Cross-cultural Analysis of E-mail Reference
by Pnina Shachaf, Lokman I. Meho, and Noriko Hara
Available online 14 December 2006
INTRODUCTION
Studies that examined virtual reference and its
potential for collaboration have by and large Today, many libraries in North America, Europe, Australia,
and Asia participate in collaborative virtual reference, exchang-
represented experiences in western
ing questions among members of a consortium.1 Some of these
English-speaking countries. This article reports consortia involve international collaboration among libraries
the results of a three-nation (Israel, Japan, and from different countries to provide direct virtual reference
Lebanon) comparative analysis to bridge this services to end users worldwide. When joining a consortium,
individual libraries reduce their expenses by sharing refer-
research gap. Similarities and differences
ence responsibilities, the cost of software, and marketing
between these countries highlight elements that activities.2
international collaborative initiatives should There are a number of advantages to the consortia model.3
Member libraries staff the desk for only a few hours a week
consider when implementing global virtual
while the rest of the schedule is covered by partner libraries.
reference services. Collaboration with an institution in a different time zone
enables a member library to provide 24/7 real time chat
reference.4 Furthermore, when an individual library addresses
questions for the entire consortium, librarians use their time
more efficiently as they handle questions from multiple
libraries and have a higher number of questions coming in,
rather than waiting for a question to arrive. International
collaborative virtual reference initiatives may further enhance
reference service, not only by expanding hours of coverage but
also by providing access to more subject specialists. In
addition, collaboration enhances the ability to answer ques-
tions regardless of languages of materials requested and
languages of transactions (for example, a question that is
submitted in French at a German library can be answered by a
French library). Collaborations enables consortia to leverage
unique knowledge and skills of librarians despite geographical
limitations. An additional benefit of joining a consortium is the
ease of creation of a shared knowledge base of all the
questions and answers previously asked for re-use in some
Pnina Shachaf is Assistant Professor,
future point in time.
School of Library and Information Science,
The problem with most if not all existing virtual collabo-
Indiana University, Bloomington, USA
rative reference services and studies is that little attention has
********@*******.****;
been given by librarians and scholars to the impact of cultural
Lokman I. Meho is Assistant Professor,
and linguistic differences among countries on these services.
School of Library and Information Science,
As we know, communication patterns are culturally embedded
Indiana University, 1320 East 10th St.,
and are significantly different from one country to another.5
LI 011 Bloomington, IN 47405-3907, USA
The role of the library in society, and user expectations from
*****@*******.****;
library services, vary as well. These are factors that may
Noriko Hara is Assistant Professor,
considerably influence the nature and quality of virtual
School of Library and Information Science,
reference service provided in global collaborations. As such,
Indiana University 1320 E. 10th Street,
it may be unwise to generalize from research on virtual
LI Room 025 Bloomington,
reference services that have been conducted largely in English-
IN 47405-3907, USA
speaking countries.
******@*******.****.
The Journal of Academic Librarianship, Volume 33, Number 2, pages 243 253 March 2007 243
Sweden. In Asia, collaborative examples include: Reference
...little attention has been given by librarians Point in Singapore, the Virtual Reference Desk of the
Shanghai Central Library, and the China Academic Distributed
and scholars to the impact of cultural and
Collaborative Virtual Reference System (CVRS), which is
linguistic differences among countries on these planned to be launched by the end of 2005.12
services. Examples of international collaborative virtual services
include a joint effort between libraries of the Australian
consortium (GAELIC) and a Florida academic library;13
Somerset libraries in the UK with Brisbane City Council
Very few studies have focused on the provision of
Library Service in Australia and Richmond County Public
collaborative online reference services in non-English speaking
Library in South Carolina;14 Chasing the Sun a virtual
countries. This is partially due to the newness of the
reference service between health science libraries in Australia
phenomenon as well as the apparent assumption that findings
and the UK;15 a collaboration among libraries in France,
from English-speaking libraries can be generalized to other
Germany, and Poland; and a collaboration among reference
countries. This study, therefore, attempts to respond to a
librarians at Slavic and East European Library at University of
lacunae in international and comparative Library and Informa-
Illinois at Urbana Champaign, the Jagiellonian University
tion Science research research of virtual reference services
Library in Krakow (Poland), and the Russian National Library
should expand to a more global context.6 The importance of the
in St. Petersburg.16
study is that it describes commonalities and differences in
virtual reference services among several countries that are Issues in International Collaborative Virtual Reference
underrepresented in the literature. The study is also important
International collaborative virtual reference requires librar-
because it identifies challenges that global collaborative virtual
ians to be aware of cultural differences and information
reference services may face upon implementation. Achieving
behavior. However, studies examining the impact of cultural
these goals is significant for designing and implementing more
diversity on information seeking behavior in libraries are
effective and efficient global collaborative virtual reference
scarce.17 Scholars addressed the cultural diversity of users in
services. Our study analyzes, compares, and contrasts asyn-
terms of country of origin and ethnic background and focused
chronous reference transactions from Israel, Japan, and
on the effect of students heterogeneity on information seeking
Lebanon. We (1) identify the types of questions that virtual
behavior, research skills, and library usage.18 Ziming Liu
reference services are receiving in these countries; (2) develop
reported that international students in the United States, who
an international questions taxonomy; and (3) identify methods
were proficient in English and whose home countries are
used by librarians to handle virtual reference questions.
strongly influenced by American culture, encounter fewer
difficulties in using American libraries; this finding was
LITERATURE REVIEW
concurred by Mengxiong Liu.19 Mengxiong Liu and Bernice
Collaborative Virtual Reference Services Redfern found that students success level and English
Successful collaborations, state-wide, national, or interna- proficiency, frequency of library use, and frequency of
reference desk inquiry are positively correlated.20 Lucinda R.
tional, have been reported in the library literature. The
largest collaborative virtual reference service today is Zoe and Diane DiMartino reported that East Asian students
QuestionPoint, a joint effort by OCLC and the Library of request more assistance in searching full-text databases than
native English speakers.21 In short, studies have found that
Congress that is based on the Collaborative Digital
Reference Service (CDRS). When CDRS was launched in students from different countries have different perceptions and
2000, the Library of Congress and fifteen partner libraries exhibit different use patterns of the library. Barriers to effective
were involved;7 by the end of 2004 more than 1,500 communication in the library include language, conceptual
awareness of library services, and philosophy of education.22
libraries in twenty countries were participating in Ques-
tionPoint. QuestionPoint interface is now available in ten Investigators have reported major differences in how
libraries are used and perceived in different countries.23
languages, supporting the provision of virtual reference
services. Using QuestionPoint involves costs and benefits Librarians in North America are challenged to provide services
for international virtual reference partnerships.8 and library instructions to international students. One source of
In addition to QuestionPoint, many state-wide consortia in this difficulty is that many Asian students have no concept of
the United States and nation-wide consortia in other countries reference services and view the library as a place to store
required texts.24 Small cities and towns in many countries have
provide virtual reference. These include, for example, in the
United States: AskColorado,9 the Metropolitan Cooperative no libraries at all. Censorship, restrictive lending policies,
closed stacks, and service expectations vary tremendously.25
Library System (MCLS) in California, QuadANJ in New-
Jersey, Maryland AskUsNow!, Ohio s KnowItNow24x7, and For example, in Japanese schools the library is a study hall and
a library clerk is in charge of the room.26 Japanese students use
Washington state s Virtual Reference Service, to name a few.
Similarly, national collaborative virtual reference services exist the library mainly for ready reference questions; they rarely use
in Australia-AskNow!,10 and in the United Kingdom-Ask a it for research although some academic libraries in Japan (e.g.,
Librarian.11 Other European collaborative examples are: the Keio University library) offer active reference services. Yoshi
Biblioteksvagten in Denmark, Ask a librarian in Finland, Hendricks claims that due to the lack of attention to reference
Die Deutsche Internetbibliothek in Germany, Lapponica services in Asian countries in general, international students
from these countries rarely ask reference questions.27
Information Service in Lapland, Al@din in the Nether-
lands, Pregunte, Las Bibliotecas responden and La Another critical component to international collaborative
Biblioteca Responde in Spain, and Fr3ga biblioteket in virtual reference is the need for an interface in multiple
244 The Journal of Academic Librarianship
languages. Providing reference services in multiple languages ship (closed stacks and departmental collections and services),
reaches out to diverse user groups and helps them seek continued to serve as an academic and national library (Beth
information on their own turf.28 The need to provide bilingual Hasfarim Haleumi ), as well as the library for the Jewish
reference services and the challenges that an individual library people worldwide. In 2003 the national library and the
faces when trying to establish this service have led Tutor.com Hebrew University libraries became two independent institu-
to develop Bilingual Librarians by Request, a subscription tions. During the early years, most of the librarians at the
service that provides library customers with online access to Hebrew University were trained on the job and their role was
Spanish-speaking reference librarians. Table 1 lists efforts to perceived to be mainly clerical. This changed with the
provide chat virtual reference services in languages other than establishment of the first graduate library school in Israel at
English.29 the Hebrew University. Two other graduate library schools
With the aforementioned issues in mind, this study makes an were later established: Bar Ilan University (also offers an
undergraduate library diploma) and the University of Haifa.33
effort to identify similarities and differences among virtual
reference services in three countries (Israel, Japan, and All three library schools were highly influenced by American
Lebanon) in order to help in the planning and implementation librarianship and library education (open shelves and active
of international collaborative virtual reference services. In reference and instruction services). Sever and Sever and
particular, this study focuses on answering the following Sidorsky Dov discuss the history of libraries and library
research question: What information about users, librarians, education in Israel, and a description of its national, academic,
public, school, and special libraries.34
and libraries should global collaborative virtual reference
initiatives consider in order to provide effective reference The University of Haifa, whose library s virtual reference
services? service is examined here, was founded during the 1960s as a
branch of the Hebrew University and later became an
METHODOLOGY independent university. The University has six main faculties
The objective of this study was to identify similarities and (Humanities, Social Sciences, Law, Science and Science
differences among three countries and discover elements that Education, Social Welfare and Health Studies, and Educa-
international collaborative initiatives should consider when tion), and a Graduate School of Business. The University of
implementing global virtual reference services. Virtual refer- Haifa is the only liberal arts university in northern Israel and
ence service in this study refers to e-mail reference through a home to the most diverse student population in the country; it
mailto link or a Web form that users can fill out where an has a heterogeneous student body of approximately 15,000
indication has been made that users can ask reference undergraduate (66 percent) and graduate students (34
questions.30 We examined recent e-mail reference transactions percent).
in university libraries in Israel, Japan, and Lebanon: University The university library is considered one of the leading
of Haifa, Keio University, and the American University of libraries in Israel. The library collection includes two million
Beirut, respectively.31 This section describes the setting of these items. It maintains about 11,000 active journal subscriptions
libraries, explains the data collection, and discusses the content and 22,000 journal titles in various formats. Hebrew language
analysis method that was applied to analyze the reference periodicals include approximately 3,300 titles. The library has
transactions. access to the full text of thousands of periodicals and hundreds
of databases.
University of Haifa
The reference department handles more than 300 reference
Academic libraries in Israel go as far back as the 1920s requests daily and provides over a hundred sessions of group
with the establishment of Hebrew University.32 Since then, its instruction per semester. The virtual reference service began in
library, which was influenced with German style of librarian- 1998 as an e-mail service dedicated for faculty members, and
since 2000 it became available through a Web form for the
University of Haifa students, faculty, and staff. The library
reports an average of two e-mail questions per day that are
Table 1
handled by one librarian.
Virtual Reference Services in non-English Languages
Keio University
Language Service
Historically, libraries in Japan originated as collections of
Spanish Informatcion en vivo! (New York Public Library)
books. Currently all universities have campus libraries and are
Q and A Cafe (San Francisco area)
regulated under different rules based on the type of universities
(e.g., private or public). According to a report from the Ministry
Library of Congress (Spanish and Portuguese)
of Education, Culture, Sports, Science and Technology, on
AskColorado
average, there are 1.9 libraries per university.35 Although certain
24/7 Reference service limitations exist, almost all libraries in Japan are open to the
public. For example, unaffiliated students and faculty members
Swedish Net University (Sweden)
can use any academic library in the country if they have a letter
Catalan Universitat de Lleida (Spain)
from their own library system.36
Chinese Public libraries of Suffolk County (New York) As far as library education is concerned 296 universities
and colleges have been reported to be offering library
French Centre regional de services aux
certificate programs.37 Undergraduate students pursuing a
bibliotheques publiques (Quebec)
library certificate are required to complete twenty credit
New Brunswick Public Library (Canada)
hours of coursework in librarianship; this certificate is
March 2007 245
required only for individuals interested in working in public nursing, and social sciences. The language of instruction is
libraries but is optional for those interested in working in English.
other types of libraries. Ueda et al. also report that eight AUB libraries hold over 600,000 volumes, growing at a rate
universities provide graduate level education in Library and of 11,000 volumes annually. They subscribe to 2,600 period-
Information Science. Of these eight universities, Keio icals, of which 200 are in Arabic, and provide access to
University, Tokyo University, Kyoto University, University approximately 16,000 full-text electronic journals in over
of Tsukuba, and Aichi Shukutoku University offer doctoral ninety periodical databases. The libraries serve a population
programs in the field. of approximately 5800 undergraduate and 1,500 graduate
Keio University, whose library s virtual reference service is students (51.5 percent male and 48.5 percent female) from
examined here, was founded in 1858 and is Japan s oldest sixty-eight different countries and 400 full-time faculty
private higher education institution. It is a comprehensive members, most of whom hold their doctorates from the United
academic university with five major campuses in the Tokyo States. The libraries are open 90 hours per week. Researchers
area: Mita, Hiyoshi, Yagami, Fujisawa, and Shinanomachi. from all over the world use AUB libraries for their unique, rich,
Nine major academic units offer undergraduate, master s, and and historical collections, particularly on Lebanon and the
Ph.D. degrees and a wide range of academic disciplines, Middle East.
including business administration, science and technology, Like many other libraries in the Arab world, most of the
medicine, and law. Keio s University libraries serve approx- senior staff members at AUB are natives with MLS degrees
from ALA-accredited programs.39 As there is no formal
imately 28,000 undergraduate students, 4,100 graduate stu-
dents, and 1,800 faculty members. training program, new staff members learn by experience. It is
The original university library located in Mita was important to note here that most of the library s processes
founded in 1890. The Old Library, a red-brick Gothic have been automated only as of 1997. Electronic reference
building, was completed in 1912 for the celebration of Keio sources and services are equally new. The reference desk,
University s 50th anniversary. It is an important cultural which is staffed by three librarians and several other support
property and an enduring symbol of the University. Each of staff and student assistants, receives approximately 100
the five campuses has a library which is called Media questions per business day, mostly in person and by
Center. As a whole, media centers (including libraries) are telephone. E-mail reference started in 2000 and currently
considered one of the foremost academic information store- receives approximately four questions per week. Only the
houses in the country. The combined libraries of the Keio professional reference librarians answer e-mail questions,
Media Centers house more than 3.7 million volumes. The which are sent to the library via direct e-mail rather than a
library on Mita campus provides access to approximately Web form.
10,000 print journals, 19,000 electronic journals, 187 data- Data Collection
bases, and 1,627 e-books.
During the winter of 2005, the three libraries were
In 2004, 43,123 reference inquiries were made at five Keio
contacted by the researchers requesting recent e-mail refer-
University Media Centers. The e-mail reference service was
ence transactions for the study. Each library agreed to provide
launched on Mita campus in 1998, and the Web form was
the researchers with the data after personal identifiers [of the
introduced in 2002, to assist Keio University graduate
users and librarians] were removed. Because different libraries
students and full-time faculty members in searches for
provided different numbers of transactions, we used fifty of
information, books, and other materials. On Hiyoshi campus,
the most recent reference transactions provided by each
the e-mail service started in late 2002, and the Web form was
library, all of which were from 2004 to 2005 academic year.
launched in 2003. Eight out of ten reference librarians at Mita
Some transactions involved multiple correspondences
and all three reference librarians at Hiyoshi respond to virtual
between the users and librarians over a period of time; these
reference inquiries. They currently receive five inquiries per
were counted as one reference transaction. Some of the
month.
transactions were incomplete and could not be included (e.g.,
American University of Beirut (AUB) a reference request without the librarian s response and vice
Modern librarianship in Lebanon started in the nineteenth versa). The final sample consisted of forty-five requests from
century with the establishment of libraries serving newly Israel, fifty-one from Japan, and fifty from Lebanon, for a
founded academic institutions. Each library was influenced by total of 146 user requests and answers (ninety-six using a
the national origins of its parent institution. For example, AUB, Web form and fifty via direct e-mail) that were made during
which was established in 1866 (as the Syrian Protestant the 2004 2005 academic year. Of the 146 transactions
College), and the Lebanese American University, founded in examined, three were in Arabic, fifty-one were in English
1924 as Beirut College for Women, were influenced by the (most from Lebanon), forty-two in Hebrew (all from Israel),
American style of librarianship.38 and fifty in Japanese (all from Japan). It should be
AUB, whose library s virtual reference service is examined emphasized again that at Keio University and the University
here, was founded as a private, independent, non-sectarian of Haifa users fill a Web form when asking a reference
institution of higher education, functioning under a charter question whereas at AUB they send a direct e-mail to the
from the State of New York. AUB s main campus is reference staff. In contrast to direct e-mail, a Web-based
comprised of 81 buildings, including a hospital, four libraries, interaction helps librarians collect potentially useful informa-
three museums, and six dormitories, as well as an athletics tion about users and their information requests, such as
field. It offers over 100 undergraduate and graduate degree academic status (e.g., undergraduate, graduate, faculty, staff,
programs in agriculture and food sciences, architecture, arts, librarian), departmental affiliation, language of materials
business, engineering, humanities, health sciences, medicine, requested, and type of materials requested.
246 The Journal of Academic Librarianship
Data Processing and Analysis
Table 2
Content analysis of reference transactions was used to Coding Scheme
uncover key themes and elements and to compare results
Category
among the three countries. Content analysis, as any other
research method, has its strengths and weaknesses. Charles H. User Tool used to asked the question
Busha and Stephen P. Harter recommend that, in order for (direct e-mail versus Web form)
content analysis to produce good results, developing an
Language of interaction
appropriate content classification scheme is an important first
(e.g., Arabic, English, Hebrew, Japanese)
step.40 They and Bryce Allen and David Reser also advise
Academic or personal status
that the categories chosen should be exhaustive, mutually
(e.g., undergraduate, graduate, faculty, staff, librarian)
exclusive, clearly and accurately defined, and conceptually
valid in relation to the research question.41 It should be noted Departmental affiliation if any
that understanding the content of e-mail transactions is aided
Country for international users
by knowledge of the communication patterns of the people
being studied. Also, because communication differs from one Whether the search history was specified
culture to another, developing a universal coding scheme can Purpose of request (e.g., research paper, dissertation)
be very challenging.42
Number of questions per request
Number of transactions per request
Content analysis of reference transactions was Type of questions asked
(e.g., topical/subject searches, known-item searches,
used to uncover key themes and elements and to
technical problems, searching instructions)
compare results among the three countries.
Language of materials requested
(e.g., Arabic, English, Hebrew, Japanese)
Type of materials requested
We began data processing by attempting to apply existing
question taxonomies to our data. However, because existing Whether users identified specific publication
schemes did not fit our data, we developed a new coding dates for the materials requested
scheme. For example, the IFLA digital reference guidelines
Librarian Number of days taken to respond to an e-mail request
were initially used to code the librarians responses.43 The
Number of messages involved in answering a request
guidelines, however, were too prescriptive, focusing on what
librarians should do instead of what they actually do. In Language of interaction used in responses
addition, the IFLA guidelines do not address user requests and (e.g., Arabic, English, Hebrew, Japanese)
do not completely account for multiple interactions to respond
Whether the language of the response was
to certain requests. Eileen Abels tactics too were initially
consistent with the language of the request
examined, but could not be applied to most of our data.44
Number of librarians involved in answering a request
We also examined question taxonomies generated from
virtual reference transactions; however, these did not seem to Whether materials requested were found in library
be generalizable beyond the particular samples they exam-
Whether information requested was found in library
ined.45 The taxonomies vary considerably from each other and
from other taxonomies that have been proposed for traditional Whether information requested was provided
reference.46 In summary, because of the lack of a question (e.g., searching instructions, citing instructions,
taxonomy that could be used in a global context, this study library policies information)
developed its own based on virtual reference transactions from Whether referrals were made, and to whom
the three countries. The resulting coding scheme describes both (e.g., another librarian within the building,
users requests and librarians responses. another library on campus, or to another institution)
We classified both requests and answers according to
Whether the response includes reiteration of the
categories described in Table 2. We initially applied the coding
4user request
scheme to a sample of thirty transactions (ten from each
country) and then compared the results in order to standardize Whether concluding remarks were made
and refine the wording and use of the categories. The coding
Whether follow up initiatives were offered to users
scheme was modified several times using different sets of thirty
transactions; the scheme presented here fits the data found in
A graduate assistant coded a sample of fifteen full transactions
the sample across the three countries. It took four iterations to
reach consensus because of the variety in the ways people in in English to verify the coding reliability. The inter-coder
different countries communicate, provide, or ask for informa- reliability was close to 93 percent.
The primary limitation of the study is the number of libraries
tion. During these iterations, we translated requests and
responses to English to clarify possible subjective interpreta- and countries represented. Another limitation of the study is
tions. Once the level of agreement among the researchers that we make no comparisons with user groups in the United
States and other English-speaking countries. Despite these
reached an acceptable rate of over 90 percent, and definitions
limitations, this study is valuable as the first to address the
and guidelines were agreed upon, we coded the entire data set.
March 2007 247
social complexity of global virtual reference. Its comparison of these students are likely to spend less time on campus and use
e-mail reference services between different countries provides virtual reference services more.
insights for enhancing existing services. The coding scheme In Japan, access to university library materials is often
developed in this study should also be valuable to researchers restricted to affiliate users; unaffiliated users can make a
interested in studying global collaborative virtual reference. interlibrary loan requests for materials not owned by their
libraries. A reference letter is required for Keio University
RESULTS AND DISCUSSION students to visit other university libraries.
The results of this study are presented and discussed in three
sections: (1) user demographics; (2) content of user requests, Type of Questions
focusing on the type of questions asked and language used; and As Table 3 indicates, the most frequent type of requests in
(3) librarians responses. Before presenting and discussing these all three countries is for known items (fourteen requests, or 31
points, it should be emphasized that the comparative analysis percent, in Israel; twenty-four (47 percent) in Japan; and
of the 146 e-mail transactions revealed several similarities and nineteen (38 percent) in Lebanon). Yet, in Israel the library
differences among the three countries. Table 3 provides a received almost as many topical questions (twelve requests, 27
summary of all user data and Table 4 does the same for percent); in Japan topical questions are not so frequent (two
librarians data. requests, 4 percent), and in Lebanon there were ten topical
requests (20 percent). Three genealogy requests and many
User Demographics
requests for reproduction (eight requests) are evident in
Most of the requests in Japan and Israel are sent by affiliated Lebanon, compared with no genealogy request in Israel and
users but in Lebanon many unaffiliated users request reference Japan, as well as no reproduction requests in Israel and only
help. In Lebanon, over 56 percent of the users were from three requests in Japan. On the other hand in Israel and Japan,
eleven different countries, including thirteen requests sent by non-reference requests (eleven requests in Israel and eight
users from the United States. requests in Japan) are more frequent than in Lebanon (three
As Fig. 1 illustrates, user demographics vary among the requests). Thus, both the types and frequencies of requests vary
three countries. Most of the questions in Japan are sent by across the three countries.
faculty members and graduate students because the service is
limited to these user groups. In Lebanon, many of the users do
not reveal their academic background, primarily because
requests are sent by e-mail rather than a Web form requesting ...the most frequent type of request in all three
such information and because many requests in Lebanon are
countries is for known items.. .
from unaffiliated users. The requests in Israel are sent primarily
by students; only a few are sent by faculty. This may be because
faculty members in Israel have many other avenues to meet their
information needs. For example, at the University of Haifa, a A comparison of the transactions from the three libraries
reference librarian is exclusively dedicated to provide library indicates that the types of questions they receive vary
services to faculty members; the librarian attends faculty considerably (see Fig. 2). For example, in Israel many more
meetings and holds reference desk hours outside the library topical requests were sent by users than in Japan or Lebanon.
building in the faculty offices. In addition, collection developers This difference in types of questions asked suggests different
in each subject area provide reference support to faculty in each user needs in these countries. One possible reason for the high
department. It is possible that these faculty members largely rate of topical requests in Israel is the fact that the library users
fulfill their information needs in other ways too. must use English terms to search for subject information, even
Users demographics in the three countries may account for if the materials are written in Hebrew (or any other language),
some of the variation in the types of questions asked. For because the library catalog uses the Library of Congress
Subject Headings.49 In addition, most of the databases in the
example, many of the users in Lebanon are unaffiliated with
AUB and are residents of other countries (see Table 3). This library collection require users to search in English. Although
may explain why genealogy questions and questions for Israeli users are required to read English materials, searching
reproduction of materials are more frequent in Lebanon, requires a higher level of English proficiency, which the
compared with Israel and Japan.47 Users demographics also librarians can provide.
influence the type of questions sent to the virtual reference Another possible explanation of the differences in the types
service in Israel, where many questions are submitted by of questions asked is differences in perceptions of the library in
graduate students (Fig. 1). Compared to their proportion in the the three countries, particularly the perception of reference
user population of the institution, more graduate students use services. The user-centered approach in Israel and Lebanon
the service than undergraduates. This may be in part because (both are influenced by American librarianship and educational
graduate students conduct more research activities than under- systems), the perception of the librarians as facilitators, and
graduate students and because in Israel a typical graduate who help answering users information needs, result in more
student is older and non-traditional in comparison to the topical questions. Unlike Israel and Lebanon, in Japan, very
other countries. The average age of a student in Israel is older few topical requests were made. This reflects the expectation in
in Japan and the United States by approximately three years. Japan that the user will be independent and will not ask for help
Most of the first year undergraduate students have completed from librarians, especially not for basic information needs. This
two (women) or three (men) years military service.48 Thus, by attitude results in higher percent of availability questions and
the time they attend graduate school many students hold full known-item search requests than topical questions. It is
time positions and some of them are married or even parents; possible that known-item requests are more frequent in Japan
248 The Journal of Academic Librarianship
Table 3 Table 3 (continued )
Count Per Country on User Variables
Israel Japan Lebanon
n = 45 n = 51 n = 50
Value
Variable
Israel Japan Lebanon
n = 45 n = 51 n = 50
Value
Variable
Years of materials Not specified 38 51 49
Format E-mail 3 13 50 Specified 7 0 1
Form 42 38 0 Technical 7 3 7
problem
Language Arabic 0 0 3
Search instructions 1 3 5
English 3 1 47
Citing instructions 0 0 1
Hebrew 42 0 0
Verification 0 3 0
Japanese 0 50 0
Non-reference 11 9 3
Status Unknown 3 5 20
Reproduction 0 3 8
Undergraduate 19 4 3
Genealogy 0 0 3
Graduate 19 17 8
Other (type 2 8 2
Faculty 2 15 6
of question)
Staff 1 1 0
Number of One 42 40 44
Librarian 0 6 8
messages Two 2 9 3
Other 1 3 5
Three 1 1 2
Affiliation Affiliated 41 43 25
Four 0 0 1
Country from Israel 45 0 0
Five 0 1 0
which request Japan 0 51 0
originated
Lebanon 0 0 18
than Israel and Lebanon because of the perceptions of the
Other 0 0 15
library and the role of the reference librarian, as Hendricks
observed.50
United States 0 0 13
Unknown 0 0 4
Language
Academic major Yes 4 34 10
International collaborative virtual reference services do
of student No 41 17 40 consider language differences. However, this consideration is
done at one level by QuestionPoint and other international
Multiple messages 3 1 36
collaborative virtual reference services when it really should be
Greeting Yes 27 17 47
done at two levels. The first level is the language of transaction
No 18 34 3 and second is the language of materials available or requested.
While it can be assumed that the language of transaction and
Purpose of request Yes 5 7 15
the language of materials will be the same in general, this study
No 40 44 35
found that some transactions are written in one language to
Known item 14 24 19 request materials in another language (or languages). Several of
the transactions analyzed were conducted in two languages.
Topical question 12 2 10
For example, in Israel a few requests were made in English and
Search history Yes 10 15 3
replied to in Hebrew; in one case the librarian replied in
No 35 36 47 English to a request made in Hebrew. Fig. 3 illustrates the
language of transactions in each of the countries.
Language of Arabic 0 0 1
Another expectation is that the language of the reference
materials English 3 0 0 transaction will be the formal language of a country. In the
requested
sample from Japan, all transactions were made in Japanese,
Hebrew 7 0 0
but in Israel and Lebanon interactions were not always
Multiple 3 0 1
conducted in the formal languages of the country. Most of
Type of materials Books 4 0 0 the transactions in Lebanon were in English, which is a
Requested foreign language (the formal languages are Arabic and
Articles 7 0 0
French but the language of instruction at AUB is English).
Dissertations 0 0 0
Likewise, most of the transactions in Israel were conducted
Multiple 3 0 1 in Hebrew, which is the formal language for 80 percent of
the population (Arabic for the Arab population (20 percent)),
Other 0 0 1
and a few were conducted in English, which is a foreign
language. The student body at the University of Haifa
March 2007 249
Table 4 Figure 1
Count Per Country on Librarian Variables Academic Background of Users Per Country
Variable Value Israel Japan Lebanon
Apology Yes 2 6 1
No 43 45 49
Reiteration Yes 13 4 43
No 32 47 7
Search history Yes 3 18 7
No 42 33 43
Concluding remarks Yes 16 9