chapman.indd 50 Database Coverage for Research in Management Information Systems Karen Chapman and Paul Brothers Karen Chapman and Paul Brothers are both Business Reference Librarians in the Angelo Bruno Business Li- brary at the University of Alabama; e-mail: kchapman@bruno.cba.ua.edu and pbrother@bruno.cba.ua.edu. This study examines the database coverage of management informa- tion systems (MIS) journals and journal articles referenced by MIS researchers. Lists of titles and references were checked for coverage in twelve databases representing a variety of vendors: five multidisciplinary databases, four business databases, and three computer science or applied science databases. The best coverage of MIS journals is found in ABI/INFORM Global and Business Source Premier. The best cover- age of articles referenced by MIS scholars is offered by the same two databases, although Business Source Premier offers significantly more full text. Business Source Premier and Web of Science provide the best coverage for any pair of the databases. ibrarians are not the only ones who know the value of infor- mation; managers in business areas from accounting to manufacturing to marketing have known for years that the right information about products, processes, and customers can help them make be�er management deci- sions. Management information systems (MIS) is the field that studies the use of information in business—what informa- tion is needed, how to get it, and how to use it. MIS focuses on the application of computer systems to business processes and, more broadly, how technology can be used effectively and efficiently to achieve the goals of the organization.1 This includes study of the interface be- tween the user and the system and the effects on each.2 MIS scholars have tracked the devel- opment of their field of study as it has evolved into a discipline. It began to take shape during the 1960s, and MIS Quar- terly, the first journal in the discipline, was founded in 1977.3 The First International Conference on Information Systems was held in 1980, and the Association for Infor- mation Systems was established in 1994.4,5 More than a hundred universities have graduate or undergraduate programs in MIS.6 Today, MIS has “a distinct subject ma�er, a distinct research perspective, and a well-developed communication system that includes respected journals.”7 Because of the nature of the discipline and its relatively recent development, MIS scholars continue to draw on research in other disciplines. Richard L. Baskerville and Michael D. Myers provided a quick overview of the development of thought related to MIS’s reference disciplines; they point out that early MIS research draws heavily on engineering, computer science, cybernetic systems theory, mathematics, management science, and behavioral decision theory. More recent suggestions have included disciplines as diverse as ar- chitecture and anthropology.8 Iris Vessey, V. Ramesh, and Robert L. Glass analyzed articles from five MIS journals for the period 1995 through 1999. They classi- fied the reference disciplines of MIS into these categories: cognitive psychology, social and behavioral science, computer science, economics, information systems, management, management science, and other. They found that information sys- tems, management, and economics were the most common reference disciplines cited by authors of the articles in their study.9 Karen Chapman and Paul Brothers analyzed references from articles in three leading MIS journals for the period 2000 through 2002. A random sample of the references to journal articles was taken, and each reference was classified by subject according to the journal’s primary subject ma�er. Management was the most frequently referenced subject, followed closely by MIS itself; other subjects with notable numbers of references were “com- puting and technology” and “psychology and sociology.”10 The purpose of this study is to examine database coverage for the discipline of MIS. Because of the continuing reliance of MIS on research in other fields, the study has been divided into two parts. The first part focuses on coverage of journals that publish MIS research. The second part fo- cuses on coverage of the journal resources used by MIS scholars in producing their research. Literature Review Bibliographic databases have assumed a prominent position in library collections, and the library literature contains numer- ous studies that compare databases or analyze their content. A subset of these studies focuses on database coverage within a particular subject area. Steve Black analyzed the social sciences cover- age of four general databases by com- paring their title lists to the list of titles covered by the Social Sciences Citation Index, as reported in Journal Citation Reports. He compared the quality of the coverage by computing the average im- pact factor for the set of titles included in each database.11 Michael Colby studied the coverage of music periodicals in six general and one humanities database. His list of music periodicals was derived from the title coverage lists of two prominent music databases. His study compared coverage overall, by type of journal, by language, and by music subfield.1 2 Péter Jacsó developed a methodology for studying database coverage of journals by counting numbers of records in the databases and used it to examine coverage of forty-two library and information science journals in six databases for a specified time pe- riod.13 Chapman’s study compared cover- age of the field of finance in three general business databases. Instead of using title lists, she retrieved references from the articles in leading finance journals and searched for those items in the databases. Her results reported numbers of items not covered, indexed only, or available full text.14 William H. Walters and Esther I. Wilder examined the bibliographic da- tabase coverage of later-life migration, a multidisciplinary field of study. They dis- covered that multidisciplinary databases provided more comprehensive coverage of the field than subject-specific data- bases and that multiple databases must be searched to identify all the literature in the field.15 Although information systems is a topic that pervades much of the library literature, the academic discipline of management information systems has received li�le a�ention. Jinjg Hu, Lai C. Liu, Kai S. Koong, and Lillian Fok studied the holdings of MIS journals at the librar- ies of universities with MIS programs. The journals then were ranked by the number of libraries where each was held.16 In their study mentioned previously, Chapman and Brothers also analyzed the references gathered from articles in three leading MIS journals in order to profile the types of materials used by MIS researchers. They found that about 65 percent of ref- Database Coverage for Research in Management Information Systems 51 erences were journal articles and about 25 percent were books or book chapters. Further examination of a sample of the journal article references showed that 93 percent were published since 1980.17 MIS scholars have studied their own literature extensively, and several articles have a�empted to define the “top” jour- nals in the field. The results of six of these studies are compared in a table on the Journal Rankings page at ISWorld Net. A bibliography of additional studies also is given.18 The two most common method- ologies are surveys of academics and/or practitioners and citation analyses.19 MIS Journals Examined Many of the journals identified by MIS researchers as among the top-ranked titles in MIS have other fields as their primary focus. For example, Harvard Business Review, identified as among the top ten MIS journals in at least six stud- ies, is typically considered a management journal.20 To identify a list of journals that are important to MIS researchers, the authors turned to Kent A. Walstrom and Bill C. Hardgrave’s 2001 report on a survey of MIS faculty. The 364 survey respondents rated each of fi�y-one jour- nals in several ways. One question asked respondents to rate a journal’s importance as a publication outlet on a scale of one to four. Overall scores were computed, and journals were ranked accordingly. A natural break occurred in the scores a�er the twentieth title, and the authors selected the top twenty titles as the basis for investigation. 21 The next step in the process was to con- firm that all twenty titles were still active and to check for possible title changes. WorldCat and Ulrich’s International Periodicals Directory were consulted for each title. All the titles were active under the title shown in the list, with one minor exception. Table 1 shows the complete list of MIS journals used in the study, along with the ISSN, publisher and sponsor, year founded, and frequency for each title. Eleven of the titles are published by scholarly or professional associations, and three are sponsored by an asso- ciation, but published by a commercial publisher. Two more are published solely by a commercial publisher, and four are published by a university division or publisher. Founding dates range from 1922 for Harvard Business Review to 1991 for European Journal of Information Systems. For the most part, founding dates are evenly distributed across the decades of the 1950s through the 1990s. Databases Examined Elaine Wagner argued that librarians should be aware of subject coverage in da- tabases “beyond the obvious.”22 Because of this field’s close ties to management and computer science, the authors select- ed databases from these fields, as well as multidisciplinary databases, to examine for coverage of MIS publications. Table 2 shows the list of databases, with each database’s vendor or producer, number of titles covered, number of titles covered full text, and general subject category as assigned by the authors. The list includes five multidisciplinary databases, four business databases, and three computer science or applied science databases. As demonstrated in table 2, the authors selected databases from a range of vendors in the various categories. Major vendors are well represented in the list, with three titles from EBSCO Information Ser- vices, two from ProQuest Information and Learning, and three from Thomson Gale. The list is rounded out with both specialty database vendors, such as Dow Jones/Re- uters, and other vendors with products across multiple disciplines, such as H. W. Wilson. The sizes of the various databases also vary dramatically, from coverage of 441 titles for Computer Source to coverage of more than 9,000 for Business Source Premier. Finally, full-text coverage also varies widely. Although it is convenient to find the full text of an article available in a database, a good researcher uses indexes and databases that lead to the best results, not just those that offer full text. 52 College & Research Libraries January 2006 TABLE 1 Journals Included in the Study Journal Title ISSN Publisher/Sponsor Year Founded Frequency Academy of Management Journal 0001-4273 Academy of Management 1958 Bimonthly Academy of Management Review 0363-7425 Academy of Management 1976 Quarterly ACM Computing Surveys 0360-0300 Association for Computing Machinery 1969 Quarterly ACM Transactions on Database Systems 0362-5915 Association for Computing Machinery 1976 Quarterly ACM Transactions on Information Systems 1046-8188 Association for Computing Machinery 1983 Quarterly Administrative Science Quarterly 0001-8392 Johnson Graduate School of Management, Cornell University 1956 Quarterly Communications of the ACM 0001-0782 Association for Computing Machinery 1958 Monthly Decision Sciences 0011-7315 Blackwell/Decision Sciences Institute 1970 Quarterly Decision Support Systems 0167-9236 Elsevier 1985 8/year European Journal of Information Systems 0960-085X Palgrave Macmillan/ Operational Research Society 1991 Quarterly Harvard Business Review 0017-8012 Harvard Business School Publishing 1922 Monthly IEEE Transactions on Knowledge and Data Engineering 1041-4347 Institute of Electrical and Electronics Engineers (IEEE) 1989 Monthly IEEE Transactions on Software Engineering 0098-5589 IEEE/Computer Society 1975 Monthly Information & Management 0378-7206 Elsevier/International Federation for Information Processing (AT) Applied Information Group 1968 8/year Information Systems Research 1047-7047 INFORMS 1990 Quarterly Journal of Management Information Systems 0742-1222 M. E. Sharpe, Inc. 1984 Quarterly Management Science 0025-1909 INFORMS 1954 Monthly MIS Quarterly 0276-7783 Management Information Systems Research Center, University of Minnesota 1977 Quarterly MIT Sloan Management Review 1532-9194 Massachusetts Institute of Technology/Sloan Management Review Association 1960 Quarterly Organization Science 1047-7039 INFORMS 1990 Quarterly Database Coverage for Research in Management Information Systems 53 Database Coverage of MIS Research The list of twenty journals that are impor- tant publication outlets for MIS research was first checked online for coverage in each database. Each title was coded as not covered, indexed only, or full text avail- able. The title was coded as full text if any issues were available full text, regardless of their date. Table 3 shows the results by journal. The most frequently covered journal on the list is IEEE Transactions on So�ware Engineering, which is indexed in eleven of the twelve databases, but not full text in any. The journal with the least cov- erage is Academy of Management Review, which is covered in four of the databases. On average, each title is covered in seven of the twelve databases. Next the databases were ranked by number of titles covered, followed by number of titles with full text. Table 4 shows the results. Three databases cover all twenty journals: Business Source Pre- mier, Expanded Academic Index, and Web of Science. Business Source Premier offers full text for fourteen journals and Expanded Academic Index for thirteen, so they are ranked first and second. ABI/IN- FORM Global and Business & Company Resource Center have good coverage, with eighteen and seventeen titles cov- ered respectively. LexisNexis Academic has by far the poorest coverage, not cover- ing any of the titles on the list. Simple title counts can be informative, but they do not capture the complexity of database coverage. Other factors also are important: Is the title indexed selec- tively or thoroughly? When did coverage begin? Are there gaps? Is coverage still TABLE 2 Databases Included in the Study Database Name Vendor/Producer Titles Covered (no.) Titles Covered Full Text (no.) Subject ABI/INFORM Global ProQuest Information and Learning 1,600 700 Business Academic Search Premier EBSCO Information Services 8,200 4,681 Multidisciplinary Applied Science & Technology Full Text H. W. Wilson 751 174 Applied Science Business & Company Resource Center Thomson Gale 3,689 2,825 Business Business Source Premier EBSCO Information Services 9,028 8,167 Business Computer Database Thomson Gale 665 361 Computer Science Computer Source EBSCO Information Services 441 315 Computer Science Expanded Academic ASAP Thomson Gale 3,723 2,203 Multidisciplinary Factiva Dow Jones/Reuters 9,000 n.a. Business LexisNexis Academic LexisNexis 5,600 n.a. Multidisciplinary ProQuest Research Library ProQuest Information and Learning 3,000 1,900 Multidisciplinary Web of Science Thomson ISI 8,700 0 Multidisciplinary 54 College & Research Libraries January 2006 active? Jacsó pointed out these and other important considerations in his article on database coverage of library and informa- tion science journals. His methodology involved counting numbers of records for the journals within each database and comparing the results across the set of databases. He looked at four com- binations of journals and date ranges: all journals for all years and for recent years, then top journals for all years and for recent years.23 The authors adopted this approach. The number of hits in the database was recorded for each journal, both overall and from 2000 to the pres- ent. Results for the individual journals were added together to arrive at a total for the database. Note that this is not an entirely accurate method of gauging level of coverage; in particular, duplicate records for the same article can inflate the results. For example, when databases have added full-text backfiles, the records containing the full text may have simply been added to the database, even though records were already in the database containing the bibliographic information for those articles. The first line in table 5 gives the results for all twenty titles for their full existence. The authors compiled previous title in- formation for the journals from Ulrich’s International Periodicals Directory da- tabase and WorldCat and included the previous titles in their counts. The second line gives the data as a percentage of the TABLE 3 Journal Coverage in Selected Databases (N = 12) Journal Title Indexed Only (no.) Full Text (no.) Total Coverage (no.) Academy of Management Journal 2 5 7 Academy of Management Review 2 2 4 ACM Computing Surveys 3 6 9 ACM Transactions on Database Systems 6 4 10 ACM Transactions on Information Systems 7 3 10 Administrative Science Quarterly 2 5 7 Communications of the ACM 4 5 9 Decision Sciences 3 3 6 Decision Support Systems 7 0 7 European Journal of Information Systems 5 2 7 Harvard Business Review 3 3 6 IEEE Transactions on Knowledge and Data Engineering 6 0 6 IEEE Transactions on Software Engineering 11 0 11 Information & Management 8 0 8 Information Systems Research 1 5 6 Journal of Management Information Systems 3 4 7 Management Science 1 4 5 MIS Quarterly 2 7 9 MIT Sloan Management Review 1 5 6 Organization Science 1 4 5 Database Coverage for Research in Management Information Systems 55 total count across all databases; this line can be used to gauge relative performance among databases more easily than the raw numbers. The rankings shown in the third line are provided simply to assist the reader in quickly identifying the highest and lowest performers. The two databases with the most coverage are ABI/INFORM Global and Business Source Premier. They offer significantly more coverage than the third-ranked database, Web of Science. Overall, the business databases outper- formed the multidisciplinary databases, which in turn outperformed the computer science–related databases. Factiva, a busi- ness database, is a noted exception with its poor performance. One explanation of the results is the presence of extensive backfiles in both ABI/INFORM Global and Business Source Premier. How do the databases perform when looking only at more recent cover- age? To answer that question, the authors used the same methodology and limited the results to articles from 2000 or later. The results are seen in the next section of table 5. Looking only at more recent articles, there is a slight shi� in the re- sults. ABI/INFORM Global drops to the fourth position, behind Business Source Premier, Business & Company Resource Center, and Web of Science, in that order. However, these four databases and the fifth-ranked database, Expanded Aca- demic Index, are clustered fairly closely. With the effects of the backfiles removed, the results are distributed more evenly across the databases. The overall pa�ern remains, with business databases outper- forming the other categories. The journals under consideration, though being those of most importance as publication outlets, are not equal in importance. However, the investigation so far has treated them equally. A data- base that covers only the less important journals may have outscored a database that covers only the more important journals if it covers them more intensively or there are simply more articles in the less important journals. To explore this situation, the totals were computed again using only the five highest-rated journals on the list: MIS Quarterly, Information Systems Research, Communications of the ACM, Journal of Management Information Systems, and Management Science. These results appear in the lower section of TABLE 4 Database Coverage of Selected Journals (N = 20) Database Name Journals Indexed Only (no.) Journals Full Text (no.) Total Covered (no.) Rank ABI/INFORM Global 6 12 18 4 Academic Search Premier 7 0 7 10 (tie) Applied Science & Technology Full Text 7 0 7 10 (tie) Business & Company Resource Center 4 13 17 5 Business Source Premier 6 14 20 1 Computer Database 2 5 7 9 Computer Source 7 3 10 8 Expanded Academic Index 7 13 20 2 Factiva 8 3 11 7 LexisNexis Academic 0 0 0 12 ProQuest Research Library 4 7 11 6 Web of Science 20 0 20 3 56 College & Research Libraries January 2006 table 5. When consid- ering coverage for all years, ABI/INFORM Global provides the greatest amount of coverage by a wide margin, with Busi- ness Source Premier following. Computer Source moves well forward in the rank- ings to replace Web of Science in the third position. Academic Search Premier also shows a higher rank- ing. When counting only articles from 2000 or later, ABI/ INFORM Global con- tinues to provide the greatest amount of coverage, but by a smaller margin. Busi- ness Source Premier and Web of Science are second and third. Expanded Academic Index achieves its highest ranking in this category. Over- all, results are dis- tributed more evenly across the databases. For both time peri- ods, two business databases provide the best coverage. In general, business databases provide the greatest amount of coverage, followed by multidisciplinary databases and then t h e c o m p u t e r s c i - ence–related data- bases. An exception is Computer Source, which advanced in the rankings for both time periods. TA B L E 5 D at ab as e C ov er ag e of S el ec te d M IS J ou rn al s A B I A ca dS P A ST B SP B C R C C om pD C om pS E xp A I F ac ti va L N A R L C W oS A ll T it le s A ll ye ar s 48 ,7 72 13 ,3 18 7, 24 1 45 ,1 28 24 ,2 44 9, 20 6 14 ,4 04 19 ,9 92 4, 70 4 0 22 ,8 78 26 ,6 85 Sh ar e of to ta l 20 .6 % 5. 6% 3. 1% 19 .1 % 10 .2 % 3. 9% 6. 1% 8. 5% 2. 0% 0. 0% 9. 7% 11 .3 % R an k 1 8 10 2 4 9 7 6 11 12 5 3 20 00 –p re se nt 7, 63 8 3, 20 3 2, 21 4 8, 49 7 8, 30 5 2, 52 8 3, 58 2 7, 21 1 3, 28 0 0 5, 36 9 8, 26 2 Sh ar e of to ta l 12 .7 % 5. 3% 3. 7% 14 .1 % 13 .8 % 4. 2% 6. 0% 12 .0 % 5. 5% 0. 0% 8. 9% 13 .7 % R an k 4 9 11 1 2 10 7 5 8 12 6 3 To p 5 T it le s A ll ye ar s 22 ,6 21 8, 91 7 3, 38 7 16 ,0 26 8, 02 4 5, 75 6 9, 99 3 7, 54 3 1, 26 2 0 6, 41 5 9, 04 0 Sh ar e of to ta l 22 .9 % 9. 0% 3. 4% 16 .2 % 8. 1% 5. 8% 10 .1 % 7. 6% 1. 3% 0. 0% 6. 5% 9. 1% R an k 1 5 10 2 6 9 3 7 11 12 8 4 20 00 –p re se nt 3, 12 2 2, 01 4 77 0 3, 00 6 2, 35 5 1, 67 8 2, 35 5 2, 56 1 1, 15 8 0 1, 98 0 2, 90 2 Sh ar e of to ta l 13 .1 % 8. 4% 3. 2% 12 .6 % 9. 9% 7. 0% 9. 9% 10 .7 % 4. 8% 0. 0% 8. 3% 12 .1 % R an k 1 7 11 2 5 (t ie ) 9 5 (t ie ) 4 10 12 8 3 Database Coverage for Research in Management Information Systems 57 Database Coverage of Resources for MIS Research The second part of the study focuses on the resources needed for MIS research. As described earlier, MIS scholars potentially may draw from several other disciplines in conducting their research. To derive a more accurate picture of which databases provide good support for MIS research, the authors elected to collect a sample of actual references used by MIS schol- ars and then test it against the selected databases. This required a method for identifying MIS articles within non-MIS journals or an approach for identifying journals that primarily publish articles on MIS. To address this problem, the au- thors turned again to the Walstrom and Hardgrave survey. Another question in the survey asked respondents to indicate whether each journal published primarily information systems research. A positive response was coded as one and a nega- tive response as zero, and scores were computed for each title. Three journals received unanimously positive responses (overall score of 1.00): Information Systems Research, Journal of Management Informa- tion Systems, and MIS Quarterly.24 The authors collected the references from every article in each issue of the three journals for the period 2000 through 2002. All items with references from those issues were included. The references to journal articles then were extracted to form a list of 9,117 citations. A random sample of 369 citations was taken for a confidence level of 95 percent and a confidence interval of plus or minus 5 percent. The sample was sorted by journal title and date, and then each citation was searched in each database. The citations were coded according to their status in the database (not included, indexing only, or full text). The 369 citations in the sample represent 148 different journal titles; 98 of the titles were referenced only once. Table 6 shows a list of the most frequently referenced titles in the sample. The most frequently referenced journal was MIS Quarterly, with thirty-six citations; Information Systems Research was second with twenty-one citations and Journal of Management Information Systems was third with eighteen. Self-citation bias could account for these three titles being the most cited. The next most-cited titles were Management Science with sixteen references, Communications of the ACM with thirteen, and Harvard Business Review with ten. Seventeen of the twenty journals listed in table 1 appeared in the sample under the current title or a previous title. The oldest article cited was published in 1951 and the most recent in 2002; the me- dian publication date for the sample was 1992 (i.e., half the articles were published in the period 1951 to 1992 and half were published from 1992 to 2002). Table 7 shows the coverage of the sample articles in each of the databases. Business Source Premier provides the most comprehensive coverage, with in- dexing or full text of 69.4 percent of the sample articles. ABI/INFORM Global is a close second, with 68.0 percent, and Web of Science is third with 58.8 percent. These are the only databases that cover over half the sample. The other business databases performed less well, with Business & Company Resource Center covering 46.1 percent and Factiva only TABLE 6 Journals in the Sample Referenced 10 or More Times Journal Title Number of Times Referenced MIS Quarterly 36 Information Systems Research 21 Journal of Management Information Systems 18 Management Science 16 Communications of the ACM 13 Harvard Business Review 10 58 College & Research Libraries January 2006 9.8 percent of the sample. Among the multidisciplinary databases, Expanded Academic Index covered 49.6 percent, ProQuest Research Library covered 31.7 percent, and Academic Search Premier covered 19.0 percent. Business Source Premier and Academic Search Premier are both produced by EBSCO Information Services, so it is surprising to see that the multidisciplinary database covered signif- icantly less of the sample than most of the other multidisciplinary databases. From that group, only LexisNexis Academic had poorer results, with only 1.1 percent coverage. The computer science databases offered limited coverage, ranging from Computer Source with 27.9 percent cov- erage to Applied Science and Technology with 10.3 percent coverage. The results for full-text coverage also are shown in table 7. In this category, Busi- ness Source Premier is dominant, offering full text for just over half of the articles in the sample. ABI/INFORM Global is second with 39.3 percent full text. The remaining databases have rather poor coverage, ranging from Computer Source with 20.6 percent down to LexisNexis Aca- demic with 1.1 percent, followed by the two databases that do not offer full text. Forty-seven of the 369 sample articles, or 12.7 percent, were not found in any of the databases. The dates of these articles ranged from 1951 through 2000. Most of the articles were distributed fairly evenly over the period 1970 through 2000, although there was a concentration of articles from the early 1980s. The median date was 1983 (i.e., half the articles were dated from 1983 or before and half were dated 1983 or later). The forty-seven articles appeared in thirty-nine distinct journals. About a third were psychology journals, and several journals were from management or business fields and from information systems or computer-related fields. The remaining journals repre- sented a variety of subjects, including education and law. As table 7 indicates, the best perform- ing database, Business Source Premier, contained coverage for less than three- quarters of the citations. However, 87.3 percent of the sample articles were found in at least one of the databases. As a next step, the authors looked at pairings of TABLE 7 Database Coverage of References by MIS Researchers Database Full Text (%) Rank Total Coverage (indexed or full text) (%) Rank Business Source Premier 51.2 1 69.4 1 ABI/INFORM Global 39.3 2 68.0 2 Web of Science 0.0 11 (tie) 58.8 3 Expanded Academic Index 15.7 6 49.6 4 Business & Company Resource Center 16.8 5 46.1 5 ProQuest Research Library 17.6 4 31.7 6 Computer Source 20.6 3 27.9 7 Academic Search Premier 12.5 7 19.0 8 Computer Database 8.9 8 18.2 9 Applied Science & Technology 0.0 11 (tie) 10.3 10 Factiva 7.3 9 9.8 11 LexisNexis Academic 1.1 10 1.1 12 Database Coverage for Research in Management Information Systems 59 databases to determine w h i c h t wo c o m b i n e d wo u l d g i ve t h e m o s t complete coverage of the articles in the sample. The results are shown in table 8. The cell at the intersection of a row and a column gives the amount of coverage for that com- bination of databases. To assist the reader, the cells on the diagonal have a gray background; these cells represent a pairing of each database with itself, which provides a baseline for comparing to what extent coverage is improved when combin- ing with other databases. The highest percentage in each row is presented in bold and italics, indicat- ing which database can be combined with the database for that row to provide the best results. Finally, a box highlights the highest percentage in the table, indicating the combination of databases that provides the best coverage of the sample articles. According to table 8, the pair of databases pro- viding the most compre- hensive coverage of the sample articles is Busi- ness Source Premier and Web of Science. These two databases provide coverage of 82.4 percent of the sample. The second best combination is Busi- ness Source Premier and ABI/INFORM Global, for coverage of 78.6 percent. In fact, most of the data- bases among those tested TA B L E 8 C ov er ag e of S am pl e by P ai rs o f D at ab as es D at ab as e A B I A SP A ST B C R C B SP C D C S E A F ac L N A P R L W oS A B I/ IN FO R M G lo ba l ( A B I) 68 .0 % 74 .3 % 71 .3 % 69 .1 % 78 .6 % 71 .3 % 75 .3 % 72 .4 % 68 .6 % 68 .0 % 70 .2 % 77 .2 % A ca de m ic S ea rc h Pr em ie r ( A SP ) 74 .3 % 19 .0 % 24 .1 % 55 .0 % 70 .7 % 23 .6 % 31 .2 % 55 .3 % 25 .5 % 19 .5 % 45 .5 % 63 .4 % A pp lie d Sc ie nc e & T ec hn ol og y (A ST ) 71 .3 % 24 .1 % 10 .3 % 49 .9 % 72 .6 % 20 .6 % 32 .8 % 52 .6 % 18 .7 % 11 .4 % 37 .9 % 59 .6 % B us in es s & C om pa ny R es ou rc e C en te r (B C R C ) 69 .1 % 55 .0 % 49 .9 % 46 .1 % 74 .8 % 51 .5 % 64 .2 % 54 .5 % 48 .0 % 46 .3 % 55 .6 % 67 .5 % B us in es s So ur ce P re m ie r ( B SP ) 78 .6 % 70 .7 % 72 .6 % 74 .8 % 69 .4 % 72 .4 % 69 .4 % 76 .7 % 69 .4 % 69 .9 % 74 .8 % 82 .4 % C om pu te r D at ab as e (C D ) 71 .3 % 23 .6 % 20 .6 % 51 .5 % 72 .4 % 18 .2 % 32 .5 % 52 .8 % 24 .7 % 18 .4 % 44 .4 % 61 .5 % C om pu te r S ou rc e (C S) 75 .3 % 31 .2 % 32 .8 % 64 .2 % 69 .4 % 32 .5 % 27 .9 % 64 .8 % 34 .4 % 28 .7 % 51 .2 % 69 .1 % E xp an de d A ca de m ic In de x (E A ) 72 .4 % 55 .3 % 52 .6 % 54 .5 % 76 .7 % 52 .8 % 64 .8 % 49 .6 % 49 .9 % 49 .9 % 58 .3 % 66 .7 % Fa ct iv a (F ac ) 68 .6 % 25 .5 % 18 .7 % 48 .0 % 69 .4 % 24 .7 % 34 .4 % 49 .9 % 9. 8% 10 .3 % 36 .9 % 61 .0 % L ex is N ex is A ca de m ic (L N A ) 68 .0 % 19 .5 % 11 .4 % 46 .3 % 69 .9 % 18 .4 % 28 .7 % 49 .9 % 10 .3 % 1. 1% 31 .7 % 59 .9 % Pr oQ ue st R es ea rc h L ib ra ry (P R L ) 70 .2 % 45 .5 % 37 .9 % 55 .6 % 74 .8 % 44 .4 % 51 .2 % 58 .3 % 36 .9 % 31 .7 % 31 .7 % 69 .1 % W eb o f S ci en ce (W oS ) 77 .2 % 63 .4 % 59 .6 % 67 .5 % 82 .4 % 61 .5 % 69 .1 % 66 .7 % 61 .0 % 59 .9 % 69 .1 % 58 .8 % 60 College & Research Libraries January 2006 show the best results when combined with Business Source Premier. The two exceptions, Academic Search Premier and Computer Source, are databases from the same vendor as Business Source Premier. For those databases, be�er results are obtained when combined with a database from a different vendor. Databases from a different vendor are likely to include coverage of publishers and sources not available to the first vendor. Discussion For coverage of journals that publish MIS research, ABI/INFORM Global and Business Source Premier are the best data- bases. These two databases rank first and second in coverage of both the full-title list and the top five journals for all years and for the top five journals for articles from 2000 to present. ABI/INFORM Global’s ranking drops somewhat when consider- ing coverage of the full list for articles from 2000 to the present. For coverage of articles referenced by MIS researchers, their posi- tions are reversed, with Business Source Premier outperforming ABI/INFORM Global by a small margin. However, Busi- ness Source Premier offers significantly more full text of the articles in the sample than any other databases tested. When combining the coverage of two databases, the best results are obtained by searching Business Source Premier and Web of Science. Almost all the databases gave their best results when combined with Business Source Premier. As dis- cussed earlier, the two exceptions are databases from the same vendor. Even ABI/INFORM Global shows its best per- formance when combined with Business Source Premier, indicating that, despite their similar performance in the rankings of individual databases, there are differ- ences in their coverage. It is interesting to note that the data- base providing the most full-text coverage of articles in the sample, Business Source Premier, offered full text for only about half the articles. The pair of databases offering best overall coverage, Business Source Premier and Web of Science, would not increase that figure because Web of Science does not include any full text. Fortunately, tools are becom- ing more widely available that allow a library to configure databases so users can link from a bibliographic record in one vendor’s database to a record with full text in another vendor’s database or at a publisher’s Web site. Thus, the amount of electronic full text actually available to the researcher will be determined by the range of databases offered and the online journal holdings of the library. Conclusion The purpose of this study is to assess database coverage of both the published output of MIS researchers and the jour- nal resources they use to support their research. Results show that indexing of both types of materials is fairly extensive. The findings also indicate that two data- bases, ABI/INFORM Global and Business Source Premier, provide the most compre- hensive coverage for each. In addition, Business Source Premier offers full text for the greatest number of MIS journals and provides the greatest amount of full text for journal articles referenced by MIS researchers. MIS researchers currently draw re- sources from many subject areas, and many of the publications considered important outlets for publication of MIS research are journals in other fields, such as management and computer science. As MIS continues to mature as a discipline, it will be interesting to observe how MIS research evolves. Databases are continu- ously changing as well, so future studies may show be�er coverage of this complex field by nonbusiness databases. Notes 1. William R. King, “Management Information Systems,” in Encyclopedia of Information Systems, Database Coverage for Research in Management Information Systems 61 vol. 3, ed. Hossein Bidgoli, 115 (New York: Academic Pr., 2003). 2. A. S. Lee, “MIS Quarterly’s Editorial Policies and Practices,” MIS Quarterly 25, no. 1 (2001): iii. 3. Richard O. Mason and James L. McKenney, “Developing an Historical Tradition in MIS Research,” MIS Quarterly 21, no. 3 (1997): 258. 4. Ibid., 259. 5. “Association for Information Systems: About AIS,” Association for Information Systems, 2001. [Accessed 8 November 2004] . Available online at www.aisnet.org/admin.shtml. 6. ISWorld Net Home Page, Association for Information Systems, 2004. [Accessed 8 November 2004]. Available online at www.isworld.org. 7. Richard L. Baskerville and Michael D. Myers, “Information Systems as a Reference Disci- pline,” MIS Quarterly 26, no. 1 (2002): 3. 8. Ibid., 1–14. 9. Iris Vessey, V. Ramesh, and Robert L. Glass, “Research in Information Systems: An Empirical Study of Diversity in the Discipline and the Journals,” Journal of Management Information Systems 19, no. 2 (2002): 129–74. 10. Karen Chapman and Paul Brothers, “A Reference Study of Leading MIS Journals: Identify- ing Formats, Frequency, and Journal Subjects,” Journal of Business & Finance Librarianship, 10, no.2 (2004):17–29. 11. Steve Black, “An Assessment of Social Sciences Coverage by Four Prominent Full-text Online Aggregated Journal Packages,” Library Collections, Acquisitions, & Technical Services 23, no. 4 (1999): 411–19. 12. Michael Colby, “Music Periodical Indexing in General Databases,” Notes 54 (1997): 27–37. 13. Péter Jacsó, “Analyzing the Journal Coverage of Abstracting/Indexing Databases at Variable Aggregate and Analytic Levels,” Library & Information Science Research 20, no.2 (1998): 133–51. 14. Karen Chapman, “Full-Text Database Support for Scholarly Research in Finance,” Journal of Business & Finance Librarianship 7, no. 4 (2002): 35–44. 15. William H. Walters and Esther I. Wilder, “Bibliographic Index Coverage of a Multidisci- plinary Field,” Journal of the American Society for Information Science and Technology 54, no. 14 (2003): 1305–12. 16. Jing Hu, Lai C. Liu, Kai S. Koong, and Lillian Fok, “Information Systems Journals: A Study of University Library Holdings,” Journal of Business & Finance Librarianship 8, no. 1 (2002): 27–43. 17. Chapman and Brothers, “A Reference Study of Leading MIS Journals.” 18. Carol Saunders, ed., “Journal Rankings,” Association for Information Systems. [Accessed 8 November 2004]. Available online at www.isworld.org/csaunders/rankings.htm. 19. For an example of a study using a survey, see Nikolaos A. Mylonopoulos and Vasilis Theoharakis, “Global Perceptions of IS Journals: Where Is the Best IS Research Published?” Com- munications of the ACM 44, no. 9 (2001): 29–33; for an example of a study using citation analysis, see Randolph B. Cooper, David Blair, and Miranda Pao, “Communicating MIS Research: A Citation Study of Journal Influence,” Information Processing & Management 29, no. 1 (1993): 113–27. 20. Saunders, “Journal Rankings.” 21. Kent A. Walstrom and Bill C. Hardgrave, “Forums for Information Systems Scholars: III,” Information & Management 39 (2001): 117–24. 22. Elaine Wagner, “Multidisciplinary Searching for Comprehensive Retrieval,” Georgia Library Quarterly 35, no. 1 (1998): 4. 23. Jacsó, “Analyzing the Journal Coverage of Abstracting/Indexing Databases at Variable Aggregate and Analytic Levels.” 24. Walstrom and Hardgrave, “Forums for Information Systems Scholars,” 120. 62 College & Research Libraries January 2006