19] M. O'Leary, At xerox, technology mirrors management styles. PC Week, June 30, 1987, 37. [20] T.J. O'Shea, Low-cost Approaches to Executive Information Systems, Journal of Information Systems Management, v. 6, no. 2, pp. 34-41, (Spring 1989). [21] A. Paller, Getting Started With Executive Information Systems, EDP Analyzer, pp. 13-14, (January, 1987). [22] A. Paller, Success and the EIS, Information Center, v. 4, no. 2, pp. 16-36 (February 1988). [23] A.H. Perlman, Building a Successful Executive Information System, Infosystems, v. 33, no. 1, pg. 66, (January 1986). [24] R.F. Powers and G.W. Dickson, MIS Project Management: Myths, Opinions, and Reality, California Management Review, v. 15, no. 3, pp. 147-155, (Spring, 1973). [25] R.H. Reck and J.R. Hall, Executive Information Systems: An Overview of Development, Journal of Information Systems Management, pp. 25-30, (Fall, 1986). [26] J.F. Rockart, Chief Executives Define Their Own Data Needs, Harvard Business Review, pp. 81-93, (MarchApril, 1979). [27] J.F. Rockart and M.E. Treacy, The CEO Goes On-line, Harvard Business Review, pp. 84-88, (January-February, 1982). [28] J.F. Rockart and D.W. De Long, Executive Support Systems: The Emergence of Top Management Computer Use, (Dow Jones-Irwin, 1988). [29] L. Runge, On the Executive's Desk, Information Center v. 4, no. 6, pg. 34 (June 1988). [30] J.A. Senn, Information Systems in Management, third edition, Wadsworth Publishing Co., California, 1987). [31] L. Volonino and G. Drinkard, Integrating EIS Into the Strategic Plan: A Case Study of Fisher-Price, DSS 89 Transactions, The Institute of Management Sciences, Providence, R.I., pp. 37-46. [32] H.J. Watson, R.K. Rainer, and C. Koh, Executive Information Systems: A Framework For Development and a Survey of Current Practices, MIS Quarterly, v. 15, no. 1, pp. 13-32 (March 1991). [33] H.J. Watson, R.K. Rainer, and M.N. Frolick, Executive Information Systems: An On-going Study of Current Practices, Journal of International Information Systems, v. 1, no. 2, pp. 37-56 (April 1992). [34] H.J. Watson, T. Boyd-Wilson, and S.R. Magal, The Evaluation of DSS Groups, Information and Management, v. 12, no. 2, pp. 79-86 (February 1987). Hugh J. Watson holds the C. Herman and Mary Virginia Terry Chair of Business Administration in the Terry College of Business at the University of Georgia. He is the author of 19 books and over 100 articles. His current research focuses on executive information systems. 1L Kelly Rainer, Jr. is an Associate Professor of Management (MIS) at Auburn University, Auburn, Alabama. He received his Ph.D. in MIS at the University of Georgia. He has published articles in MIS Quarterly, Journal of Management Information Systems, Information and Management, Decision Support Systems, as well as others. He is co-editor, with Hugh Watson and George Houdeshel, of Executive Information Systems (Wiley). His current research interests include executive information systems, end-user computing, structural equation modeling, and research methodological issues.