Information Processing and Management 46 (2010) 233243

Contents lists available at ScienceDirect

Information Processing and Management

journal homepage: www.elsevier.com/locate/infoproman

Interdisciplinarity and the information-seeking behavior of scientists

Hamid R. Jamali a,*, David Nicholas b
Department of Library and Information Studies, Faculty of Psychology and Education, Tarbiat Moallem University, No 49, Mofateh Ave, P.O. Box: 15614, Tehran, Iran Centre for Information Behaviour and the Evaluation of Research (CIBER), Department of Information Studies, University College London, Henry Morley Building, Gower Street, London WC1E 6BT, UK
b a

a r t i c l e

i n f o

a b s t r a c t
Adopting an intradisciplinary perspective, this article evaluates the information-seeking behavior of academics from different subelds of physics and astronomy. It investigates the effect of interdisciplinarity (reliance on the literature of other subjects) and the scatter of literature on two aspects of the information-seeking behavior: methods used for keeping up-to-date and for identifying articles. To this end a survey of 114 PhD students and staff at the Department of Physics and Astronomy at the University College London was carried out. The ndings showed that the subelds that are more interdisciplinary or have a more scattered literature are more likely to use general search facilities for nding information. The study also showed that cross-disciplinary use of the literature is not necessarily an indicator of scattered literature. The study reveals intradisciplinary differences among physicists and astronomers in terms of their information-seeking behavior and highlights the risk of overlooking the characteristics of information-seeking behavior of specialized subject communities by focusing on very broad subject categories. 2010 Elsevier Ltd. All rights reserved.

Article history: Received 9 October 2008 Received in revised form 25 October 2009 Accepted 26 December 2009 Available online 25 January 2010 Keywords: Information-seeking behavior Interdisciplinarity Scatter of literature Physics Astronomy

1. Introduction One of the most well-established ndings in regard to scientic information-seeking behavior is that there are major disciplinary differences in behavior. For example, physical scientists have different patterns of information-seeking behavior compared to humanities scholars or social scientists (see e.g. Brockman, Neumann, Palmer, & Tidline, 2001; Brown, 1999; Nelson, 2001; Rusch-Feja & Siebeky, 1999; Smith, 2003; Sparks, 2005; Talja & Maula, 2003; Tenopir, 2003; Tenopir & King, 2002; Tomney & Burton, 1998). But what are the reasons for these disciplinary differences? Several factors have been identied. For example as Talja, Vakkari, Fry, and Wouters (2007) showed different scientic elds may have different research cultures and research culture inuences the use of e-resources. Research culture in Talja et als work refers to characteristics such as research-group membership and the degree of establishment of the research area. Different elds may rely on certain types of information resources and apply specic techniques for locating resources. The availability of different types of resources (e.g. e-journals) in a eld could also have impact on the information-seeking behavior of the scholars of that eld. Two other factors have been shown to have impact on the information-seeking behavior of scholars are the interdisciplinary nature and the scatter of literature of a scientic eld (Talja & Maula, 2003; Tenopir, 2008; Vakkari & Talja, 2005). The interdisciplinarity of a eld refers to the reliance of its scientists on the literature of other disciplines. In other words, the extent to which researchers of a eld use the literature of other disciplines is regarded as a measure of interdisciplinarity

* Corresponding author. Tel.: +98 9127248119. E-mail addresses: h.jamali@gmail.com (H.R. Jamali), david.nicholas@ucl.ac.uk (D. Nicholas). 0306-4573/$ - see front matter 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.ipm.2009.12.010

234

H.R. Jamali, D. Nicholas / Information Processing and Management 46 (2010) 233243

of that research eld. The level of scatter of the literature of a eld is the degree to which information on a subject is distributed (scattered) among the resources where such information may be expected to be found. Although the two concepts of scatter of literature and interdisciplinary overlap, they are different. For instance, researchers in condensed matter physics may use the literature of chemistry, but this does not mean that they have a scattered body of literature as the chemistry literature they use might be published in only one or two specic chemistry journals. Therefore, their eld could be interdisciplinary but with a relatively concentrated literature. Scientists who work in subject areas of an interdisciplinary nature are thought to have different information seeking patterns compared to those who practice in well established subject areas with clear boundaries (Bates, 1996). However, the nature of this difference needs further investigation and clarication. This paper proceeds to investigate this issue in the context of physics and astronomy. The article focuses on the impact of interdisciplinarity and scatter of literature on two aspects of information-seeking behavior: (a) Methods used for keeping up-to-date, which is a very important aspect of the information-seeking behavior of scientists. (b) Methods used for identifying journal articles of interest; articles are the main source of research information in many scientic elds including physics and astronomy (Gould & Pearce, 1991 as cited in Lawal, 2002). A case study approach is adopted and the topic is investigated in the light of academic physicists and astronomers working at the University College London (UCL). The study adopts an intradisciplinary perspective (examining the subelds of physics and astronomy) in order to get a closer and clearer picture of the issues being investigated. The reason is that most of the research in the eld of scientic information-seeking behavior deals with rather broad subject categories such as natural sciences, physical sciences and social sciences; and sometimes somewhat narrower (but still broad) categories as physics, chemistry etc (see for example Sparks, 2005). However, nowadays a single discipline like physics or chemistry is so heterogeneous it needs to be broken down to specialized subelds, many of which will have quite different characteristics. 2. Related works About a decade ago scholars such as Palmer (1999, p. 242) highlighted the fact that we had a limited understanding of how interdisciplinarians gather and use information. During the last decade some studies (e.g. Talja & Maula, 2003; Talja et al., 2007; Vakkari, 2006, 2008; Vakkari & Talja, 2005) have dealt with the issue of scatter of literature and interdisciplinarity and their relation with the information-seeking behavior of scholars. Mote (1962) was the rst researcher who made the distinction between subjects with high and subjects with low scatter. Before the popularity of digital information resources and services and in the age of print resources, a study by Packer and Soergel (1979) showed that scholars in high scatter subject areas tended to use current awareness methods and spent more time on searching. Somewhat in contrast to Packer and Soergel, Bates (2002) later hypothesized that scholars in high scatter elds use chaining and browsing as their primary search methods, whereas directed keyword searching is a more effective method for nding relevant materials in low scatter elds. In recent years, Talja and Vakkari and their colleagues have done quite a few studies (mainly on the users of FinELib) in which they have investigated the effect of scatter of literature on the information-seeking behavior of scholars. The key ndings of these studies are:  High scatter leads to a more intensive use of both journal and reference databases. In other words, those who use literature from several elds use more databases of all types compared to those who use literature mainly from their own eld (Talja & Maula, 2003; Vakkari & Talja, 2005).  Increase in scatter of literature increases only the importance of searching in reference databases as a method of information seeking. This could have a few reasons including: (a) there are a larger number of relevant databases in high scatter elds implying greater effort in keeping up-to-date; (b) mixed-journal databases containing journals from several elds could facilitate search across elds especially when a eld has a great degree of vocabulary control; (c) researchers in some high scatter elds probably reduce their search load by rst searching databases for references and then continue onto the full-text journals (Vakkari & Talja, 2005).  Perceived availability of the material provided is a stronger predictor of the use than users discipline and the availability of resources underlies the disciplinary variation in the use of digital libraries (Vakkari, 2006).  High degree of scatter increases the number of used journal databases in less established research areas whereas it has less inuence on the use of journal databases in established research areas (Talja et al., 2007).  Users who consider themselves highly interdisciplinary are more likely to nd e-articles by following citation links, value scientic monographs and conference proceedings more than others and textbooks less, and are more likely to be older than 36 (Tenopir, 2008). With regard to the information-seeking behavior of physicists and astronomers, a survey at the University of Oklahoma by Brown (1999) showed that physicists and astronomers used citations at the end of articles (94%), retrospective searching of indexing/abstracting tools (56%), personal communication (50%) and browsing older volumes (19%) for identifying arti-

H.R. Jamali, D. Nicholas / Information Processing and Management 46 (2010) 233243

235

cles. Nicholas, Huntington, Rowlands, Jamali, and Haynes (2005) surveyed the users of the Institute of Physics (IoP) journals and found out that the most frequent method used to locate journal articles was visiting a journals website. Younger respondents were more likely to rely on the Web of Science, while older respondents were likely to depend on their personal collection. In general, younger users were more likely to depend on and use online methods. Two surveys of astronomers by Tenopir, King, Boyce, Grayson, and Paulson (2005) indicated that they are highly reliant on e-journals. They concluded that the availability of a mature electronic journals system (such as Astrophysics Data System, ADS) from their primary professional society has surely inuenced their early adoption of e-journals. Another study by Nicholas, Huntington, and Jamali (2006) that utilized both log analysis technique and online survey to study the information-seeking behavior of ScienceDirects authors (as users) revealed some information about physicists in comparison to scientists in other elds. Physicists, compared to respondents from other subjects, were more likely to be browsers than searchers they obtained much information by requesting journal homepages and journal issues. Physicists were also more active, making a higher number of requests in a session. In brief, it could be said that physicists and astronomers have some distinct attributes including: they are heavy users of e-print archives (Hurd, 1996; Kling & McKim, 2000; Fry, 2003), they are early adopters of, and as a matter of fact among developers of e-journals and e-resources (Dixon, 1999, p. 3; Singleton, 1997, p. 152, Kurtz et al., 2000, 2005; Tenopir et al., 2005), they have one of the best organised literature in science (Gould & Pearce, 1991, cited in Lawal, 2002), and nally they are known as innovators in methods of scholarly communication (Wertman, 1999). 3. Methodology This article is based on a mixed-method research project (Jamali, 2008) whose aim was to conduct an intradisciplinary study on the information-seeking behavior of physicists and astronomers. What is presented in this article is based on the quantitative part of the study. The broad aim of the quantitative part of the original study was to nd out about certain aspects of information-seeking behavior of scientists such as keeping up-to-date and nding articles as well as to triangulate some of the ndings of the qualitative part of the study. A self-administered web-based questionnaire (see the Appendix A) was designed for conducting the survey. The questionnaires went online on the 3rd of May, 2006 for a month. To conduct the survey, a personalized email was sent to all staff and PhD students in the department with a link to the questionnaire. This was followed by two sets of reminder emails with about ten day intervals. The sample included all PhD students and staff in the Department of Physics and Astronomy at UCL. The Department had the following seven research groups at the time of the study that were considered as the research areas and units of analysis in this study. They are: Atmospheric Physics (AP); High Energy Physics (HEP); Condensed Matter and Materials Physics (CMMP); Astronomy and Astrophysics (AA); Theoretical Molecular Physics (TMP); Atomic, Molecular, Optical and Positron Physics (AMOP); Optical Science Laboratory (OSL). The two main attributes studied in this research, i.e. interdisciplinarity and scatter of literature, were measured by asking two questions (Q7 and Q8 below) of the respondents. Of course this is based on their perception of their eld and may not a 100% match the reality. Possibly, a more reliable measure of interdisciplinarity and scatter of literature would be obtained by conducting scientometric studies, especially citation studies. Conducting that type of study is out of scope of the current research. A search in the literature resulted in no signicant such study that would give us an actual measurement of interdisciplinarity nature and scatter of literature of the different subelds of physics and astronomy, so we could compare it with the respondents perceptions presented in this study. However, the assumption is that the respondents know their eld and know how much they rely on the literature that belongs to other scientic elds. Crosstabulation was used for the statistical analysis and Pearson chi-square and Cramers V tests were used to check the signicance of all results. 4. Findings 4.1. Characteristics of the sample There were 129 staff and 113 PhD students (total 242 people) in the department that were invited to take part in the survey. The survey achieved 47.1% response rate with 114 respondents. This is a good response rate given that academic Webbased surveys participation rates range from 3% to 62% for electronic surveys (Hemminger, Lu, Vaughan, & Adams, 2007). Table 1 shows the size of the research population and the number/percentage of respondents. 57% of respondents were research students and 43% were members of staff. Seven main subelds of physics and astronomy (research group entities inside the department) were used to categorize the respondents (Table 2). CMMP accounted for 31.6% of the respondents. CMMP is the biggest research group in the department and it encompasses a considerable number of smaller research groups that research very specic topics. After CMMP, AA accounted for the second highest number of respondents with 22 (19.3%) respondents. This research group also covers many smaller research groups such as hot stars, star formation and so on. The smallest number of respondents belonged to the OSL with three respondents who all do instrumentation-kind of research. OSL at the time of data collection had 11 members so three respondents means about 0.33% of the population, and it is relatively a representative sample. However, it is still a small number of respondents and the ndings related to this group should be interpreted with extra caution.

236

H.R. Jamali, D. Nicholas / Information Processing and Management 46 (2010) 233243

Table 1 Distribution of the respondents by academic status. Academic status PhD Student Research Fellow Senior Researcher Lecturer Reader Professor Total No 65 20 6 11 2 10 114 % 57 17.5 5.3 9.6 1.8 8.8 100

Table 2 Distribution of the respondents by research group. Research group Atmospheric Physics (AP) High Energy Physics (HEP) Condensed Matter and Materials Physics (CMMP) Astronomy and Astrophysics (AA) Theoretical Molecular Physics (TMP) Atomic, Molecular, Optical and Positron Physics (AMOP) Optical Science Laboratory (OSL) Total No 11 18 36 22 11 13 3 114 % 9.6 15.8 31.6 19.3 9.6 11.4 2.6 100

100 90 80 70 60 50 40 30 20 10

36

6 56

33

27

36

46

67

32 27 14 53 45 36 6 33 18 33 23 15
Often

18 31
Interdisc. use

Sometimes Rarely

8
0

Never

AP HEP

CMMP AA

TMP AMOP

OSL

Fig. 1. Percentage breakdown of use of interdisciplinary literature by research group.

4.2. Interdisciplinarity Respondents were asked to specify how often they needed to use the literature of other disciplines, implying how interdisciplinary their research was. Fig. 1 shows how often respondents in each group used literature of other elds. The data show that groups such as CMMP and TMP are more reliant on the literature of other elds and therefore have a more interdisciplinary nature. Fifty-three percent of respondents from CMMP and 45% from TMP stated that they often use the literature of other elds. Other groups such as HEP rely mainly on their own literature and cannot be counted as interdisciplinary. Only 6% of respondents from HEP stated that they used the literature of other elds often while 33% of them never used the

H.R. Jamali, D. Nicholas / Information Processing and Management 46 (2010) 233243

237

100

9
90 80 70 60 50 40

17

36

9 41

27

8 38

100

9 9 73 53 17 28

36

Scatter of Lit.

54 9
Scattered

39
30 20

41 18
Reasonable Don't know

18
10

8
0

Concentrated TMP AA AMOP OSL

AP HEP

CMMP

Fig. 2. Percentage breakdown of the research group and scatter of the subelds literature.

literature of other elds. A Pearson chi-square test showed that this result was signicant (v2 = 46, Sig = 0.01). However, a Cramers V test (=0.37) showed that the relation between the research eld and interdisciplinarity is not very strong. 4.3. Scatter of literature Respondents were asked how scattered they perceived the literature of their elds to be. The characteristics of literature in different subelds of physics and astronomy are not the same as we can see in Fig. 2. The literature of some subelds of physics, such as CMMP and TMP, is more scattered and some other subelds such as HEP and OSL have clearer boundaries and a more concentrated literature in that they have a clear set of specialized journals as well as databases. All of the respondents from the OSL group maintained that the literature of their eld is concentrated. A Pearson chi-square test approved the signicance of these results (v2 = 39, Sig = 0.01). A Cramers V test (=0.34) however conrmed that the strength of the relations between scatter of literature and research eld is not considerably strong. 4.4. Keeping up-to-date 4.4.1. By interdisciplinarity of the eld Scientists rely on different methods for keeping abreast with the developments of their elds. Some may rely on the word of mouth and what they hear from their friends and colleagues. Some rely on meetings held within their departments or research groups and some go to conferences for that purpose. Besides interpersonal communication methods, there are also print and electronic resources to use. Using email alerts (whether of journals ToC, or of e-print or a database), conducting regular or semi-regular searches in databases such as arxiv.org, and browsing e-resource collections are other methods that could be used for keeping up-to-date. Those with the most interdisciplinary research used database searching more than the others. Nineteen percent of those who said they often need to use other disciplines literature and 11% of those who said they did it sometimes searched databases for keeping up-to-date. The gure was zero for those who never or rarely needed other disciplines literature. A Pearson chi-square test showed that this result was statistically signicant (v2 = 35, Sig = 0.03, Cramers V = 0.39) (see Fig. 3 relates). 4.4.2. By scatter of literature Those who believed their literature was scattered were more likely to search databases and browse e-journals for keeping up-to-date. Those with concentrated literature had the highest rate of e-print archive usage (36%). Those who were not sure about the scatter of literature of their subelds tended to rely on word of mouth more than the other groups did; a third did

238

H.R. Jamali, D. Nicholas / Information Processing and Management 46 (2010) 233243

100 90

42

19

20

14

Most important method

Word of Mouth

80 70 60

14 6 19

9
Meetings

11 33 26

ToC Email Alerts Search Email Alerts E-print Arch Email

8
50 40 30 20 10 0

8 8 8 25 8

Searching Database

8 25

30

31
Conferences Browsing E-print Arc Browsing E-Journals

Never Often

Rarely Sometimes

Fig. 3. Percentage breakdown of top ranked methods for keeping up-to-date by interdisciplinarity of subelds.

100

18
90 80 70 60 50 40 30 20 10 0

33

23

13 9 9

Most important method Word of Mouth Meetings ToC Email Alerts Search Email Alerts

5 5 5 8 37

9 11 11 16 22 39 7 23 7 26

E-print Email alert Searching Database Conferences Browsing E-print Arc Browsing E-Journals

21

22

Concentrated Don't know

Reasonable Scattered

scatter of literature
Fig. 4. Percentage breakdown of the top ranked methods for keeping up-to-date by scatter of subelds literature.

so (Fig. 4). A Pearson chi-square test showed that the result was signicant (v2 = 32, Sig = 0.03, Cramers V = 0.38). Again these ndings are in line with those of Vakkari and Talja (2005). 4.5. Finding research articles Articles are an important source of information for scientists. The average number of readings per year per science faculty member continues to increase, while the average time spent per reading is decreasing (Tenopir, King, Edwards, & Lei, 2009).

H.R. Jamali, D. Nicholas / Information Processing and Management 46 (2010) 233243

239

100
25 25 60 39 33 14 36 50 63 8 25 58 32
Sometimes

21

100

90 80 70 60 50 40
15 43

Interdisc. use

30 20 10 0
13 13 13 10 15

29

29

Rarely

Often

18 7 7
Never

Fig. 5. Percentage breakdown of the most used methods for identifying articles by interdisciplinarity of the eld.

Scientists identify the articles they read through different methods. Some articles might be recommended to one by friends or colleagues. A researcher might identify an article through searching a general database, such as Web of Science or a subject-specic database like ADS. An article might also be identied through email alerts a user receives or through browsing an e-journals website. Following the references in other resources might also result in nding more articles. A cross-tabulation of interdisciplinarity with the most used methods for identifying articles (Fig. 5) show that those people with the most interdisciplinary research, i.e. those who used the literature of other elds often, relied on Google scholar (100%), ToC email alerts (63%) and searching general databases such as Web of Science (58%). On the other hand those who used the literature of other elds never or rarely relied more on recommendations (63%), and searching subject databases (54%). A Pearson chi-square test showed that the result was signicant (v2 = 34, Sig = 0.03, Cramers V = 0.41). 4.5.1. By scatter of the elds literature Fig. 6 gives the percentage breakdown of the most used method for identifying articles by the scatter of the subelds literature. The more scattered the literature, the higher the likelihood of using Google, Google Scholar, general databases and subject databases. On the other hand the less scattered the literature of a subeld the higher the chance that respondents used ToC email alerts. This makes sense because databases cover several subject areas and they tend to be a better means for nding articles in several elds simultaneously. The cross-tabulation of the most used method for identifying articles by the scatter of the elds literature and by the interdisciplinarity of the eld generally conrms the ndings by Bates (2002) and Vakkari and Talja (2005). A Pearson chi-square test showed that the result for this crosstabulation was not statistically signicant (v2 = 46, Sig = 0.61). 5. Discussion and conclusions Table 3 proles the subelds of physics and astronomy in terms of their interdisciplinarity and the level of scatter of their literature and it also shows the main methods they relied on for keeping up-to-date and for nding articles. The ranking for the interdisciplinarity and scatter of the table are based on the percentage of the answers of the survey respondents to the questions on their use of the literature of other disciplines and the scatter of the literature of their eld. The larger the number of stars the more interdisciplinary and the more scattered the literature of the eld. As we can see from the table, the subelds of physics and astronomy are different in terms of their reliance on different methods used for keeping up-to-date as well as methods used for nding articles. The ndings of the present study raise a question about the past hypotheses proposed by Bates (2002) and Vakkari and Talja (2005) related to the effect that scatter of the literature of a eld and its interdisciplinarity have on the informationseeking behavior of scientists. Bates (2002) suggested that the use of chaining and browsing are the main search methods

h Sc e gl oo G g in e gl ch ar oo a G Se g td in ec ch bj ar su a g Se ld in ra ch ne ar ge Se g es in nc ch re ar fe in re Se ch ng ar ki se ac or Tr ng si ow Br ts ns er tio al da C en To m om ec R

240

H.R. Jamali, D. Nicholas / Information Processing and Management 46 (2010) 233243

100

10
90

22

12 19

13 17

Most used method

Searching Google Sch Searching Google

18
80

33
70 60 50 40

5 28 14 19 33 23 14 26 17

Searching subject da Searching general da Tracking references Browsing or searchin

30 20 10 0

9 10 11 5
Concentrated Don't know Reasonable

17
ToC alerts

12
Recommendations

Scattered

Scatter of literature
Fig. 6. Percentage breakdown of the most used methods for identifying articles by scatter of subelds literature.

Table 3 Information-seeking characteristics of research groups. Main method used for keeping up-to-date AA AMOP AP CMMP HEP OSL TMP Browsing e-print archive Browsing e-journals Browsing e-print, e-journal, Toc alert, Word of mouth Browsing e-journals Browsing e-print archive Conferences Browsing e-journals Main method used for nding articles Search in subject database Browsing and searching e-journal sites Tracking references Search in general databases Search in subject databases Tracking references Tracking references Interdisciplinarity of the eld ** *** ***** ******* * **** ****** Scatter of literature **** ** *** ******* ***** * ******

used by scholars in high scatter elds and directed keyword searching is the more effective method for nding relevant materials in low scatter elds. On the other hand, Vakkari and Talja (2005) argued otherwise and maintained that the increase in scatter increases only the importance of searching in reference databases. A few points should be discussed here: (a) The subject categories used in Vakkari and Taljas research were very broad and included categories such as Natural Sciences that encompassed physics, chemistry, food industry and some other disciplines. One must appreciate that these categories (e.g. natural sciences) are broad academic or scholarly areas and each encompasses several more specic and diverse scholarly domains with their own research cultures and trends. Several authors (Bawden, 2006, p. 676; Case, 2002; Fry & Talja, 2004, p. 21; Kling & McKim, 2000; Wilson, 1981) have pointed out the need for narrowing the research focus for in-depth studies of well-dened groups to determine the underlying factors of behavior, something in which this study tried to contribute. (b) With regard to the effect of interdisciplinarity and scatter of literature, we can see that although there is support for the ndings of Vakkari and Taljas (2005) study, one must be cautious in generalising this nding as there are many contributing factors that could affect the information-seeking behavior of scholars of a specic eld. While in an area such as CMMP, which has a highly scattered literature, searching in general databases is the main searching method, in another area with scattered literature such as OSL tracking references is the main mean for identifying articles. Although people in OSL use the literature of other elds they believe their literature body is reasonably concentrated.

H.R. Jamali, D. Nicholas / Information Processing and Management 46 (2010) 233243

241

(c) Arguing against Bates (2002), the cross-disciplinary use of literature does not necessarily imply the scatter of literature. For example researchers in a subeld of physics might use the chemistry literature, but the chemical papers they use may be mainly published in one or two specic chemistry journals allocated to a specialized subeld of chemistry. This means that the literature they rely on is not scattered, but concentrated in a few journals. Therefore, the interdisciplinary use and the scatter of literature should be considered two separate issues. With regard to the information-seeking behavior of physicists and astronomers, in general, the study lends support to the ndings of Brown (1999) that showed high usage of citation tracking. It also conrms some of the ndings by Nicholas et al. (2006) that showed physicists compared to scientists in some other subjects were more likely to be browsers. However, as we have found specic subelds are different and talking of physicists here might be over-generalisation of the data. It should be borne in mind that this study has some limitations and one must be cautious in generalising the ndings. A limitation of this study is that we have focused on scientists opinions about the interdisciplinarity and scatter of literature of their elds. It is difcult to say to what extent the subjective opinion reects the objective nature of their elds. Another limitation was the size of the population. The number of respondents of some of the groups, especially OSL, was very small and this has denitely had impact on the results, which could be see in relatively low results of Cramers V tests. Some of the ndings were not statistically signicant and a study with a larger population should be conducted for further clarication. However, this study was meant to explore the issues of interdisciplinarity and scatter of literature with an intradisciplinary perspective, an aim that has been mostly achieved. Obviously, further international research is required to reach a rm conclusion on the issues discussed in this article. Especially as the past studies have shown that besides the factors investigated in this article, quite a number of factors including research orientation, i.e. basic or applied (Rice & Tarin, 1993), collaborative culture (Kling & McKim, 2000) and communication patterns (Trm & Vakkari, 2004) inuence the use of e-resources.

Acknowledgement The senior author would like to thank the Iranian Ministry of Science, Research and Technology for funding the study. The authors also would like to thank the anonymous reviewers for their constructive comments.

Appendix A (1) In your subeld, how important is rapid awareness of new papers? (Not at all important/A little important/Somewhat important/Quite important/Absolutely critical/I do not know). (2) How dependent are you on each of these methods for keeping up-to-date with developments in your subeld?(Scales: Very dependent/Quite dependent/Not very dependent/Not at all dependent). Browsing electronic journals. Browsing print journals. Browsing preprint archive. Receiving journals table of contents email alerts. Receiving email alerts from preprint archives. Receiving search email alerts (like the service of Web of Knowledge). Newsletters. Departmental or groups seminars and meetings. Conferences. Word of mouth and colleagues. Regular or semi-regular searching on a database or Internet. Other, please specify. (3) Please rank the top three methods you depend on for keeping up-to-date. (4) How often do you use each of these methods for identifying research articles?(Frequency: Daily/23 times a week/ About once or twice a month/Less than once a month/Never). Recommendation from friends. Table of contents email alerts. Browsing or searching journals websites. Following up references at the end of papers. Searching in a general database such as Web of Knowledge. Searching in a subject-specic database such as ADS, Spires, Inspec. Searching Google for works or authors (this does not include when you search Google to nd a journals website). Searching Google Scholar. Other. (5) By which of the above-mentioned methods do you identify the highest number of articles you read?

242

H.R. Jamali, D. Nicholas / Information Processing and Management 46 (2010) 233243

(6) Please, think of the scholarly article you read most recently. How did you identify the last article you read? Through a colleague. Through email alert. Through browsing a journals website. Through references of another paper. Through a search on Google. Through a search on Google Scholar. Through a search in an abstract database (e.g., Web of Science, Spires). I had read it before and was rereading it, so I already knew about it. Other, please specify.In this Section 1 would like to know about your perception of your research area. (7) For your research, how often do you need to look for and use the results of research by people in other disciplines (e.g. chemistry etc.)? (Frequency: Never/Rarely/Sometimes/Often). (8) What do you think of the journal literature of your subeld? It is very scattered in many journals and searchable though several databases. It is reasonable, not very scattered and not very concentrated. It is quite concentrated in a few journals and searchable through a few databases. I do not know. References
Bates, M. J. (1996). Learning about the information seeking of interdisciplinary scholars and students. Library Trends, 45(2), 155164. Bates, M. J. (2002). Speculations on browsing, directed searching, and linking in relation to the Bradford distribution. In H. Bruce, R. Fidel, P. Ingwersen, & P. Vakkari (Eds.), Emerging frameworks and methods: Proceedings of the fourth international conference on conceptions of library and information science (CoLIS 4), July 21-25 (pp. 137149). Seattle, WA: Libraries Unlimited, Greenwood Village. Bawden, D. (2006). Users, user studies and human information behaviour: A three-decade perspective on Tom Wilsons On user studies and information needs. Journal of Documentation, 62(6), 671679. Brockman, W. S., Neumann, L., Palmer, C. L., & Tidline, T. J. (2001). Scholarly work in the humanities and the evolving information environment. Washington, DC: Digital Library Federation, Council on Library and Information Resources. Brown, C. M. (1999). Information literacy of physical science graduate students in the information age. College and Research Libraries, 60(5), 426438. Case, D. O. (2002). Looking for information: A survey of research on information seeking, needs, and behaviour. Amsterdam: Academic Press. Dixon, A. (1999). The twelve ages of electronic journals. Vine(111), 39. Fry, J. (2003). The cultural shaping of scholarly communication within academic specialisms. Unpublished PhD Thesis, University of Brighton, Brighton. Fry, J., & Talja, S. (2004). The cultural shaping of scholarly communication: Explaining e-journal use within and across academic elds. Paper presented at the American society for information science and technology annual meeting on managing and enhancing information: Cultures and conicts, 1318 November, Providence, Rhode Island. Gould, C. C., & Pearce, K. (1991). Information needs in the sciences: An assessment. Mountain View, CA: Research Library Group. Hemminger, B. M., Lu, D., Vaughan, K. T. L., & Adams, S. J. (2007). Information seeking behavior of academic scientists. Journal of the American Society for Information Science and Technology, 54(14), 22052225. Hurd, J. M. (1996). High energy physics. In S. Crawford, J. M. Hurd, & A. C. Weller (Eds.), From print to electronic: The transformation of scientic communication (pp. 6576). Medford, NJ: Information Today. Jamali, H. R. (2008). Information-seeking behaviour of physicists and astronomers: An intradisciplinary study, University College London, Unpublished PhD thesis. Kling, R., & McKim, G. (2000). Not just a matter of time: Field differences in the shaping of electronic media in supporting scientic communication. Journal of the American Society for Information Science and Technology, 51(14), 13061320. Kurtz, M., Eichhorn, G., Accomazzi, A., Grant, C., Demleitner, M., & Murray, S. (2005). Worldwide use and impact of the NASA Astrophysics Data System digital library. Journal of the American Society for Information Science and Technology, 56(1), 3645. Kurtz, M., Eichhorn, G., Accomazzi, A., Grant, C., Murray, S., & Watson, J. (2000). The NASA Astrophysics Data System: Overview. Astronomy and Astrophysics Supplement Series, 143(1), 4160. Lawal, I. (2002). Scholarly communication: The use and non-use of e-print archives for the dissemination of scientic information. Issues in Science and Technology Librarianship (36). <http://www.istl.org/02-fall/index.html> Visited 25.11.07. Mote, L. J. B. (1962). Reasons for the variation of information needs of scientists. Journal of Documentation, 18, 169175. Nelson, D. (2001). The uptake of electronic journals by academics in the UK, their attitudes towards them and their potential impact on scholarly communication. Information Services & Use, 21(34), 205214. Nicholas, D., Huntington, P., & Jamali, H. R. (2006). Authors as users: A deep log analysis linking demographic and attitudinal data obtained from scholarly authors with their usage of ScienceDirect. London: CIBER, University College London. Nicholas, D., Huntington, P., Rowlands, I., Jamali, H. R., & Haynes, J. (2005). An investigation of physicists information-seeking behaviour, publishing practices, and views on the scholarly publishing system: With special reference to the Institute of Physics and its journals, report of an international survey. London: CIBER, University College London. Packer, K. H., & Soergel, D. (1979). The importance of SDI for current awareness in elds with SevereScatter of information. Journal of the American Society for Information Science, 30, 125135. Palmer, C. L. (1999). Structures and strategies of interdisciplinary science. Journal of the American Society for Information Science, 50(3), 242253. Rice, R. E., & Tarin, P. (1993). Staying informed: Scientic communication and use of information sources within disciplines. In S. Bonzi (Ed.), Proceedings of the 56th ASIS annual meeting (Vol. 30, pp. 160164). Medford, NJ: Learned Information. Rusch-Feja, D., & Siebeky, U. (1999). Evaluation of usage and acceptance of electronic journals. D-Lib Magazine, 5(10). <http://www.dlib.org/dlib/october99/ rusch-feja/10rusch-feja-full-report.html> Visited 25.11.07. Singleton, A. (1997). Journal and the electronic programme of the Institute of Physics. The Serials Librarian, 30(3-4), 149161. Smith, E. T. (2003). Changes in faculty reading behaviours: The impact of electronic journals on the University of Georgia. Journal of Academic Librarianship, 29(3), 162168. Sparks, S. (2005). JISC disciplinary differences report. London: Rightscom Ltd. Talja, S., & Maula, H. (2003). Reasons for the use and non use of electronic journals and databases: A domain analytical study in four scholarly disciplines. Journal of Documentation, 59(6), 673691. Talja, S., Vakkari, P., Fry, J., & Wouters, P. (2007). Impact of research cultures on the use of digital library resources. Journal of the American Society for Information Science and Technology, 58(11), 16741685.

H.R. Jamali, D. Nicholas / Information Processing and Management 46 (2010) 233243

243

Tenopir, C. (2003). Use and users of electronic library resources: A overview and analysis of recent research studies. Report for the council on library and information resources. <www.clir.org/pubs/reports/pub120/pub120.pdf> Visited 25.11.07. Tenopir, C. (2008). Digging deeper into e-journal use: Outcomes, value, and interdisciplinary reading, the 2008 annual CONCERT meeting, November 1415, 2008. Tenopir, C., & King, D. W. (2002). Reading behaviour and electronic journals. Learned Publishing, 15(4), 259265. Tenopir, C., King, D. W., Boyce, P., Grayson, M., & Paulson, K. (2005). Relying on electronic journals: Reading patterns of astronomers. Journal of the American Society for Information Science and Technology, 56(8), 786802. Tenopir, C., King, D. W., Edwards, S., & Lei, W. (2009). Electronic journals and changes in scholarly article seeking and reading patterns. Aslib Proceedings, 61(1), 532. Tomney, H., & Burton, P. F. (1998). Electronic journals: A study of usage and attitudes among academics. Journal of Information Science, 24(6), 419429. Trm, S., & Vakkari, P. (2004). Discipline, availability of electronic resources and the use of Finnish National Electronic LibraryFinELib. Information Research, 10(1), Paper No. 204. <http://informationr.net/ir/10-1/paper204.html> Retrieved 06.06.06. Vakkari, P. (2006). Trends in the use of digital libraries by scientists in 20002005: A case study of FinELib. In: Proceedings of the ASIS&T annual meeting. Austin, Texas, November 39, 2006. Vakkari, P. (2008). Perceived inuence of the use of electronic information resources on scholarly work and publication productivity. Journal of the American Society for Information Science and Technology, 59(4), 602612. Vakkari, P., & Talja, S. (2005). The inuence of the scatter of literature on the use of electronic resources across disciplines: A case study of FinELib. In A. Rauber, S. Christodoulakis, & A. M. Tjoa (Eds.). Research and advanced technology for digital libraries: Proceedings of the 9th European conference, lecture notes in computer science (Vol. 3652, pp. 205217). Berlin: Springer. Wertman, E.R. (1999). Electronic preprint distribution: a case study of physicists and chemists at the University of Maryland. Unpublished MSc, Virginia Polytechnic Institute and State University, Blacksburg, VI. Wilson, T. D. (1981). On user studies and information needs. Journal of Documentation, 37(1), 315.