Information-Seeking Behavior in the Digital Age: A Multidisciplinary Study of Academic Researchers

Xuemei Ge

This article focuses on how electronic information resources influence the information-seeking process in the social sciences and humanities. It examines the information-seeking behavior of scholars in these fields, and extends the David Ellis model of information-seeking behavior for social scientists, which includes six characteristics: starting, chaining, brows- ing, differentiating, monitoring, and extracting. The study was conducted at Tennessee State University (TSU). Thirty active social sciences and humanities faculty, as well as doctoral students, were interviewed about their use of electronic information resources for research purposes, their perception of electronic and print materials, their opinions concerning the Ellis model, and ways the model might apply to them. Based on the interview results, the researcher provides suggestions on how current information services and products can be improved to better serve social sciences and humanities researchers. The author makes recommenda- tions for improving library services and technologies to better meet the needs of social sciences and humanities scholars.

odern modes of technology 4. FTP (file transfer protocol) have changed the information 5. Online catalogs environment in which social 6. Electronic journals sciences and humanities re- 7. Databases searchers work. The pursuit of knowledge 8. Web portals has been revolutionized, mainly through David Ellis proposed a behavior model the vast expansion of data accessible via of information-seeking behavior based the Internet. Increased knowledge of the on observations of social scientists. The information-seeking behaviors of social model includes six fundamental charac- sciences and humanities researchers is teristics of : starting, crucial to meeting their information needs. chaining, browsing, differentiating, monitor- The electronic information resources ing, and extracting.1 examined in this study include: This study seeks to understand how 1. Electronic mail electronic information resources affect 2. Listservs the information-seeking processes in the 3. Web sites social sciences and humanities. It also

Xuemei Ge is Metadata Librarian in Brown-Daniel Library at Tennessee State University; e-mail: xge@ tnstate.edu. © Xuemei Ge

435 crl-34r2 436 College & Research Libraries September 2010 endeavors to discover how technology ing.4 Stephen E. Wiberley, Jr. and William contributes to and perhaps alters the infor- G. Jones revealed that temporal factors mation-seeking process and explores the have a significant impact on humanists’ applicability of Ellis’s model in the digital adoption of electronic information tech- information environment. The findings of nology and identified and described four this study provide suggestions on how types of time intervals that influence hu- current information services and systems manists’ behavior. Three are types of time can be improved to better serve social sci- spent: anticipated start-up time, actual ences and humanities researchers as they start-up time, and use time; the fourth is navigate this new data-rich environment. time of life: that is, the stage or trajectory The study addresses the following of a scholar’s project or career.5 Margaret specific research questions: Stieg Dalton and Laurie Charnigo studied 1. What roles do electronic informa- historians’ attitudes toward and use of tion resources play in social sciences and electronic materials and found that the humanities researchers’ information seek- application of electronic resources have ing behavior? increased historians’ use of online cata- 2. How do social sciences and humani- logs and indexes in their efforts to iden- ties researchers use electronic information tify appropriate primary and secondary resources in their information seeking? sources of information.6 Susana Romanos 3. What factors affect these researchers’ de Tiratel investigated the information- use of electronic information resources? seeking behavior of Argentine humanities 4. To what extent is Ellis’s behavior and social sciences scholars in 2000, and model applicable to the digital informa- found no substantial differences between tion environment? them, concluding they share similar infor- mation-seeking behaviors.7 Peiling Wang Literature Review wrote about disciplinary and cultural Research findings concerning the use of differences among information seekers in electronic resources by social scientists the Internet age, concluding that there are and humanities scholars vary, but recent differences across disciplines and cultures studies show an increase in the use of in terms of how they rank the importance electronic resources. A study of Brazilian of these resources and how much they social sciences researchers found that, use them.8 In her 2007 paper about the although print resources are still the most information-seeking behaviors of aca- frequently used, electronic resources are demic researchers in the Internet age, a becoming increasingly popular. Access to user study in the United States, China, networked computers is the main obstacle and Greece, Wang further discussed the to the use of databases and other elec- information needs, information-seeking tronic resources.2 Hannah Francis focused behaviors, and resource use of selected on a study that described the information- special interest groups.9 In their review seeking behavior of social sciences faculty of scholarly information practices in the at the University of the West Indies (UWI). online environment, Carole L. Palmer, One of this 2005 study’s findings was that Lauren C. Teffeau, and Carrie M. Pirmann social scientists prefer journal articles in began to address the problem by report- electronic format over print.3 David Ellis ing on the state of knowledge on scholarly and Hanna Oldman’s study explored the information behavior, focusing on the information-seeking behavior of research- information-seeking activities involved in ers in the field of English literature at the research process and how they differ British universities. The article concluded across disciplines. The report found that with recommendations for further study information practices may be enhanced of the use of electronic resources in rela- or advanced by new information re- tion to information literacy and brows- sources and tools. What has changed in Information-Seeking Behavior in the Digital Age 437 the digital environment is not the value Most of the information-seeking be- of these kinds of sources, but rather how havior categories in Ellis’s model are sup- they are searched, accessed, and used ported by capabilities available in com- in the scholarly process.10 Clara M. Chu mon Web browsers. Thus, an individual studied the information needs of literary could initiate surfing the Web from one of critics while producing literary criticism a few favorite Web sites (starting); follow and developed a six-stage model of pro- hypertext links to related information ducing literary criticism, which includes resources in both backward- and forward- ideas, preparation, elaboration, analysis linking directions (chaining); scan the Web and writing, dissemination, and further pages of the sources selected (browsing); writing and dissemination.11 bookmark useful sources for future refer- The study of electronic information- ence and visits (differentiating); subscribe seeking behavior in the social sciences and to e-mail-based services that alert the humanities dates back to the 1980s, but user of new information or development David Ellis was the first to model the pro- (monitoring); and search a particular cess of information-seeking behavior of source or site for all information on that social scientists: how they search for and site on a particular topic (extracting).14,15 interact with the materials, as opposed In the book, Looking for Information, to the sources they use and the manner Donald O. Case indicates that the Ellis in which the materials are obtained. Ellis model makes no claim of considering described six fundamental characteristics the many factors and variables gener- of information seeking practiced by social ally involved in information seeking. scientists: starting, chaining, browsing, For example, the type of need and what differentiating, monitoring, and extracting. sort of information or “help” might Starting refers to the information-seeking satisfy it, or the availability of sources patterns of researchers beginning work in and their characteristics.16 Lokman I. a new area. Chaining describes the process Meho and Stephanie W. Haas’s study of following chains of citations or other on information-seeking behavior of forms of referential connection between social sciences faculty studying state- materials or sources identified during less nations revealed a frequent use of “starting” activities. Browsing is defined , with 88 percent as “semi-directed or semi-structured of participants responding that they use searching in an area of potential inter- electronic resources.17 Access problems est.” Differentiating involves “identifying were a major issue for selected materials, different sets of sources in terms of the with 83 percent of faculty reporting they differing probability of their containing travel to special collections or archives to useful material.” Monitoring is an activity locate historical documents. In another limited to those people following devel- study, Lokman I. Mehe and Helen R. opments in specialized areas. Finally, Tibbo revised Ellis’s information-seeking extracting refers to “the activity of going behavior model, using a specialized case through a particular source and selec- study of social sciences faculty research- tively identifying relevant materials from ing stateless nations. They developed a that source.”12 It should also be noted model which differs from Ellis’s, grouping that Ellis conducted another study about all information-seeking behavior into four information-seeking patterns of academic interrelated stages: searching, accessing, researchers in 1993. Ellis’s comparison of processing, and ending.18 the different activities reported by social scientists led to the conclusion that these Methodology six categories were sufficient to represent Research Design the different information-seeking patterns This study adopts a qualitative approach of researchers.13 to information-seeking behavior, using 438 College & Research Libraries September 2010 the interview method as the primary tool model was presented to the interviewee, for data collection. The presence of an and comments were solicited. The final interviewer improves response rates and question offered the interviewee another quality of answers, since interviewers can opportunity to add any comments to the clarify questions as well as ask follow-up interview. The questions in the Interview questions. The interviews were recorded Guideline attempted to address the and transcribed. researcher’s questions about what roles electronic information resources play in Interview Guideline social sciences and humanities research- The interview design was semi-structured ers’ information seeking, how social and included both closed and open-ended sciences and humanities researchers use questions. After a review of the existing electronic information resources in their literature, interview questions were de- information seeking, which factors affect veloped that described information use researchers’ use of electronic informa- patterns, methods of locating informa- tion resources, and to what extent Ellis’s tion, and use of information technology, behavior model is applicable to the elec- with a particular emphasis on electronic tronic information environment. resources. An Interview Guideline was designed to systematically collect data Population and Sampling (see Appendix A). The Interview Guide- The study was conducted at Tennessee line consisted of a series of thirteen ques- State University in Nashville, where the tions intended to drive the interview researcher works. Tennessee State Univer- process. Initial interview questions were sity is an urban land-grant university with developed from domains that emerged particularly strong programs and depart- from literature review in concert with the ments in social sciences and humanities: researcher’s experience. The first question Education, Sociology, Business, Commu- asked the interviewee to briefly describe nications, History, Geography, Political a recent research project which required Science, and Women’s Studies, Africana library research or access to scholarly Studies, Literature, Philosophy and In- information. This question grounded the terdisciplinary Studies. Tennessee State interviewer and the interviewee in a spe- University has more than 460 full-time cific research situation when responding and part-time faculty members, many of to subsequent questions about informa- whom publish regularly. During the time tion resources used to support research. of the study, the university library had Questions 2 through 9 were structured fewer than 400,000 monographs, 1,700 questions about the use of eight types journals and possessed more than 100 da- of electronic resources, such as the Web, tabases. Currently, the university library e-mail, and FTP. Each question had three has approximately 350,000 monographs, subsequent sections regarding usage 168,866 ebooks, and 1,098 print and online frequency and number of years of use, as periodical subscriptions, but possesses well as an importance rating. Question 10 rich electronic resources, including more asked the interviewees to compare their than 170 electronic databases. For the use of print resources with their use of past three years, the undergraduate en- electronic resources. Question 11, an rollment has been less than 8,000, while open-ended question, gave the interview- the graduate enrollment, which includes ees the opportunity to freely comment both master’s and doctoral programs, has on the mentioned electronic resources averaged less than 2,000. and to elaborate on how and why each The study included two types of inter- source was selected. Question 12, also viewees: faculty members and doctoral an open-ended question, segued the students. The researcher contacted 46 interview to Ellis’s behavior model. The potential participants via e-mail to inquire Information-Seeking Behavior in the Digital Age 439 about their interest in the project. Thirty Findings and Discussions active and productive social sciences and Interview Results humanities faculty and doctoral students Interviews were conducted between June at Tennessee State University agreed to and December of 2004 and took place in participate. participants’ departmental offices. The length of the interviews ranged from 45 Data Collection and Analysis Procedure to 90 minutes, with the majority lasting Before the interview, each participant about an hour. At the end of the data signed an informed consent form ad- collection process, all interview data, dressing voluntary participation. The relevant portions of the taped interviews, interviewer provided a short introduc- and notes were transformed into MS tion to the research topic. After giving Word files. Transcripts ranged from 1,400 definitions of the electronic resources for to 3,600 words and generated enough the study, the researcher followed the data to provide a detailed and accurate Interview Guideline (see Appendix A). account of researchers’ perceptions of Participants were asked about their work, their information-seeking activities. their research habits and methodologies, Topics discussed included researchers’ and their usage of eight different elec- use of electronic information resources, tronic information resources for research their perceptions of electronic and print purposes. The researcher took notes on a resources, the problems they encountered response sheet listing the questions asked. in the research process, their methods of With the permission of the participants, keeping abreast of new developments in all interviews were recorded on tape. To the field, their help-seeking behaviors, preserve anonymity, both the notes and and their opinions as to whether David El- tapes were given an identifying number lis’s information-seeking behavior model retained only by the researcher. Indi- was applicable to their research. The vidual responses were identified by the overall goal was to assess how technology identification numbers assigned during contributes to the information-seeking data analysis and all recorded data were process for social sciences and humanities transcribed into text using these identifi- researchers and to determine how well cation numbers. The interview data were Ellis’s model applies to new technologies. coded and tabulated to facilitate analysis The research verified the Ellis model and and comparison using both the quantita- revealed potential new features. tive and the qualitative analysis methods. The Participants Limitations of the Study The 30 participants interviewed were Tennessee State University is mainly diverse in terms of gender, rank, dis- teaching oriented and does not have cipline, and research topics. Social sci- many doctoral programs in social sciences ences researchers in the study included and humanities. individuals who conduct research in the

Table 1 Participants by Rank and Gender (n = 30) Doctoral Assistant Associate Professor Row Total Student Professor Professor Male 4 (13.3%) 4 (13.3%) 4 (13.3%) 7 (23.3%) 19 (63.2%) Female 5 (16.7%) 4 (13.3%) 1 (3.3%) 1 (3.3%) 11 (36.6%) Column Total 9 (30.0%) 8 (26.6%) 5 (16.6%) 8 (26.6%) 30 (99.8%)* * Due to rounding 440 College & Research Libraries September 2010

Table 2 Participants by Discipline (n = 30) Broad Category Discipline Number of Participants Social Sciences Ed. Administration 10 (33.3%) Teaching & Learning 5 (16.7%) Hotel Management 3 (10.0%) Sociology 2 (6.7%) Business Administration 1 (3.3%) Geography 1 (3.3%) Political Science 1 (3.3%) Humanities History 2 (6.7%) African Studies 2 (6.7%) Women’s Studies 1 (3.3%) Literature & Philosophy 1 (3.3%) Communication 1 (3.3%) fields of Educational Administration, Use of Electronic Information Resources Teaching & Learning, Hotel Manage- in Research ment, Sociology, Business Administra- The data obtained from the interviewees tion, Geography, and Political Science. provides insight into the role of electronic Individuals conducting research in information resources in information- the fields of History, Africana Studies, seeking behaviors. The following cat- Women’s Studies, and Literature & egories were examined during the study: Philosophy were classified as humani- Use of Electronic Information Resources ties researchers. A participant from the for Research. Among the eight types of Communications department was also electronic information resources, the considered a humanities researcher, Web was used by 29 (96.7%) participants since the participant teaches Theater for research and information-gathering, classes and conducts research in the databases were used by 27 (90.0%) par- humanities (see table 1 and table 2). ticipants, e-journals were used by 26

Table 3 Numbers of Users and Average Years of Use for Various Electronic Resources for Research (n = 30) Type of Internet Number of Average Years Standard Range Resources Users of Usage Deviation (Min–Max) Web 29 (96.7%) 6.8 2.8 2–12.5 Databases 27 (90.0%) 6.1 3.2 1.5–14 E-journals 26 (86.7%) 4.8 3.1 1–12 E-mail 25 (83.3%) 6.7 4.3 1.5–17 Online Catalogs 24 (80.0%) 7.4 5.0 1–17 Listserv 10 (33.3%) 5.4 5.1 1–16 Portals 10 (33.3%) 6.6 3.7 2 –12 FTP 9 (30.0%) 6.7 4.0 3–12 Information-Seeking Behavior in the Digital Age 441

Table 4 Frequency of Participant Use of Electronic Resources for Research Internet Multiple Daily Weekly Monthly Rarely Resources Times a Day Web 4 (13.8%) 10 (34.5%) 13 (44.8%) 0 (0%) 2 (6.9%) E-mail 2 (8.0%) 8 (32.0%) 10 (40.0%) 2 (8.0%) 3 (12.0%) Listservs 0 (0%) 6 (60.0%) 1 (10.0%) 2 (20.0%) 1 (10.0%) Databases 0 (0%) 7 (25.9%) 15 (55.6%) 4 (14.8%) 1 (3.7%) E-journals 0 (0%) 4 (15.4%) 15 (57.7%) 5 (19.2%) 2 (7.7%) Online Catalogs 1 (4.2%) 2 (8.3%) 13 (54.2%) 6 (25.5%) 2 (8.3%) Portals 0 (0%) 2 (20.0%) 5 (50.0%) 2 (20.0%) 1 (10.0%) FTP 0 (0%) 0 (0%) 0 (0%) 3 (33.3%) 6 (77.7%)

(86.7%) participants, e-mail was used by resource used. Problems associated with 25 (83.3%) participants, online catalogs Web use that was reported by participants were used by 24 (80.0%) participants, include information overload, difficulty in listservs and portals were each only used conducting precise searches, the mixture by 10 (33.3%) participants, and FTP was of substantial and irrelevant sites, and only used by 9 (30.0%) participants (see difficulty in evaluating the credibility and table 3). actual sources of data. Frequency of Use and Number of Years of Databases ranked second in impor- Use. More than 48 percent of the Web users tance. Many participants originally used interviewed visited the Web as an infor- databases in their traditional index forms, mation-gathering tool daily or multiple but later migrated to “transformed” data- times a day. Forty percent of e-mail users bases, which they perceived as a powerful interviewed used e-mail as an information- tool. The majority of researchers preferred gathering tool daily or multiple times a subject databases containing an abundance day. Sixty percent of listserv users were of journal titles to find articles, as opposed daily listserv readers. More than 80 percent to searching just one e-journal. Several us- of database users accessed them on a daily ers had only utilized the free databases that or weekly basis. More than 70 percent of are available on the Web (such as the free e-journal users accessed e-journals daily basic version of the Education Resources to weekly. More than 80 percent of cata- Information Center site, or U.S. Census log users used online catalogs weekly or Bureau databases), and failed to benefit monthly. Seventy percent of portal users from the library’s quality-controlled, fee- chose that avenue daily or weekly. More based databases. Participants reported than 75 percent of FTP users rarely used encountering difficulties stemming from it as the default research strategy. Data problems with the functionality of the shows that the number of years of use does library’s ILS or the university’s network not necessarily relate to frequency of usage issues, inadequate search skills, termino- (see table 3 and table 4). logical problems, and deficient knowledge Importance Rankings of Various Electronic of suitable databases for their academic Information Resources. On a five-point scale field. Many were concerned about the (1 being the least important and 5 being availability of older, current, and full-text the most important), the Web received journal articles through databases. the highest ranking, with a score of 4.5 Electronic journals were rated as the on average, thus qualifying it as the most third most important resource. Some important popular electronic research participants only used free e-journals on 442 College & Research Libraries September 2010 the Web because they were unfamiliar Listservs were still fairly new to some with paid subscription e-journals avail- participants. Some non-users reported able at the library. Some participants that they are not familiar with the source were unclear as to the relationships or haven’t been able to find any good between e-journals and databases and listservs in their respective fields. Faculty didn’t see any differences between them. and other academics take advantage of Some preferred e-journals because they listservs to ask or answer questions, could browse journals by tables of con- browse current information in their tents and read full-text articles. There fields, locate information on conferences, were also concerns about the availability discover new publications, and locate rel- for some older, current, and full-text evant calls for papers. Listservs, especially articles, as well as rare or lesser-known moderated ones, tend to be more focused journals. Online catalogs were rated as and relevant to serious research. the fourth most important electronic Web portals were rated as the seventh resource. Many participants browsed most important electronic research re- the library’s online catalogs to locate source. About two-thirds of the partici- the library’s existing print and online pants were not familiar with portals. The resources, and some searched the online participants who did use portals liked catalogs of other libraries, union/con- them because they provided shortcuts sortial catalogs (such as Athena, which that were ideal for people working on includes academic and public libraries specific projects and those who wanted in the Nashville area), and publisher or to keep up with special research interests. vendors’ online catalogs. The most common problems with portals E-mail was rated as the fifth most were the variance in quality and how important electronic resource, whereas quickly they became outdated, leading listservs were the sixth most important. to issues such as dead links. E-mail had become a common com- FTP was rated as the least important munication and networking tool for electronic research resource in this study. participants. It was a method for them Users only occasionally need to transfer to make contacts with experts, conduct or download files using FTP, since today’s interviews or surveys, and network browsers can easily handle most of their with colleagues. Problems mentioned by downloading tasks. FTP was predomi- participants are junk mail and unstable nantly viewed as outdated and obsolete e-mail accounts. (see table 5).

Table 5 Importance Rankings for Various Types of Electronic Information Resources for Research Rank Internet Sources Importance Standard Range Score Deviation (Min–Max) 1 Web 4.5 0.9 1–5 2 Databases 4.4 0.8 2–5 3 E-journals 4.2 0.9 2–5 4 Online Catalogs 4.1 0.9 2–5 5 E-mail 3.6 1.4 1–5 6 Listserv 3.2 1.0 2–5 7 Portal 3.2 1.2 2–5 8 FTP 2.4 1.1 1–4 Information-Seeking Behavior in the Digital Age 443

Table 6 Percent of Participants’ Research Needs Satisfied by Print vs. Electronic Information Resources Type of Source Average Percentage Standard Range (Min–Max) of Research Needs Deviation Satisfied by Type Print Information 42 19 10–90 Electronic Information 58 19 10–90

Use of Electronic Resources vs. Use of It may provide perspective and insight Print Resources to discuss the specifics of the cases cited in It was generally agreed that social scientists table 7. Case 1 is based on a full professor tend to rely heavily on periodicals, while in the Teaching & Learning Department. humanities researchers rely more on books Accustomed to using print resources for and primary sources. While both types of most of his academic career, he was unfa- researchers use a wide range of information miliar with new technologies and found sources, their use of electronic sources is electronic information resources difficult increasing. Overall, the participants used to understand and manage. electronic resources to satisfy 58 percent The participant in Case 2 is a senior re- of their research needs and print sources searcher in the Communications Depart- to satisfy 42 percent (see table 6). Indeed, ment. This professor primarily teaches many researchers displayed a marked pref- Theater classes and avoided electronic in- erence for electronic resources over print. formation resources for the same reasons as the professor in Case 1. In addition, Special Cases the participant had some concerns about Generally speaking, participants’ opin- the availability of electronic information ions on electronic information resources resources for his research projects. were positive. However, even though on Case 3 involved a full professor in the average 58 percent of the researchers’ in- History Department who was mainly con- formation needs are satisfied by electronic cerned about the availability of the older information resources, in certain cases, materials in the discipline. For example, participants’ information needs are mainly the participant commented: satisfied by print. These special cases are useful to bear in mind when discussing the Another thing is that six thousand varying opinions of participants. years of human history is not avail-

Table 7 Percent of Participants’ Research Needs Satisfied by Print vs. Electronic Resources (Special Cases) Case Disciplines Percentage Satisfied Percentage Satisfied by by Print Resources Electronic Resources Case 1 Teaching & Learning 90 10 Case 2 Communication 80 20 Case 3 History 70 30 Case 4 African Study 70 30 Case 5 Literature and Philosophy 60 40 Case 6 Education Administration 60 40 444 College & Research Libraries September 2010

able online. … If you think about lo- Case 6 is a doctoral student in the cal history, court records and deeds, Education Administration Department none of them are available online. who had concerns about the availability … If you think about the priority of of older materials in electronic format. the government of the state of Ten- nessee, who is going to pay millions Use of Print vs. Electronic Resources by of dollars to have people digitize all Rank and Gender of the court cases that go back to the Our data demonstrates diverse usage 1700s? For historians, unless they do patterns for electronic information re- the research for very recent history, sources among users of different academic you have to really get to the print. ranks. Doctoral students and assistant professors are more enthusiastic users A faculty member in the Africana Stud- of electronic information resources, rely- ies Department served as the example for ing on electronic resources more heavily Case 4. The professor was not familiar for their research than associate and full with the library’s electronic information professors. These junior researchers are resources and found it hard to evaluate presumably younger and more comfort- electronic resources on the Web. The na- able with emerging technologies. Indeed, ture of his research relied heavily upon doctoral students satisfied 61.7 percent of field studies and he preferred the print their research needs (12.2 SD) with elec- data to the digitized. tronic information resources, and assistant Case 5 cites a professor in the Language professors satisfied 70.0 percent of their & Philosophy Department, who was a research needs (15.1 SD) with electronic frequent user of electronic information information resources. Conversely, senior resources, but who was concerned about researchers, perhaps less comfortable the availability of electronic resources that with new technology, chose to satisfy the supported his discipline at the university. majority of their research requirements The professor had previously requested with print resources, while associate a database for his field from the library professors satisfied 52.0 percent of their and was planning to request a few more. research needs (22.8 SD) and full profes-

Figure 1 Percent of Research Needs Satisfied by Print vs. Electronic Resources for Different Academic Rankings

80%

70%

60% Print 50% Resources

40% Electronic Resources 30%

20%

10%

0% Doctoral Assistant Associate Professor Student Professor Professor Information-Seeking Behavior in the Digital Age 445 sors satisfied 52.5 percent of their research 2. Accessibility—Most electronic infor- needs (21.9 SD) with print resources (see mation resources are available anywhere, figure 1). anytime to anyone with a computer, and The author also examined how gender participants appreciated this ease and influences participants’ use of electronic convenience. information resources. Our data revealed that male researchers approached elec- I started my research back in the tronic information resources slightly more ’70s. I used to drive to a lot of librar- than female researchers. Due to an uneven ies, and to search their card catalog distribution of participants in terms of to see what they have. It is so conve- gender and rank, comparisons of means nient and effective to use e-sources. and standard deviations could only be So much information is online. I made for two groups: doctoral students can at least find what a library has and assistant professors. Male doctoral through their online catalogs. If I students used electronic information re- really need something, I can get it sources for 63.8 percent of their research through interlibrary loan. (12S) needs (SD 18.0), while female doctoral students used electronic information re- 3. Usability—The majority of scholars sources for 60.0 percent of their research responding to the questionnaire cited the needs (SD 8.2). Meanwhile, male assistant usability features of electronic information professors used electronic information resources. They enjoyed the convenience resources for 80.0 percent of their research of saving and printing, the frequency of needs (SD 8.2), while female assistant updating and the powerful potential of professors used electronic information search functions. They especially appreci- resources for 60.0 percent of their research ated the ease of sharing information and needs (SD 14.1). sending papers to other researchers.

Reasons for Use of Electronic Resources Electronic resources are convenient, According to the responses, electronic easy to use and easy to access. (8E) information resources have a number of advantages over print. It is easy to do the searches. With the 1. Availability in Electronic Format—The computer, I can do Boolean searches amount of information available in electron- and use multiple terms. The com- ic format has vastly increased over recent puter does all the searches for me years. Participants appreciated the options that I used to do the long way. (5Ho) provided by this increased availability. 4. Source Quality—Professors and doc- It takes a lot of space to store my toral students also cited the specificity, journals. I just don’t subscribe to quality, and reliability of certain electronic those very common journals now, resources, such as government sites and such as American Journal of Sociology, e-journals. because usually there is an online version, and I read the online jour- I use databases to search for peer- nals. It saves money and space. (12S) reviewed journal articles. (4E, 8E, 11B,…) I use e-journals to search for Well, the major historical journals peer-reviewed journal papers. (12S,....) are available online. American His- torical Journal, Journal of Modern If you see something like New York History and New York Review of Times, National Journal on the Web, Books are the e-journals that I read you technically assume these have frequently. (14Hi) gone through the same referee pro- 446 College & Research Libraries September 2010

cess, or if you see American Political try to copy and paste things I find Science Review on the Web, you as- online. Hardcopy is nicer. But I am sume what is there has gone through adjusting myself and learning to the same peer-reviewed process. You read more online, because we can see these publications on the Web; get so much through e-resources, you assume they go through the same and we can’t get away from it. (13S) peer-reviewed process, same referee process in that publication. (16P) I love electronic resources. ... Any- thing to avoid stacking papers, I am 5. Disciplinary and Research Topic Speci- delighted. It gives you more oppor- ficity—All of the respondents in this study tunities to find information. (23Ho) used electronic information resources for their research to some extent, but some Obstacles to the Use of Electronic Resources used electronic information resources Data was analyzed to gain insight into more, depending on the nature of available reasons for non-use or low use of elec- information for their disciplines. Many par- tronic information resources for research ticipants would have used more electronic purposes. Several factors were mentioned resources if more had been made available by participants: in their discipline or area of interest. 1. Lack of Availability—Participants felt that some information was not avail- I use their catalogs to find what they able in electronic format. For example, have in sociology and the historical the amount of electronic information re- information. For example, for my sources available to humanities research- project about history of sociology, I ers is low, compared to that available to use the university’s online catalogs social sciences researchers. Information to find who wrote the dissertations collected from this study supports the about sociology at the beginning of notion that the creation of digital archives the program for those universities. for infrequently held materials would be (12S) an enormous benefit to certain faculty.

If I am working on a very current Most of the books don’t have an cutting edge topic like teaching electronic version. Therefore, we e-poetry or digital poetry, it works rely on print. (13S) multi-dimensionally and includes a wide scope of activities. E-poetry 2. Lack of Accessibility—Electronic cannot exist in a traditional journal. resource availability varies by institution. The e-poetry is mixed with words, The library’s services, the individual’s images, job descriptions, and many awareness of the resources, and the per- other interactions. These are things son’s research skills also influence acces- that cannot exist in a traditional sibility of information. The data made journal. (18W) evident that many of the participants encountered significant obstructions to 6. Belief in Efficacy—Many research- , such as unavailability ers clearly possessed positive attitudes of desired sources or unstable or hard-to- toward electronic information resources, use ILS systems. believing these resources to be time- saving, convenient, and effective. There are times that the library systems are down, and there is no I mostly use electronic resources way to use the library’s E-resources now and am learning to use more. when I need to. That is one problem I download things I read online. I with E-resources. (3E) Information-Seeking Behavior in the Digital Age 447

Many times I can’t find full-text accurate information for their research. for many articles through the E- Not all participants are confident that resources, and can’t access most electronic information resources pro- recent and archived issues. (22Ho) vided accurate, reliable, and high-quality information. 3. Usability Issues—Content organiza- tion, interface, and choice of computer There is so much false information system all affect the usage of electronic and garbage on the Web. Informa- information resources. The findings of tion on the Web increases rapidly, the study proved that a well-designed and a lot of it is just garbage in and library homepage, good information garbage out. (24A) literacy skills, and user education are all important. For example, two profes- 5. Disciplinary and Research Topic Con- sors (13S, 14Hi) complained that the straints—Researchers’ disciplines or re- TSU library didn’t have ProQuest. In search topics may influence their usage of fact, the library subscribed to a number electronic resources. As can be seen in table of ProQuest databases but listed them 8, the two broad disciplinary categories in separately on its database Web page. this study show different usage patterns. The author mentioned this to the public Some disciplines and research projects services librarians and the Web master require less extensive information-gath- reformatted the Web page to list the ering from published resources, relying databases under the providers, such as instead on field studies and interviews. ProQuest, Sage, etc. Incomplete catalog- ing of digital material can also be an is- Historians use probably less digital sue, since electronic information sources information than most of the social change so quickly. It is important to pro- scientists, because they are using vide more consistent digital cataloging, primary sources and archives so especially Persistent Uniform Resource much. (14Hi) Locators (PURLs). When you do field studies, they are It is time-consuming for me to kind of connected to the technology search the electronic resources, and in the field. But technology doesn’t sometimes navigating between links do everything, and some of these is also confusing. (21T) have to be done by people. Human intelligence is needed. … Technol- 4. Uneven Source Quality—The results ogy is good in many ways, but it still of the study strongly indicated that many can’t do everything for field studies. participants suffered from information I like primary material and collect overload and were in need of specific and row data for my field studies. (24A)

Table 8 Percent of Research Needs Satisfied by Print vs. Electronic Resources for Social Sciences and Humanities Researchers Broad Disciplinary Category Percent of Needs Satisfied Percent of Needs Satisfied by Print Resources by Electronic Resources Social Sciences 36.7 (Mean) 63.3 (Mean) 18.0 (SD) 18.0 (SD) Humanities 60.0 (Mean) 40.0 (Mean) 14.0 (SD) 14.0 (SD) 448 College & Research Libraries September 2010

6. Perceived Ease of Use—Participants monitoring, browsing, differentiating, liked print materials because they were and extracting stages. Examples follow: convenient, portable, and comfortable to use. Many participants mentioned Starting the discomfort of reading on a computer Sometimes you say that here is screen, preferring to print out materials something I am interested in, here instead. Print materials can also be easier is something I really want to go look to access, browse, and manage. at, so you start your research with that purpose. But I think other times, I can read print materials without you are going to be out there looking a computer and Internet, and take at things, reading things, hearing them with me and read them any- things, and seeing things, and some where comfortably. (21T) thoughts might just pop up at you. Some thoughts might just hit you in 7. Lack of Awareness—The data es- the way that you say, that is interest- tablished that participants may not be ing and I have never thought about fully aware of all the relevant electronic that; and I think I want to go look at resources available to them through the that and explore that. (16Hi) library. Specific recommendations (for instance, for a particular database or Web Chaining site) can influence a researcher’s use of For me, the most important aspect sources. of the model is “chaining.” The reference lists from e-journals are I am not familiar with this resource, excellent starting points. (8E) but it sounds interesting. (4E) I found a bibliography about reli- 8. Personal Constraints—Personal gion and spirituality on the Web constraints are situational. Some users yesterday. I bookmarked the site, believed that they were too busy or “too and also saved it on the disk. I can old” to learn to effectively use electronic try to find some materials from that information resources, or that learning to bibliography later. (10A) use them was too difficult. Browsing TSU library has many databases. I I do a lot of browsing. Now I browse am only using a few of them, and I on the Internet a lot, but before I am not familiar to the rest of them. might have gone to a library to look I don’t have time to go through all at the latest copies of various jour- of them and try to figure out what nals. I look at the tables of contents, they are all about. (11B) abstracts and references. Now I can do that online. (21Ho) Interview Results and the Ellis Model The six characteristics of the Ellis model Monitoring seem to be fairly applicable to partici- I monitor a chat room for the former pants’ use of electronic information re- members of Franciscan priests, and sources. For example, “browsing” Web I get a lot of quotes from that chat pages or e-journals does not seem radi- room. Then I don’t have to interview cally different in nature from browsing ta- those people. I also monitor four bles of contents in journal articles or book print journals regularly. (12S) chapters. Many responses clearly confirm the relevance of Ellis’s information- I think because of the Web and other seeking model, especially the chaining, electronic resources, monitoring Information-Seeking Behavior in the Digital Age 449 almost precedes starting. Sometimes read the abstracts and then scan the one doesn’t have an idea until he/she contents of the materials in order picks it up in the medium. ... I think to decide whether or not to use the we rely more on monitoring and materials. If I need the materials and browsing now. I think monitoring they are full-text linked, I will print has become a constant step. With them out right away. If not, I go to the changing media, it is no longer the library and copy the materials, a linear process. I monitor even be- or request the materials through the fore I start a specific research topic, interlibrary loan. (9T) because that is where I am going to gain a lot for the topic in some ways. All of the interviewees agreed that Ellis’s I think browsing and monitoring model describes the basic process of search- have become things that are no ing for information. They have used part longer part of the process, but that or all of the stages in the model at one time supersede the process. I have them or another and many of the respondents’ going on all the time. (18W) answers fit the model extraordinarily well. Analysis of the data, however, indicates Differentiating that some revisions to the basic model Now differentiating and extracting are needed, because several information- have become a lot more compli- seeking activities or tasks could not be cated. Evaluating, depending on categorized into the six characteristics. Two where you get the materials from, new characteristics emerged inductively can be difficult. One has to figure during the course of data analysis. out where the material comes from, “Preparation and planning” could be a who produced it, and how current new characteristic. Effective searching it is. The whole idea about how requires planning, attention to detail, and current information is has changed successful search strategies. Faced with an radically. (18W) overload of information, it is important to find out which sources are useful, and to I definitely do differentiating. I will discover effective and simple search proce- look for the article by the same au- dures. To undertake an effective search, one thor, particularly in the educational must use a range of tools and technology field. Some authors are well-known that enable information to be identified, in the field. When I was working located, and obtained. For example, to con- on my dissertation, I would read duct an effective search, one needs to figure five, six or seven articles from the out what one is really researching, come same author to see if that author up with keywords and synonyms, and changed his/her perception over use phrases, truncation, Boolean searches, time. (23Ho) and field searching, as appropriate. Data revealed that “preparation and planning” Extracting could be an additional stage in the model: I take differentiating and extracting as two different levels of sorting. Before starting, I explore and diag- You initially sort, and then sort finer nose the research problems, create a when you get close to what you visual or mental picture to see how need. (17Hi) I should proceed with my research, and develop valid solutions. (1E) Once I confirm the topic of my re- search, I go to the Web site (mainly I enjoy using electronic resources, TSU online databases) to search for but sometimes they can be over- sources related to my research. I whelming. For example, if I don’t 450 College & Research Libraries September 2010

pick the proper search terms, some- is more reliable for me too. I always times I get thousands of results, and prefer a hard copy. Once I have it, I it is hard to sort out the relevant always have it. (21T) results. Sometimes I don’t get any results, or not enough. This is one If I can get articles electronically, I problem for e-sources. Electronic will save the articles on my hard resources can save time, but some- drive if it is something that I want times they can be time-consuming to keep. If it is something really also. (6T) important, I might print it out. But more often I would save it. (23Ho) As researchers uncover more resources in an increasingly data-rich world, infor- Thus, our data showed that “prepara- mation management becomes more and tion and planning” and “information man- more important and challenging. Many agement” are major information-seeking participants find it difficult to organize the activities that could be potentially be digital materials they collect. Researchers added to David Ellis’s behavioral model. can cope with large quantities of informa- tion through a variety of strategies, but Conclusions effective tools are needed for information This study explored the degree to which management. “” specific electronic resources are significant thus could potentially qualify as a new to the social sciences and humanities re- characteristic, as indicated by these par- searchers. The study found that electronic ticipant responses: information resources played an essential role in these researchers’ information- “Revising” is one stage that I use for seeking pursuits. Among the eight types my information gathering. I go back of Internet information technologies and update or discard information rated, the Web, databases, and e-journals as the knowledge base expands. are ranked first, second, and third in For instance, I gathered some in- importance, followed by online catalogs formation about six months ago and e-mail. Social sciences researchers for an ongoing research project. As use electronic information resources information expands so fast today, more often than humanities researchers. I found more relevant materials for Doctoral students and assistant professors the project. Some of the information (both academically junior) have a higher I found about six months ago is not rate of usage of electronic information re- important to me anymore, so I went sources than their more senior colleagues. back to reorganize the material I All of the participants surveyed utilized have, and discarded some of it. (8E) electronic resources for their research at some point in the research process, and Even for the materials I find through will continue to employ them as a means e-resources, I like to print them out of gathering information. Easy access to and read the hard copies. It is much information anytime and anywhere is easier to flip the pages back and important to these researchers, making forth for the hard copies, and I can them desire even more electronic informa- also make notes or clip parts of tion resource availability. In certain disci- them and paste them to different plines, however, electronic resources are places. It is much easier for me to perceived to be less available and/or less organize hard copy materials. (10A) necessary to the researcher’s field of exper- tise. The study also explored the degree to It is much easier for me to read, store which Ellis’s model remains relevant in the and organize print material, and it age of electronic resources and confirmed Information-Seeking Behavior in the Digital Age 451 that the characteristics proposed by Ellis’s conducted, the library now has added model continue to play viable roles in two smart classrooms offering library ori- research activities. These characteristics entations and bibliographic instruction; take place in both traditional research professors can request these online. In ad- environments which rely on print and dition, the Embedded Librarian Program the electronic information environment. was successfully implemented in 2007. Many participants mentioned that these This program provides collaboration with characteristics do not necessarily occur in faculty in teaching information literacy the sequence outlined in the study’s inter- skills, either online or “on the ground.” view, or can take place concurrently with In 2004, when the study was conduct- the other characteristics. As researchers ed, the library had DRA, which lacked progress from one activity to another, and sufficient functionality, but the library their use of the characteristics will depend migrated to Innovative Millennium on their individual needs and situations. system in 2005, and was able to provide In addition to the six original characteris- better services. The migration to Innova- tics (starting, chaining, browsing, monitor- tive Millennium system allowed more ing, differentiation, and extracting), this functionality and user friendly features study suggests two new characteristics: that include the addition of ebook records, preparation and planning and information serials MARC records, and book reviews management. These new stages of research to make the online catalog more user development reflect social sciences and friendly. The redesign of the Web OPAC humanities researchers’ methods for lo- page, the addition of a link resolver, the cating relevant information. The two new availability of federated searching, and characteristics identified in this study sug- the addition of more databases, ebooks, gest a need for additional research tools and e-journal packages, have combined and for more flexible and user-friendly to increase the library traffic from on- information systems. The findings of this campus and remote users. The library has investigation indicate that obstacles to submitted a proposal for a new Learning the increased use of electronic resources Commons. The library offers orientation include perceptions of availability and us- classes to new faculty at the beginning of ability. With research data becoming more each semester and invites academic de- accessible than ever before, there are many partments to visit the library for informa- new opportunities for libraries to expand tion literacy sessions tailored to their field. and innovate their functions in today’s Major transformations have occurred electronic information environment. The in the digital world since this research areas for innovation and entrepreneurship was conducted in 2004. Web 2.0 tech- in the field include nologies such as Facebook, Flickr, blogs, interfacing, , user YouTube, and Twitter have emerged. instruction, standardization, classification, Social networking software offers a new information management, preservation, approach for academic libraries, since it and organization of networked informa- allows library users to build relationships tion sources that focus more on users. Aca- with library staff and other library users. demic libraries must integrate technology Libraries are now looking to stay con- and traditional services, making libraries nected to users through various forms of the gateways to the vast wealth of print social media. The Brown-Daniel Library and electronic information now available, at Tennessee State University started us- which will encourage library usage and ing Twitter and Facebook in 2009, and the enhance patron satisfaction. library dean has started a blog. Indeed, such changes are taking place The rise of Facebook, Twitter, blogs, at libraries nationwide. At the Tennessee and other social computing tools has State University, where this research was undoubtedly influenced the information- 452 College & Research Libraries September 2010 seeking behaviors of scholars. New tools Given the changes that have taken place such as RSS aggregators, Google Books, in the field of electronic information mobile computing, Zotero, and the open resources since this research was com- access archives and repositories are also pleted, it would be of interest to conduct likely to change the ways that research- additional studies to investigate how ers access and use technology. These researchers’ use of library and Internet new and emerging technologies may be resources continue to adapt and evolve having a significant research impact that as they continue to conduct research in this article does not address, therefore the fluid world of digital information meriting further exploration and study. resources.

Notes 1. Ellis, David. “A Behavioral Approach to Information Retrieval Design.” Journal of Docu- mentation 45, no. 3 (1989): 171–221 2. Sely Costa and Jack Meadows, “The Impact of Computer Usage on Scholarly Communica- tion among Social Scientists,” Journal of Information Science 26, no. 4 (2000): 255–62. 3. Hannah Francis, “The Information-Seeking Behavior of Social Faculty at the University of the West Indies, St. Augustine Campus,” Journal of Academic Librarianship 31, no. 1 (Jan. 2005): 67–72. 4. David Ellis and Hanna Oldman, “The English Literature Researcher in the Age of the Internet,” Journal of Information Science 31, no. 1 (2005): 29–36. 5. Stephen E. Wiberley, Jr. and William G. Jones, “Time and Technology: A Decade-Long Look at Humanists’ Use of Electronic Information Technology,” College & Research Libraries 61, no. 5 (2000): 421–31. 6. Margaret Stieg Dalton and Laurie Charnigo, “Historians and Their Information Sources,” College & Research Libraries 65, no. 5 (2004): 400–25. 7. Susana Romanos de Tiratel, “Accessing Information Use by Humanists and Social Scientists: A Study at the Universidad de Buenos Aires, Argentina,” Journal of Academic Librarianship 26, no. 5 (2000): 346–54. 8. Peiling Wang, “Information Behaviors of Academic Researchers in the Internet Era: An Interdisciplinary & Cross-cultural Study,” 1st International Scientific Conference eRA: the Infor- mation Technology to Science, Economy, Society & Education, September 16–17, 2006. 9. Peiling Wang, D.A. Dervos, Yan Zhang, and Lei Wu, “Information-seeking Behaviors of Academic Researchers in the Internet Age: A User Study in the United States, China and Greece,” in Proceedings of the 69th Annual Meeting of the American Society for Information Science & Technology, Milwaukee, WI, October 19–24, 2007. 10. Carole L. Palmer, Lauren C. Teffeau, and Carrie M. Pirmann, “Scholarly Information Practices in the Online Environment,” Dublin, OH: OCLC, 2009. 11. Clara M. Chu, “Literary Critics at Work and Their Information Needs: A Research-phases Model,” Library & Information Science Research 21, no. 2 (1999): 247–73. 12. David. Ellis, “A Behavioral Approach to Information Retrieval Design,” Journal of Docu- mentation 45, no. 3 (1989): 171–212. 13. David Ellis, “Modeling the Information-Seeking Patterns of Academic Researchers: A Grounded Theory Approach,” The Library Quarterly 63, no. 4 (1993): 469–86. 14. Chun W. Choo, Brian Detlor, and Don Turnbull, “A Behavioral Model of Information Seeking on the Web: Preliminary Results of a Study of How Managers and IT Specialists Use the Web,” in Proceedings of the ASIS Annual Meeting 35 (1998): 290–302. 15. Chun W. Choo, Brian Detlor, and Don Turnbull, “Information Seeking on the Web: An Integrated Model of Browsing and Searching,” 2000. Available online at http://firstmonday.org/ htbin/cgiwrap/bin/ojs/index.php/fm/article/view/729/638. [Accessed 10 May 2010]. 16. Donald O.Case, Looking for Information. San Diego, CA: Academic Press, 2002. 17. Lokman I.Meho and Stephanie W. Haas, “Information-Seeking Behavior and Use of Social Science Faculty Studying Stateless Nations: A Case Study,” Library and Information Science Research 23, no. 1 (2001): 5–25. 18. Lokman I. Meho and Helen R.Tibbo, “Modeling the Information-Seeking Behavior of Social Scientists: Ellis’s Study Revisited,” Journal of the American Society for Information Science and Technology 54, no. 6 (2003): 570–87. Information-Seeking Behavior in the Digital Age 453 Appendix A: Interview Guideline

Task Analysis This is the core part of the interview. The participants will be asked to respond to a series of questions, such as their information use patterns, methods of locating information, problems encountered in the research process, methods of keeping abreast of new developments in the field, help-seeking behavior, and use of information technology, with a particular emphasis on electronic resources.

Initial interview questions.

Rank of the Participant: ______Department: ______No. ______

[Greet, explain the project briefly, consent form, & set recording] 1. Could you briefly describe one of your recently completed research projects, in which you have used various information resources?

2a. Do you use e-mail to gather information for your research? q Yes. q No. Reason for not using the source______[If no, go to question 3]

2b. How often do you use e-mail to gather information for your research?

2c. How long have you been using e-mail to gather information for your research?

2d. How would you rate the importance of e-mail as a tool for gathering information for your research? Let’s use a scale of 1 to 5, 1 for the least important and 5 for the most important. 1 2 3 4 5

3a. Do you use listservs to gather information for your research? q Yes. q No. Reason for not using the source______[If no, go to question 4]

3b. How often do you use listservs to gather information for your research?

3c. How long have you been using listservs to gather information for your research?

3d. How would you rate the importance of listservs as a tool for gathering information for your research? Let’s use a scale of 1 to 5, 1 for the least important and 5 for the most important. 1 2 3 4 5

3e. Would you please give me a few examples for your most frequently used listservs for the research?

4a. Do you use the Web to gather information for your research? q Yes. q No. Reason for not using the source______[If no, go to question 5]

4b. How often do you use the Web to gather information for your research?

4c. How long have you been using the Web to gather information for your research? 454 College & Research Libraries September 2010

4d. How would you rate the importance of the Web as a tool for gathering information for your research? Let’s use a scale of 1 to 5, 1 for the least important and 5 for the most important. 1 2 3 4 5

5a. Do you use FTP to gather information for your research? q Yes. q No. Reason for not using the source______[If no, go to question 6]

5b. How often do you use FTP to gather information for your research?

5c. How long have you been using FTP to gather information for your research?

5d. How would you rate the importance of FTP as a tool for gathering information for your research? Let’s use a scale of 1 to 5, 1 for the least important and 5 for the most important. 1 2 3 4 5

6a. Do you use online catalogs to gather information for your research? q Yes. q No. Reason for not using the source______[If no, go to question 7]

6b. How often do you use online catalogs to gather information for your research?

6c. How long have you been using online catalogs to gather information for your research?

6d. How would you rate the importance of online catalogs as a tool for gathering information for your research? Let’s use a scale of 1 to 5, 1 for the least important and 5 for the most important. 1 2 3 4 5

7a. Do you use e-journals to gather information for your research? q Yes. q No. Reason for not using the source______[If no, go to question 8]

7b. How often do you use e-journals to gather information for your research?

7c. How long have you been using e-journals to gather information for your research?

7d. How would you rate the importance of e-journals as a tool for gathering informa- tion for your research? Let’s use a scale of 1 to 5, 1 for the least important and 5 for the most important. 1 2 3 4 5

7e. Would you please give me a few examples for your most frequently used electronic journals for the research?

8a. Do you use databases to gather information for your research? q Yes. q No. Reason for not using the source______[If no, go to question 9]

8b. How often do you use databases to gather information for your research?

8c. How long have you been using databases to gather information for your research? Information-Seeking Behavior in the Digital Age 455

8d. How would you rate the importance of databases as a tool for gathering informa- tion for your research? Let’s use a scale of 1 to 5, 1 for the least important and 5 for the most important. 1 2 3 4 5

8e. Would you please give me a few examples of your most frequently used databases for the research?

9a. Do you use portals to gather information for your research? q Yes. q No. Reason for not using the source______[If no, go to question 10]

9b. How often do you use portals to gather information for your research?

9c. How long have you been using portals to gather information for your research?

9d. How would you rate the importance of portals as a tool for gathering information for your research? Let’s use a scale of 1 to 5, 1 for the least important and 5 for the most important. 1 2 3 4 5

9 e. Would you please give me a few examples for your most frequently used portals for the research?

[For nonusers of e-resources, go to question 12 and 13.]

10. How are the above information resources used in your research process? {This is the most important question of all. Make sure they elaborate on how the e- sources were used, and why.} e-mail: listserv: Web: FTP: online catalogs: databases: electronic journals: portal:

11. In comparison between print information resources and electronic information re- sources, which percentage of your research information needs is satisfied by each type?

Print information______

Electronic information______

12. There is a model that depicts six types of information seeking. I am interested in your experiences in finding information for your research. Could you look at the model [give the interviewee a copy of the model] and provide your comments?

13. Is there anything you would like to add to what we have discussed?

{Thank you very much!}