Christopher (Cal) Lee,
School of Information and Library Science,
University of North Carolina at Chapel Hill

Title:

Curation Junction, What’s your Function? Defining What it means to Do Digital Curation in order to Teach It

Abstract:

Information professionals are increasingly responsible for taking care of digital collections across their full life – from pre-creation design and planning to provision of access, potentially over long time scales.   The activities required to fulfill these responsibilities, often labeled “digital curation,” can occur within various professional, disciplinary, institutional, or organizational contexts (e.g. social science data archive in a university, commercial collection of scanned page images, state archives, private art museum, independent research center).  Graduate library and information science programs are uniquely positioned to prepare professionals who will be responsible for digital curation, but this will require new curricula and educational offerings, which may cut across traditional professional boundaries.

The DigCCurr (Digital Curation Curriculum) and DigCCurr2 projects -- both funded by the Institute for Museum and Library Services -- are developing conceptual frameworks, educational offerings, professional field experiences and doctoral-level research opportunities, in order to prepare the next generation of digital curation professionals.  One of the fundamental activities of DigCCurr is a detailed elaboration of the components of digital curation – i.e what it means to “do” digital curation.  Based on collection and analysis of data from numerous sources (documents, surveys, interviews), we have developed a Matrix of Topics for Digital Curation Curriculum.

This paper will summarize the motivation, methodology and data that have served as the basis for the DigCCurr matrix and will explain the components of the matrix, which is based on six dimensions: mandates, values and principles; functions and skills; professional, disciplinary or institutional /organizational context; type of resource; prerequisite knowledge; and transition point in the information continuum.   This paper will place strong emphasis on the second dimension: functions and skills.  What do our numerous data sources suggest the next generation of digital curation professions will need to do, and what are the implications for educators at schools that prepare information professions?

fdurrant2000@yahoo.com

Title
Teaching the Use of  Social Networks for Collaborative Library Activities – a study at the University of the West Indies, Dept of Library and Information Studies.

ABSTRACT

Among the large number of social networks which are developing, some social networking sites are dedicated to the needs of librarians or are meeting the needs of librarians.  These sites connect people based on the books which they share.

Research Problem
The popularity and intensive use of social networks, has encouraged librarians to establish presences on general sites such as Facebook, MySpace.  Others such as LibraryThing are designed to facilitate the execution of library activities.  As there is need to test the The Department of Library and Information Studies at the University of the West Indies has been studying the implications of the growing membership of social networks for libraries with a view to determining the best ways in which these applications can be incorporated into the regular activities of libraries.  This question is worth further study as libraries incorporate or consider incorporating Web 2.0 technologies for executing their functions and services.

Previous Scholarship
The literature on the topic of social networks and libraries has shown that  previous scholarship on the issue of social networking has placed emphasis on privacy and trust  and related concerns.  boyd and Ellison’s defining article identifies the main areas of research on as impression management, self presentation and friendship performance, networks and network structures, bridging online and offline networks, privacy, safety and trust. They also identified other research areas as race and ethnicity, religion, gender, sexuality shaping of identity, and use of social networks in libraries, schools and universities.

Online social networking sites have attracted membership involving users or potential library users.  The experience in incorporating such large numbers provides indications that libraries can possibly use SNSs to interact with users and to support collaborative activities.

At  the Department of Library and Information Studies, we have recognized the potential value of social networks to libraries, and have been undertaking research into the exponential growth of these networks, and the perception of people including librarians who make up the communities served by the libraries.  Social networks are therefore an area of research for some staff and students. In the last semester second year LIS undergraduate students used social networks as the subject of their research proposals, and graduate students are also doing research including collection development and social networks in academic libraries.

Data sources
Sites on Facebook and on LibraryThing will be used by our students to assess the experience in sharing books and interacting with others with similar interests, and to provide implications for practice.

Implications for practice
In recognition of the successes of these sites in facilitating interaction among large geographically dispersed groups of users, sharing content, collaborating with peers, and linking communities, the presentation will examine and assess  library applications of social networking sites, and their contribution to the development of effective interactive information resources.

Assistant Professor

Indiana University School of Library & Information Studies - Indianapolis (SLIS-Indy

katschil@iupui.edu

Title:
Promoting Leadership Potential Among New MLS Professionals:  Early Results from the Career Transitions Executive Leadership Program (CTELP)

Conference Theme:
Preparing students to work in new digital frontiers

Submitted to:
R. David Lankes, rdlankes@iis.syr.edu

Abstract:

In fall 2007, the Indiana University School of Library & Information Science at Indianapolis (SLIS) partnered with the Indiana Library Federation (ILF) and Indiana State Library (ISL) to implement a three-year, IMLS-funded project called the Career Transitions Executive Leadership Program (CTELP).  CTELP is built around a leadership-focused paradigm grounded in community apprenticeships (‘residencies’), and leadership and management education.  Each semester during the three-year program, approximately two to five highly-qualified MLS students are admitted to CTELP, typically during their last semester of graduate school.  MLS student ‘residents’ complete a full-time, one-semester library residency during which they participate in management-level operations and organizational decision making at the host site library.  Residencies are intended to provide experience and insights into library leadership and management that goes far beyond a typical internship.  Residents complete high level projects or advance initiatives such as program planning, grant writing, and so on.  Residency opportunities are available in all types of libraries including public, academic, law, health sciences, school media centers and others.

CTELP is characterized by its focus on sustained mentorship relationships.  Each student resident is paired with the library director or associate director at the host site, who contributes significant time toward mentoring the resident’s professional development.  The network of host libraries and professional mentors provides an invaluable tier of support, mentoring students and working with SLIS and partners to improve the scope and quality of LIS education.

Simultaneously with the on-site residencies, CTELP students participate in a one-semester, three-credit leadership course, conducted primarily online by SLIS faculty.  The coursework provides a forum through which students explore and reflect on issues related to effective organizational management and leadership, mentorship, and civic engagement, as well as a forum for students to support one another during their residencies.  CTELP students receive a generous stipend and tuition support, as well as nine graduate credits for their residency and coursework.

Through their engagement with professional mentors, residency projects, and coursework, CTELP students gain experience and insight into the challenges of leading an organization in areas such as organizational effectiveness, diversity, ethical decision making, human relations, employee performance, process improvement, and resource allocation.  In addition, CTELP prepares graduates to assume leadership and managerial positions in libraries and information centers upon the completion of their MLS degrees.

During their resident’s semester, host site mentors complete mid-term and end-of-the-term assessment surveys on which they record their perceptions of the resident’s leadership skills, attitudes, and potential.  All residents also complete pre- and post-residency leadership self-assessment surveys, and post- assessments of the residency experience itself.  Residents are then tracked semi-annually for three years to determine to what extent they engage in leadership activities within their organizations and communities as they embark on a professional career post-MLS.  This paper describes CTELP and presents early survey data, illustrating important similarities and discrepancies between students’ and practitioners’ perceptions about library leadership and the roles of MLS education in preparing library leaders for the future.

Terrell G. Russell and Cassidy R. Sugimoto

unc@terrellrussell.com, csugimoto@unc.edu
School of Information and Library Science, UNC-Chapel Hill, Chapel Hill, NC 27599

Abstract

Academic genealogy is valuable because it provides context, history and has the potential to predict future trends in the field.  However, it is most commonly done casually and without the rigor to provide a platform for discussion beyond the anecdote.  This paper presents a novel technique for calculating genealogical scores for individuals and academic “families.”  This data-driven technique provides a platform for greater contextualization and insight into an academic’s legacy.

1.    Introduction
Academics have long considered their place in the pursuit of knowledge.  They are aware of who have come before them and they are continually working at the edge of knowledge to carve out their own place.  The literature review is an integral part of any research and contextualizes the current work and provides a framework for understanding what is new.  Academic genealogy does the same; it provides the observer a better view of the influencing powers in a researcher's life and course of study.

However, much of the information regarding academic genealogy is anecdotal and orally disseminated, making it difficult for new researchers in a discipline to identify their place in an historical context and for those studying a field to have a comprehensive picture of the academic family tree.  There has also been very little in the way of quantifying family trees as a method to gain a contextualized perspective of the impact of an individual or branch within a particular academic community or discipline.

This lack of quantification hinders a researcher's ability to see patterns and prevents data-driven discussion around academic genealogy.  The MPACT project (http://www.ils.unc.edu/mpact) now calculates seven family tree metrics discussed in this paper and offers a first attempt at providing this much needed quantification.  MPACT provides these metrics, a visualization of an individual's academic family tree and open access to the data for querying and further investigation by third parties.

2.    Literature Review
Kealy & Mullen (1996) proposed the idea of macro-mentorship—in which mentorship is considered on the level of an organization of mentoring relationships, rather than purely the one-to-one, mentor-mentee relationship.  In their description of macro-mentorship, they offer the idea of the academic family tree as one example of macro-mentorship.  They argue that by making observations at the macro level we can implement change to enhance mentorship on the micro level (the one-to-one relationships).  They encourage the study of academic lineages to help individual scholars place themselves and others within a broader, historical context and for institutions and disciplines to examine their distinctive cultural histories.

However, there have been few published studies on academic lineage/genealogy.  The majority of the published work has traced the lineage of a particular individual—either a contemporary scholar tracing back through their ancestry or a study exploring the family tree of a certain “patriarch” in a given field (Stuart & Pierce, 2006; Bennett & Lowe, 2005; Stella, 2001; Tyler & Tyler, 1992).  There are also numerous web projects of this kind (i.e., http://www.cs.usask.ca/~mould/lineage/lineage.html).

Some studies have limited the focus to a particular department within a university, such as the study of chemistry faculty at the University of Wisconsin-Madison (Rocke & Ihde, 1979) and the psychology department at the University of Western Ontario (Innis, 1988).  Similarly, there have been a few databases devoted to the study of faculties, for example, the chemistry faculty at the University of Texas at Austin (http://www.lib.utexas.edu/chem/genealogy/).

Even fewer published studies exist on the disciplinary level.  One such study, on American field primatologists, utilized a combination of survey methodology and formal/informal communication to graph 672 names in the field of primatology (Kelley & Sussman, 2007).  Two papers within the field of psychology have focused on tracing academic lineages to identify contributions and ideological perspectives over time (Williams, 1993; Robertson, 1994).  The field of physical education teacher education (PETE) was studied by identifying the key “elders” in that field and tracing their lineages (Mitchell, 1992).  Additionally, there have been a few disciplinary web projects, including the Artificial Intelligence Genealogy Project at the University of Texas at Austin (http://aigp.csres.utexas.edu/~aigp/) and the Mathematics Genealogy Project at North Dakota State University (http://www.genealogy.math.ndsu.nodak.edu/index.php).

No previously published research has generated a technique or methodology for calculating and quantifying academic genealogy.  This paper presents seven novel metrics using the MPACT database and the field of Library and Information Science.  This database (http://www.ils.unc.edu/mpact) at the University of North Carolina at Chapel Hill captures dissertations, advisees, committeeships, school, year, title, abstract, and discipline for academics.  It was originally focused on the field of Library and Information Science but has recently broadened its scope to include other disciplines.  The dissertations and academic connections collected in the MPACT database were gathered through a process that includes extensive querying of UMI's database as well as interlibrary loan and physical inspection (see Marchionini, Solomon, Davis, & Russell, 2006 for a more thorough discussion of the collection techniques).  When documents proved hard to find, personal contact was made with the university where a dissertation was published as well as with the authors themselves on some occasion.  This is a time-intensive collection process but allows MPACT to house the only known queryable repository of committeeships at this time.  As the interdisciplinarity of the dataset improves, MPACT hopes to become the platform upon which the interrogation of academic genealogy can flourish.

3.    Description of Metrics
In order to quantify the output of academic genealogy, seven different metrics have been proposed:  A score—the sum of the number of times an individual has served as an advisor on a dissertation; A+C score—the sum of the number of times an individual has served on a dissertation in any capacity (either as an advisor or committee member); T score—the sum of the total number of descendants in an individual’s family tree; D score—a calculation of decayed influence on descendants in an individual's family tree; G score—the sum of the number of generations of descendants of an individual; W score—the sum of descendants in the largest generation in an individual's family tree; and the TA score—the sum of the number of descendants in an individual’s family tree who went on to become advisors themselves.  With the exception of the A+C score, all the above metrics are comprised of only advisorships as the effects of committeeships on academic lineage are more personal and limited in scope with respect to an entire discipline. These metrics provide seven different angles by which one can quantify the impact of an individual’s lineage, identify prominent branches, and describe the nature of academic lineage within a discipline.

3.1. A score (advising score)
The A score is calculated by summing the number of times an individual has advised on a dissertation.  This number is limited to advisorships alone and does not take into consideration times when the individual served on a dissertation committee as a committee member.  An A score of zero indicates that an individual has never served as an advisor on a dissertation committee.

For example, Figure 1 presents an academic lineage within the MPACT database (an unmarked version of the family tree for this individual, Jesse H. Shera, is presented in Appendix A).  In order to calculate the A score for the head of this academic tree, one would count the number of direct links from that individual.  As shown in Figure 1, counting from the left to the right, there are five links directly connecting the single individual to the next generation.  Therefore, the individual would have an A score of 5.

The A score identifies the degree to which a given advisor has impacted the field on the individual level, by calculating their direct impact on a single generation of scholars.  A ranked output of the highest A scores in the discipline represents those who are individually the top “producers” in terms of doctoral student output (see Appendix B for a list of the top 50 A scores within the MPACT database).

3.2. A+C score (advising and committeeship score)
The A+C score is calculated by summing the number of times an individual has served on a dissertation committee in any capacity.  Unlike the A score, this number also includes those times in which an individual served as a committee member on a dissertation.  An A+C score of zero indicates that an individual has never served on a dissertation committee.

Figure 2 presents the calculation for the A+C score.  Similar to the A score, the number of direct links from the individual to the next generation are calculated (in this example, it remains 5).  However, added to this are the number of direct committeeships found in the MPACT database (not shown in the figure).  The sum of these two numbers comprises an academic’s A+C score.  For the example shown below, Jesse H. Shera directly served on 7 committees, making his A+C score 12.

The A+C score, like the A score, is indicative of the direct mentoring contributions of an individual.  A ranked output of the highest A+C scores would also represent those who are top producers on the individual level (see Appendix B for a list of the top 50 A+C scores within the MPACT database).

3.3. T score (tree score)
The T score is calculated by summing the number of descendants in an individual’s family tree, not including the individual themselves.  The T score is the simple sum of all the descendants in the tree regardless of generational distance.  A T score of zero indicates that an individual has never served as an advisor on a dissertation committee, and therefore has no academic descendants in their tree.

Figure 3 shows a calculation of the T score.  There are five links from the individual to the first generation of descendants, ten links from the first generation to the second, six links from the second to the third, and seven links from the third generation to the fourth generation.  The sum of all these links is 28, the T score for the example.  Another way to examine this would be to count the number of boxes (proxies for the people) in the family tree.  The total number of boxes, not including the head individual, would make up the same score (28).

The T score provides a description of the total size of a family tree.  Thus, as a cumulative measure, no advisee will ever have a T score larger than their advisor's.  The individuals with the highest T scores will therefore represent the patriarchs of the field—those at the head of long-standing and productive academic lines (see Appendix B for a list of the top 50 T scores within the MPACT database).

3.4. D score (decaying tree score)
The D score, similar to the T score, is based on the number of descendants in a family tree.  However, instead of providing a value of one for each descendant (like the T score), the D score is derived from summing partial values.  That is, each descendent generation from the individual of interest is given one-half the value of the preceding generation.  Therefore, an individual receives a value of one for those they directly advised, a value of one-half for those advised by their advisees, and a one-quarter value for each of their advisees, with continued decay through the generations.  A D score of zero indicates that an individual has never served as an advisor on a dissertation committee, and therefore has no academic descendants in their tree.

Figure 4 provides an example of the calculation of a D score for Jesse H. Shera.  As can be seen in the example, each person in the first generation is counted once (5).  The value of the second generation is halved (10 becomes 5).  The value of the third generation is halved again (1/4) for a total of 1.5.  The seven individuals of the fourth generation provide additional value of .875.  The sum total of these four generations is the D score, 12.375.

The D score also provides a description of the total size of a family tree.  The D score, however, gives greater weight to those who have a high number of direct advisees.  Therefore, unlike the T score, an individual can earn a higher D score than his own advisor since his advisees will count more for him than for his advisor (see Appendix B for a list of the top 50 D scores within the MPACT database).

3.5. G score (generations score)
The G score is the sum of the number of generations of descendants in an individual's family tree.  An individual who has never advised will have zero generations in their family tree and a G score of zero.

As shown in Figure 5, the G score is a simple sum of the number of generations following an individual.  In the case of Jesse H. Shera, there are four generations of scholars within his family tree, giving him a G score of 4.

The G score characterizes the depth of a family tree.  Those with the greatest G scores will represent the longest academic lines in a discipline (see Appendix B for a list of the top 50 G scores in the MPACT database).

3.6. W score (width score)
The W score examines the width of a family tree, by measuring the largest generation in an individual's family tree.  A generation is represented by all the people who are the same number of generations below an individual.  A W score of zero indicates that an individual has no advisees.  A W score of one indicates that no generation of descendants had more than one person in it.

Figure 6 provides an example of the calculation of the W score.  As shown in the Figure, the number of individuals in each generation is calculated—in this case, the generations have 5, 10, 6, and 7 individuals in them, respectively.  The generation with the greatest number of individuals within a tree is used to calculate the W score.  In this case, the largest generation has 10 members, giving the tree a W score of 10.

The W score is reflective of the culture of a family tree—for instance the W score can identify those trees in which a large burst of generational productivity can be observed (see Appendix B for the list of the top 50 W scores within the MPACT database).

3.7. TA score (advisors on tree score)
The TA score is calculated by counting the number of descendants in an individual's tree who have served as an advisor themselves.  An individual who has never advised a student who has in turn advised a student of their own would have a TA score of zero.  A TA score of one or higher indicates that an individual has at least one academic grandchild.

An example of a TA score is shown in Figure 7.  Each individual who has one or more links (representing having served at least once in the position of an advisor) receives a count of one.  Therefore the TA score is the total number of individuals who served at least once as an advisor in a family tree—in this case there are five individuals in this tree (not including the head of the tree) who have served in the position of an advisor.  The TA score for Jesse H. Shera, therefore, is 5.

The TA score is indicative of the impact that an individual has made on an academic discipline.  Whereas other scores count each descendant, this score only counts those descendants who remained in academe and contributed by advising dissertations themselves.  The TA score is thereby the best single indicator of the impact an individual has had on producing a lineage of academics (see Appendix B for a list of the top 50 TA scores within the MPACT database).

4.    Discussion and Future Work
The metrics presented above provide only a stepping stone into further discussion and comparison to other metrics.  The interdisciplinarity of the design will allow this database to provide a unique perspective on the study of academic genealogy.  For now, however, the field of LIS is nearly completely mapped and can be seen as a valuable example of the type of analysis that can be done with this type of data.

The list of top A and A+C scores provide a detailed lists of the patriarchs/matriarchs in the field.  Beyond its inspirational value, it allows someone new or unfamiliar with the field to immediately find those mentors who have contributed their knowledgebase to many generations of scholars.  The duplication in the names on the top of the lists from the A and A+C scores (as well as with later scores) makes it clear very quickly who the people who shaped the field of LIS are and the particular ways in which they shaped the field.  As the field matures, it will be valuable to notice whether the duplication continues to exist.

The list of Top G scores reinforces the fact that LIS is a relatively young field with the deepest tree in this dataset only having five generations.  This could also be, in part, a reflection of the fact that LIS is a rich multidisciplinary field with doctoral graduates who find rewarding work in many other areas.  For fields that are more focused and/or not as applicable in industry, these scores may be much higher representing much longer, insular branches of academic scholarship.

The Top D scores provide a more temporally local version of the Top T list.  It seems that there is an avenue for further research by tweaking this particular function of decay (currently set at 0.5 per generation) to find a sweet spot of “influence.”  As long as the Top D list looks extremely similar to the Top T list (where advisees can never overtake their mentors), it provides little additional value.  A manipulation of the rate of decay may allow the Top D list to become more interesting and valuable.

The Top TA list shows that having students who produce their own students is very hard and rarely done.  The top six scores on that list are five sevens and a single nine.  As a measure of impact on a field, this TA score represents a very hard ladder to climb; a ladder that takes a lifetime of good people (and good funding?) to scale.  The mentors atop this list have truly provided great service to the academic discipline.

As the database of collected mentorship connections grow richer and denser across disciplines, the seven metrics (A, A+C, T, G, W, D, TA) should allow insight and research never possible before.  Some of this work may include thematic comparisons of branches of family trees.  Does classification research appear obviously prominent when perusing the list of people in the MPACT database?  Where does human-computer interaction research start to appear on the scene?  Are those branches shorter since the researchers came from other fields (engineering, computer science, education)?

Additional work can also be done comparing the emergent features of different disciplines.  Do old disciplines show differing characteristics from younger ones?  Can we see the academic trail of biochemistry branching and merging from chemistry and biology?  What are the correlations between mentoring and family trees and the Big Science model, whereby large numbers of students are funded and working in a lab under a single advisor (who has the grant money)?  Does the W score increase because the advisor has 10-30 graduate students under him at any one time?  How do newly emerging humanities-based fields, such as Women's Studies, compare in this regard?

The technique proposed in this research provides a firm foundation in identifying, quantifying, and contextualizing mentorship and academic genealogy within the academic community. It is hoped that this technique will be utilized to answer the questions above and other questions that have not yet appeared.
Works Cited
Bennett, A.F., & Lowe, C. (2005). The academic genealogy of George A. Bartholomew. Integrative and Comparative Biology, 45(2), 231-233.
Innis, N.K. (1988). Academic Genealogy of the Department of Psychology, University of Western Ontario. London: Department of Psychology, University of Western Ontario.
Kelley, E.A., & Sussman, R.W. (2007). An academic genealogy on the history of American field primatologists. American Journal of Physical Anthropology, 132(3), 406-425.
Kelly, W.A., & Mullen, C.A. (1996). Re-thinking mentoring relationships. (Available from the ERIC Document Reproduction Service No. ED 394 420).
Marchionini, G., Solomon, P., Davis, C., and Russell, T. (2006). Information and library science MPACT: A preliminary analysis. Library & Information Science Research, 28(4), 480-500.
Mitchell, M.F. (1992). A descriptive analysis and academic genealogy of major contributors to JTPE in the 1980s. Journal of Teaching in Physical Education, 11(4), 426-442.
Mould, D. (2008). Academic Lineage. Retrieved January 25, 2008 from http://www.cs.usask.ca/~mould/lineage/lineage.html.
North Dakota State University (2007). Mathematics Genealogy Project. Retrieved January 25, 2008 from http://www.genealogy.math.ndsu.nodak.edu/index.php.
Rocke, A.J., & Ihde, A.J. (1979). A Badger Chemist Genealogy: The Faculty at the University of Wisconsin-Madison. Journal of Chemical Education, 56(2), 93-95.
Stella, V.J. (2001). My mentors. Journal of Pharmaceutical Sciences, 90(8), 969-978.
Stuart, D.G., & Pierce, P.A. (2006). The academic lineage of Sir John Carew Eccles (1903-1997). Progress in Neurobiology, 78(3-5), 136-155.
The University of Texas at Austin (2007). AI Genealogy Project. Retrieved January 25, 2008 from http://aigp.csres.utexas.edu/~aigp/.
The University of Texas at Austin. (2007). Academic Genealogy of Chemistry Faculty. Retrieved January 25, 2008 from http://www.lib.utexas.edu/chem/genealogy/.
Tyler, V.M., & Tyler, V.E. (1992). The academic genealogy of Schwarting, Arthur, E., Pharmacognosist. Journal of Natural Products, 55(7), 833-844.
Williams, R.B. (1993). Contributions to the history of psychology: tracing academic geneaology. Psychological Reports, 72(1), 85-86.

Abstract:
This paper suggests that a Post-Lacanian perspective provides a useful theoretical and methodological approach for understanding reflexive information architectures—one key example being those that draw on second-generation Web technologies. In following, the Post-Lacanian perspective is useful for LIS education and research in several ways. Understanding information architectures in various contexts adds value to overall research in the organization of information. Many of those contexts are information environments that include reflexive information technologies. The Post-Lacanian perspective—complemented by older LIS research—is useful to help understand those technologies with a diversely rich perspective. This paper explains that perspective by describing its contribution to information ethics, design, and organization. This paper also contends that it provides a useful tool for practical work. It uses theoretical and empirical analysis as evidence.

Introduction
In May of 2007 the crafting website Ravelry launched. The site, “a place for knitters, crocheters, designers, spinners, and dyers to keep track of their yarn, tools, and pattern information” (Forbes, 2007), includes features characteristic of many social networking sites: it allows its members to become virtual friends, provides forums for discussion, and generates content through user contributions. It has become a popular website with over 50,000 active users, 3,000 active community groups, and 30,000 new forum posts every day (Petrovski, 2008, p. 52). The site was originally conceived and constructed by two people, and now is maintained and developed by four. While Ravelry incorporates aspects similar to other social networking sites, it also includes distinctive features—one example being an organizational system for users to list their personal craft supplies and books. This blending of old and new is a blending which leverages traditional LIS organizational techniques and knowledges from other disciplines as diverse as engineering and psychology (Morville & Rosenfeld, 2006, pp. 3-12). Like many modern websites, Ravelry consists of a complex information architecture that uses many of these traditional LIS organizational techniques while also expanding and manipulating them into newer hybrids.

This hybridity signals one difference between the organizational systems of modern websites and those traditionally studied in LIS; these sites include technologies with materialities that are more reactive to their user community than many traditional organizational systems that use taxonomies, hierarchies, and other classifications. These newer organizational systems are more reflexive to their users (Campbell & Fast, 2004). This reflexivity presented by World Wide Web technologies has generated interest among diverse scholars and practitioners, though it is often described in terms of those disciplines’ vocabularies (Arbogast de Hubert-Miller, 2006; G. Burnett, 2000; G. Burnett, Dickey, Kazmer, & Chudoba, 2003; Gary Burnett, 2002; Campbell, 2007; Robins, 2002; Jenkins, 2006; Sunstein, 2007).

Reflexivity means that the information architecture reacts—often within seconds—to a user’s interaction, which in turn alters future user interactions. One example is social bookmarking: as users label their own content, the taxonomy of a larger information architecture is changed, affecting future taxonomy use (Golder & Huberman, 2006). Although differences exist between newer and older information architectures, these reflexive sites don’t necessarily encourage a drastic change from past organization of information research. They do, however, promote new research questions. Information architectures that categorize library holdings generate different research questions than those from information architectures that are used for social bookmarking.

This paper argues that a Post-Lacanian perspective provides an important perspective for understanding the reflexivity of modern Web information architectures. The perspective uses a rich vocabulary that can be recycled from its original context to account for the reflexivity of Web technologies in relation to their users. It can be used to collapse streams of research that often are often separate streams of study: literacy, information retrieval, user behavior, information ethics, documentation, technology use, social studies of information, and organization of information. Because the objects of these studies co-occur in reflexive information technologies, combining the methods and theories of these researches allows new insight of interaction with these technologies. For example, many Internet search engines reuse user-entered keywords in order to retrieve more relevant results in the future.

This theoretical work is important for both education and research because it provides practical methods of analysis for the classroom and theoretical perspectives for research; it reformulates former perspectives and approaches to understand reflexive information architectures. One part of this paper makes that case through empirical research. A pilot study analyzing information organization and behavior from Ravelry provides data. A second part of this paper uses theoretical analysis to redefine past research. The theoretical and conceptual work of the past is valuable for these new research questions. Re-envisioning the modern Web as a new information phenomenon is unnecessary, but a theoretical propaedeutic helps to understand newer and more reflexive information contexts. This project expands past LIS researches for reflexive Web systems: theory from the cognitive/sociocognitive (Anderson, 1985; Ellis, 1992; Hjorland, 2002), engineering (Harman, 1997; Over, 2001), ethnographic (McKechnie, 2000; Pettigrew, 2000), critical (Bowker, 2005; Bowker & Star, 1999), and other perspectives can be reused for these reflexive Web architectures.

One premise of this argument is that many methods used in information studies research in the past are contextualized within that research but can be modified in useful ways, much like current Web hybridities are often modified from traditional organizational techniques (Morville & Rosenfeld, 2006). Theories like Chatman’s Information Poverty (1996), Savolainen’s Everyday Life Information Seeking (1995), and Bate’s Berrypicking (1989), do not cease to provide intellectual focus; different technologies do not signal the end of traditional concerns or a break with the past (Smith, 1994). This paper bridges past theoretical work with the current intellectual work in order to ground new researches in the intellectual lineage of this discipline.

The following section develops key terms of the Post-Lacanian perspective while discussing how those terms may revision established information studies theory. It suggests strengths and weaknesses of its use. While beyond the scope of this paper to demonstrate this revisioning for all past research, focusing on milestones in past research provides inductive examples.

Retheorizing for Reflexivity
This perspective provides a compliment to existing LIS research. While reusing past research, it does not provide a metatheoretical perspective for all information studies. The Post-Lacanian perspective inflects research through its unique vocabulary. It highlights certain affinities while suppressing others. The following foregrounds these research affinities by providing a description of its theorizations and methods while simultaneously describing how it reuses past information studies research and methods.

Theories of human information behavior are a basic component of nearly all information studies research—either overtly or tacitly. The Post-Lacanian triad of Imaginary, Symbolic, and Real provide a way to retheorize those theories while complicating their relationships. The Imaginary register of that triad roughly resembles theories of human cognition as understood in the cognitive paradigm (Anderson, 1985; Ellis, 1992). As an individual makes sense of their environment, the conscious choices they make with their sensory input produces the Imaginary register. Bates’s seminal berrypicking article provides an apt example (1989). Some question this perspective’s legitimacy for understanding information behavior, suggesting it is too simplistic to explain human behavior (Frohmann, 1992a, 1992b; Day, 2005, 2000).

The Symbolic register attempts to theorize the complex relationship between material environments and perceived cognition. Because users make decisions that depend on subjective experience and material environments, each individual will interpret their behavior and goals differently. Some important material environments are classifications, information systems, and most importantly in this research, information architectures—amalgams of technologies, techniques, and vocabularies. Therefore those environments are a part of the Symbolic register. Those environments of the Symbolic register help shape the decision-making taking place in the Imaginary, providing tangible and material impedance to user behavior. While information studies behavior may place the Symbolic on the system of information, the Post-Lacanian perspective focuses on the grammar of language (Lacan, 1977, 2006; Leupin, 2004; Fink, 1999, 1995, 1998; Skelton, 1995). Focusing on grammar over information system lends itself to a unique view of information structure: grammar structures behavior but is simultaneously structured by that behavior through its repetition by the person. The Post-Lacanian perspective becomes palpable within information studies as the information architectures that users interact with begin to readjust their behavior to the user. To sum, information architecture alters the user’s behavior while remaining malleable to new contingent uses of the architecture which would recreate the rules anew: reflexivity. This loop of architecture reflexing to information behaviors is a key point in Post-Lacanian theory that provides a way to understand the complex relationship between the user and the organization of information in many modern information technology environments. The information architecture is “structured like a language” (Lacan, 1977, p. 203).

Indicating information architecture is structured like a language is to suggest that relationships between its components begin to resemble those of language. Rhetorical devices like metaphor embed between components of the architecture. Modern Internet technologies provide examples. User input into the Google search engine transforms keywords into an output that consists of links and text. A linguistic displacement occurs as the engine produces results that suggest an inherent connection between the input and the search results, a transformation of language. Reflexive technologies place even more importance on the natural ability of humans to reuse language in new ways. A bond between Symbolic and Imaginary is demonstrated here.

The final component of the triad, the Real, bears the most resemblance to various strains of ethical research in the organization of information. It describes what the ethical orientation of the system itself is. Kittler uses a metaphor to describe this orientation (Kittler, 1999). If the Imaginary is a typewriter a person manipulates to create a screenplay, and the genre of screenplay (drama, comedy, horror) is the Symbolic form that affords certain types of descriptions, the film is the Real which provides an atmospheric side-effect of the medium and is largely unnoticeable.

Žižek indicates the Real may only be approached through comparative interpretation (2005). For example, it’s unlikely that early 20th century writers considered the static or snow of the film as an important part of their screenplays, yet today it is noticeably applied to create a quality that represents the aura of an era’s communication. It is unlikely the audience would have noticed that the black and white screen projected certain stereotypes and affinities at the expense of others. In following that idea, information architectures contain an unknowable Real that is important for shaping its culture and knowledge. Looking back to the nascent web of 1991 or comparing library OPACs to the vastly different interfaces available in today’s web would help to theorize the various registers of the Real. Because the Real is structured by the vestigial traces of the Imaginary and Symbolic registers, identifying its work becomes an ethical project bound up the Post-Lacanian perspective.

This section provided an introduction to some aspects of this perspective while describing its relationship to information studies research. It is a vocabulary to begin interpreting and start building more useful concepts. This list of terminology and theoretical ideas is far from exhausted. The following describes the beginnings of empirical research that expands on these ideas. It uses both qualitative interviews and critical content analysis. Members from the Ravelry user community provided material for interpretation.

Empirical Application of the Post-Lacanian
Theorizing the Imaginary of Ravelry
Elsa works full time, parents one child, and has been knitting for several decades. She considers herself a real knitter, a phrase she used to suggest that knitting is an essential part of her life. In that regard, she suggests that she can sense a kinship in other real knitters. Elsa is also a podcast enthusiast, an activity that overlaps with her affinity for knitting. She said that one of the favorite parts of her day is drinking coffee and listening to knitting podcasts in the morning.

Those podcasts helped her discover Ravelry in early 2008. After hearing about the site during various programs, she decided to explore. She first signed up for a user account through Ravelry’s invitation system. Unlike many Internet sites, Ravelry does not provide immediate access to its content and requires that participants wait several days or weeks after they register personal information in order to use the site. Elsa, and other interview participants, suggested that the wait created anticipation. But in fact, she had forgotten that she had event registered until she received her email invitation several weeks later. Elsa then returned to the site and started exploring.

Recounting her experiences since that time, Elsa said that Ravelry had little impact on her previous knitting habits. She continues to knit a similar quantity and variety of projects. For her, the site was a supplement to her already plentiful knitting hobby. In accord, she uses the site at her leisure, choosing to ignore the technologies that she finds less useful to her previous activities, such as those that allow her to list her variously gauged needles online. If she finds little immediate use for a technology, she doesn’t attempt to use it. To illustrate, Elsa did suggest that she enjoys several functions of the site that complement her existing hobby. She lists the skeins of yarn stored at her house in an online database. Because of that, she said that she is now better able to find her yarn due to its documentation on the site—Ravelry acts an inventory. She also enjoys uploading pictures of her finished projects so that people can “ooh and aah” at them, a feature she enjoys because she considers the members of Ravelry more appreciative than people like her husband or work colleagues.

The cognitive perspective focuses on describing these activities, attempting to understand why Elsa reasons that some activities are relevant to her Ravelry use while others are not. And as suggested above, this also exemplifies a description of the Imaginary register. Elsa makes conscious choices towards certain ends; she can clearly articulate and manipulate her environment. But it is Elsa’s insistence that she only uses the site as a supplement that the Symbolic register can be perceived. In her interview, Elsa had suggested she currently uses Ravelry to participate in discussion forums. She estimated most her time on Ravelry is devoted to writing messages to other users about a wide range of topics. Some of these discussions are about food preparation; some are about politics; some are about knitting and craftwork. The longer Elsa used Ravelry, the more that her original desire to use Ravelry in support of her knitting was supplanted by her desire to participate in discussions facilitated through Ravelry’s architecture. Another interview participant helped demonstrate the site’s ability to supplant original desire with a different desire for community interaction. That participant recently needed to reduce the time she spent on the site; when forced to choose, she cut back from her online knitting activities in order to participate in online discussions.

The Symbolic register helps explain this phenomenon. Like many social networking sites, Ravelry’s vocabulary and site architecture support community interaction—the knitting is a support to this vocabulary and acts as an initial enticement. Ravelry’s knitting features become the objet petit a of the site. The Symbolic linkage supporting community interactions is mobilized but largely unaffected by the knitting content. This is alluded to by the recent proliferation of social networking sites on the World Wide Web. Their content could often be replaced; the same information architecture framework is reused in numerous social networking sites. And it is. As developers work on new ways to reuse content from those sites, they focus on the linkages instead of the content (Keith, 2008). Studies of social networking sites demonstrate the value the community linkages and social identity over content (Mika, 2005). It is only when the subtraction of users and their community interlinking they have generated that an amalgam of broken links become noticeable in the site. In other words, Elsa’s knitting is minor for the overall site structure so long as it supports a social networking purpose. Evaluating her user dislike of listing her needles is not fully explained by the fact that she says it doesn’t support her hobby. A question to answer is why her current Symbolic relationship with Ravelry’s architecture does not support the listing of her needles.

The importance of an interlinking supporting communication is further exemplified when Elsa later spoke about the mail functions within Ravelry. Elsa said that she doesn’t enjoy using that function because it doesn’t give her as great a flexibility to connect to the community. In other words, the site architecture does not support the same rich community when the mail function is used. Nearly all interview participants had similar complaints about the mail. They simply could not connect to the community in the same way.

Information professionals should recognize that the users’ Imaginary register is not always useful for interpretation; they often describe symptoms of a larger structural anomaly. Elsa says that she uses the site because it supports her knitting. Yet she no longer uses the site to knit; she uses the site to communicate with a larger community about a wide range of social topics. The Symbolic register of the architecture refocused Elsa’s desire to a new goal, which she clearly does not articulate—or perhaps is not aware. To provide more description, the next section describes a second relationship between the Imaginary and the Symbolic.

The Symbolic and the Imaginary Permeate Among Users
Bridget uses Ravelry for several hours each day. She has been using the site several months longer than Elsa, and she had originally sampled similar sites like Craftster (Kramer, 2006).  Bridget also suggested she originally wanted to use Crafster to support her hobby: “I really didn't like the way the site was organized and I found it really hard to maneuver around, especially since it involves all crafts and at the time I was only interested in knitting resources.” She eventually became a member of the more user friendly Ravelry, and like Elsa, her  “relationship with Ravelry is ever evolving.” She classifies her relationship in three chronological periods: “dabbling and discovery, obsession and burnout, and balance and self-discovery.” She moderates several discussion forums; she organizes several related activities related to Ravelry. For example, she organizes gift exchanges amongst the Ravelry communities that she moderates. As she discussed Ravelry, she described the sites technical abilities in greater detail than other interview participants; she knew about functions of which few other interview participants knew.

These statements reveal at least two interesting facts. First, the Symbolic register is modified during continued site use and varies slightly amongst users. The original desire of the site needed to support the Symbolic register of an uninitiated participant. The later use of the site needed to support the Symbolic register of a user who has integrated much more of the site’s grammar and architecture into their experience. The Symbolic varies slightly amongst users because Elsa never reported to have experienced Bridget’s burnout period. The Symbolic varies faintly between people and in time spent on the site. Second, her site expertise, along with a new feature that she called magic linking, suggests a palpable experience of the Symbolic as it slowly incorporates pieces of the Imaginary. At times, Bridget is aware of processes that will blend into the hidden Symbolic register. To exemplify, Bridget describes magic linking as follows.

This allows the poster to link to a person, project, pattern, or yarn, without having to find the url and do any html work. You simply type [name](person or pattern, etc) and it will find the page that you are looking for and link to it. If you link to a person a message goes to their inbox so they know they are being talked about.

Two parts of this description suggest the work of the Imaginary moving through the Symbolic. First, Bridget’s description suggests an importance to recognizing the magic link as a type of URL link. The part that Bridget likes is that it refines her choice and provides ease. Although she has fewer linking choices, she enjoys that her linking process is faster. The Symbolic of the magic link is its rhetorical function: the link becomes a use of prosopopoeia, or personification. This is a clear example of the idea that the architecture functions like a language. The magic link is a rhetorical transformation and a form of community exchange. The transformation invites certain uses of the site while concealing others. Although Bridget still recognizes that the magic link is only one form of link, other interview participants did not. They described this as a normal functioning of the Ravelry technology.

Second, Bridget suggests that the reflexive link to alert another Ravelry member that they have been linked is unavoidable. She knows that she exerts control over the magic link—knowing that if she wanted to reference a different type of resource like an external web site, the magic link is inappropriate. Yet she doesn’t seem to suggest that the link to the other user is somehow not necessary. She recognizes the personification of the magic link, but not that the rhetorical chiasmus has become a pervasive form of community networking. For Bridget, the magic linking function is intimately associated with a natural site function that alerts users they have been magic linked. That is the Symbolic register, a troping that is natural in the unconscious of the individual and the social community. It becomes an unwritten rule of proper social functioning. Those types of concealed tropes help structure and orient the site towards its social functioning. Theoretically, to de-architecture the links that prop up the communal functioning of Ravelry would be to unravel the social structure and information architecture of Ravelry. Member content can be used or removed, to some extent, as long as that action doesn’t affect the overall focus on community, a focus that supplants any knitting desire as secondary to the ability to communication about that knitting content. These descriptions of the rhetorical work in the Symbolic register through the articulation of the Imaginary helps reveal the Real of Ravelry in the following section.

Ravelry and the Real
One part of Ravelry rarely used by any interview participants was a frogging feature. As members upload information about knitting projects to the site, they are able to list them as finished, in-progress, hibernating, or frogged. When Bridget magic links to a user profile, she is providing community access to view any projects that have been marked as frogged. The word frogging is an onomatopoeic play on the phrase “rip it, rip it.” In knitting vernacular, frogging a project means ripping it apart after a mistake has been noticed. Frogging a knitting project is similar to throwing away a painting that has become too muddled to salvage. Ravelry member interviews about the frogging features clarify how the Real register does work in information architecture.

Rhoda indicated that since she joined Ravelry, she completes more of her knitting projects. She said she feels social pressure from her Ravelry community to do so. Further, she suggested that she feels one way to complete an item is to frog the entire project. She feels “guilt when her projects are in progress or hibernating.” For Rhoda, at least, frogging is one form of completion. But the frogging feature mostly supports uploading images of frogged projects. Yet the process of frogging does not lend itself to images simply because a fully frogged project has no materiality, and a partially frogged project is in progress or hibernating. Without content to support the community Symbolic register, the frogging features are little used. During interviews, Rhoda had plenty of stories to tell about frogging, but Ravelry does not support these types of stories well. Ravelry doesn’t support this type of storytelling about frogging. Frogging is a phantom category that is a leftover from knitting in other contexts. This Real, the resistance of the site to certain types of content, is perceivable with the juxtaposition of a secondary vernacular perspective of frogging.

To simplify, for Rhoda completing projects is good. The functions of Ravelry orient completed content for good site use: social interaction. This seems to be an ethical value that was imposed by her Ravelry use. She and other users repeated that they knit for fun as relaxation. That process was relaxing and the outcome unimportant. Yet Rhoda specifically states that she completes more projects today because of guilt. Prior to this, her knitting activities were not activities that could be complete or not complete, they were simply a process. But the Symbolic and Imaginary combine to generate a vestigial Real that recreates knitting with a different ethical value of good. The Real supports and influences the site as the remainder of the Symbolic and Imaginary.

Practicalities of the Symbolic/Imaginary
The benefit of recognizing the Imaginary and Symbolic are practical in at least two ways. First, it suggests a new model of user behavior—one that is much more aligned with the understanding of the organization of information as the unconscious persuader of a savvy user that reflexively interacts with that user. Second, an ethic for designers emerges. Elsa’s interaction with the site afforded her access to a new activity.  Further, via other functions of the site she has since invested her time and effort into that hobby which reinvests itself into a further dependence on the original infrastructure of the site.  The designer is faced with a moral dilemma of deciding whether the access they have provided is a burden they ethically condone. With a site devoted to leisure and craft this seems relatively minor; when large infrastructures are devoted to public dissemination of governmental information the stakes are much higher.

Further, there is a benefit to understanding the language of information architectures. These examples suggest three types of tropes of a list of thousands, many of which are not currently part of popular navigational systems. The information architecture acts like a language. It uses transformative tropes to provide sensibility for its users. Several prevalent tropes like chiasmus, prosopopoeia, and anastrophe were discussed in this paper. Using these tropes as a heuristic opens design possibilities. One example could be dehortatio, a dissuading technique. Imagine linkages to forums that are completely different than the information currently being viewed by a user. The possibilities for new forms of knowing in various contexts are enormous. A dehortatio trope could be used to design interactions that incorporate alternative points of view and inform public policy making. While not all information architectures are identical to Ravelry—or have its same user base—the Post-Lacanian perspective provides an interpretive framework that becomes immensely useful when applied to systems that are primarily organization for human use.

Post-Lacanian Precautions
The material covered here is foundational. The Post-Lacanian tradition has been continually developed over the last sixty years. To attempt a full description in one paper is impossible. And in relation, some of the concepts have been generalized or abridged in order to provide a foundation for further conversation. Indeed, some of the descriptions used here have been used abnormally in order to provide a more helpful overview of the Post-Lacanian position. The previous work may be thought of as a ladder that will eventually need to be discarded (Wittgenstein, 1981, p. 6.54). For example, the Real is one concept that could be more fruitfully developed. It is more complex than the partial concept this paper described.

Concluding Remarks on the Post-Lacanian Information Architecture
The Post-Lacanian perspective foregrounds several research areas relevant to information studies: the reimagining of seminal information studies theory with a unique perspective, the ability to interpret the confluence of users and architectures, and the foregrounding of particular ethical decisions. Further, the perspective espouses a philosophical perspective for communicating to practitioners who will later develop information architectures. The designer helps to fulfill a desire in the user—a key component of the perspective. The users’ desire of the system reshapes that architecture. The designer becomes an employer of tropological Symbolic relationships. The types of organizational tools affect the material practice of practitioners. And because this perspective was shown to incorporate both the user and designer in the design process, it revalues the relationship of designer, user, and system. For example, the designer creates a search box employing an enthymatic metaphor that the user participates in. This participation further restructures the shape and process of the entire site. The algorithm is the trope. Further and in relation, this suggests that there is a historically oriented telos at work within the information architecture. As the user and architecture and designer intermingle with each other, their actions become inscribed into the site, building a sort of technological momentum (Hughes, 1994). This momentum doesn’t build toward certain ends, but it does privilege certain outcomes over others. For example, Bowker & Star demonstrate how users are literally built into the DSM-IV and become prototypic personas for later subjects that interact with that same informational system (Bowker & Star, 1999).

Because the system becomes oriented in certain ways, it provides an entry into the second research area: the ethics of the information architecture itself. The second research area, an ethics of architecture design is highlighted through this paper’s discussion about the Real. Through analysis of the Symbolic, Imaginary, and Real, a vocabulary is provided that provokes questions of how material affordances are assembled within the site, which Symbolic transformations ensue and make possible certain actions and knowledges, and how the Imaginary desire of those involved alter both the Symbolic and the Real.

This perspective is significant for both research and education. In support of research, this analysis benefits information studies theory. Various interdisciplinary discussions have focused on comparing older library technologies to newer technologies. These discussions are often superficial and unhelpful. Analysis of non-LIS systems informs theoretical ideas that may be useful for information studies systems. These discussions do not need to be limited to choosing one technology over another. The Post-Lacanian perspective complicates those either-or questions by suggesting that technologies and people are linguistically bound. Applying this perspective informs analysis of systems of information as they occur in a material environment, which should help inform information studies theory. In support of education, Post-Lacanian analysis provides a heuristic to understand the complex and contingent behavior of users. Although in-depth analysis requires a lengthy theoretical commitment of research into Post-Lacanian research and tropological vocabularies, that knowledge can be made practical and passed on to practitioners at a more rudimentary and useful level. Indeed, it can be extremely practical—issues of evaluation and ethics should be. Education through innovation; innovation through new researches; researches rooted within past perspectives; past perspectives leveraged through Post-Lacanian theory.

References
Anderson, J. R. (1985). Cognitive psychology and its implications (2nd ed.). New York: W.H. Freeman.
Arbogast de Hubert-Miller, B. (2006). The IA of potentiality: Toward a grounded theory of information architecture philosophy, theory and research. Bulletin of the American Society for Information Science, 33(1). Retrieved September 23, 2008, from http://www.asis.org/Bulletin/Aug-06/arbogast_de_hubert-miller.html.
Bates, M. J. (1989). The design of browsing and berrypicking techniques for the online search interface. Online Review, 13(5), 407-424.
Benjamin, W. (1968). The work of art in the age of mechanical reproduction. In H. Zohn (Tran.), Illuminations (pp. 217-252). New York: Schocken.
Bowker, G. C. (2005). Memory practices in the sciences. Cambridge: MIT Press.
Bowker, G. C., & Star, S. L. (1999). Sorting things out: Classification and its consequences. Cambridge: MIT Press.
Burnett, G. (2000). Information exchange in virtual communities: A typology. Information Research, 5(4), 5-4.
Burnett, G., Dickey, M. H., Kazmer, M. M., & Chudoba, K. M. (2003). Inscription and interpretation of text: A cultural hermeneutic examination of virtual community. Information Research, 9(4), 9-1.
Burnett, G. (2002). The scattered members of an invisible republic: Virtual communities and Paul Ricoeur's hermeneutics. The Library Quarterly, 72(2), 155.
Campbell, D. G. (2007). IA Research: The Future State of the Art. Bulletin of the American Society for Information Science, 33(5), 9-10.
Campbell, D. G., & Fast, K. V. (2004). Panizzi, Lubetzky, and Google: How the modern web environment is reinventing the theory of cataloguing. Canadian Journal of Information & Library Sciences, 28(3), 25-38.
Chatman, E. A. (1996). The impoverished life-world of outsiders. Journal of the American Society for Information Science, 47(3), 193-206.
Day, R. E. (2000). The "conduit metaphor"and the nature and politics of information studies. Journal of the American Society for Information Science and Technology, 51, 805-811.
Day, R. E. (2005). Poststructuralism and information studies. Annual Review of Information Science and Technology, 39(1), 575-609.
Ellis, D. (1992). The physical and cognitive paradigms in information retrieval research. Journal of Documentation, 48(1), 45-64.
Fink, B. (1995). The Lacanian subject: Between language and jouissance. Princeton University Press.
Fink, B. (1998). The master signifier and the four discourses. Key Concepts of Lacanian Psychoanalysis, 29–47.
Fink, B. (1999). A clinical introduction to Lacanian psychoanalysis. Harvard University Press.
Forbes, J. M. (2007). Ravelry: about our site. Ravelry. Retrieved September 12, 2008, from http://www.ravelry.com/about.
Frohmann, B. P. (1992a). The power of images: A discourse analysis of the cognitive viewpoint. Journal of Documentation, 48(4), 365-86.
Frohmann, B. P. (1992b). Knowledge and power in library and information science: Toward a discourse analysis of the cognitive viewpoint. Conceptions of Library and Information Science: Historical, Empirical and Theoretical Perspectives, Taylor Graham, London, 135-48.
Golder, S. A., & Huberman, B. A. (2006). Usage patterns of collaborative tagging systems. Journal of Information Science, 32(2), 198-208.
Harman, D. K. (1997). The TREC conferences. Morgan Kaufmann Multimedia Information And Systems Series, 247-256.
Hjorland, B. (2002). Epistemology and the socio-cognitive perspective in information science. Journal of the American Society for Information Science and Technology, 53(4), 257.
Hughes, T. P. (1994). Technological Momentum. In Does Technology Drive History? The Dilemma Technological Determinism (pp. 101-113). Cambridge: MIT Press.
Jenkins, H. (2006). Convergence culture: Where old and new media collide. New York: New York University Press.
Keith, J. (2008, March 10). Building portable social networks. South by Southwest Interactive, Austin, TX. Retrieved October 22, 2008, from http://2008.sxsw.com/interactive/programming/panels_schedule/?action=show&id=IAP060351.
Kittler, F. A. (1999). Gramophone, film, typewriter (G. Wintrop-Young & M. Wutz, Trans.), Writing science. Stanford: Stanford University Press.
Kramer, L. (2006). Craftster. Craftster - Crafting Community & Forums: Crafts, Patterns, Projects & Craft Ideas. Retrieved October 21, 2008, from http://www.craftster.org/.
Lacan, J. (1977). The four fundamental concepts of psycho-analysis, International psycho-analytical library. London: Hogarth Press.
Lacan, J. (1988). The Seminar of Jacques Lacan: Book II: The ego in Freud's theory and in the technique of psychoanalysis 1954–1955 (J. A. Miller, Tran.). Cambridge: Cambridge University Press.
Lacan, J. (2006). Ecrits: The first complete edition in English (B. Fink, Tran.). New York: W.W. Norton & Co.
Leupin, A. (2004). Lacan today: Psychoanalysis, science, religion (p. 150). New York: Other Press.
McKechnie, L. (2000). Ethnographic observation of preschool children. Library and Information Science Research, 22(1), 61-76.
Mika, P. (2005). Flink: Semantic Web technology for the extraction and analysis of social networks. Web Semantics: Science, Services and Agents on the World Wide Web, 3(2-3), 211-223.
Morville, P., & Rosenfeld, L. (2006). Information architecture for the World Wide Web (3rd ed.). O'Reilly.
Mueller, M. (2002). Ruling the Root: Internet Governance and the Taming of Cyberspace Cambridge, Mass: MIT Press.
Olson, H. A. (2002). The power to name: Locating the limits of subject representation in libraries. Dordrecht, The Netherlands: Kluwer Academic.
Over, P. (2001). The TREC interactive track: An annotated bibliography. Information Processing & Management, 37(3), 369-381.
Petrovski, L. (2008, Fall). Ravelry.com. Vogueknitting International, 52-53.
Pettigrew, K. E. (2000). Lay information provision in community settings: How community health nurses disseminate human services information to the elderly. Library Quarterly, 70(1), 47-85.
Robins, D. (2002). Information architecture in library and information science curricula. Bulletin of the American Society for Information Science, 28(2). Retrieved September 23, 2008, from http://www.asis.org/Bulletin/Jan-02/robins.html.
Savolainen, R. (1995). Everyday life information seeking: Approaching information seeking in the context of "way of life." Library and Information Science Research, 17(3), 259-294.
Skelton, R. (1995). Is the unconscious structured like a language? In International Forum of Psychoanalysis (Vol. 4, pp. 168-178). PEP Web.
Smith, M. R. (1994). Technological determinism in American culture. In M. R. Smith & L. Marx (Eds.), Does Technology Drive History? The Dilemma Technological Determinism (pp. 1-35). Cambridge: MIT Press.
Sunstein, C. R. (2007). Republic.com 2.0. Princeton: Princeton University Press.
Wittgenstein, L. (1981). Tractatus logico-philosophicus (C. K. Ogden, Tran.) (Revised Ed.). London: Routledge.
Žižek, S. B. (2005). Interrogating the real. London: Continuum.

Anatoliy Gruzd

Graduate School of Library and Information Science

University of Illinois at Urbana-Champaign

agruzd2@illinois.edu

Abstract

As a way to gain greater insight into the operation of Library and Information Science (LIS) e-learning communities, the presented work applies automated text mining techniques to text-based communication to identify, describe and evaluate underlying social networks within such communities. The main thrust of the study is to find a way to use computers to automatically discover social ties that form between students just from their threaded discussions. Currently, one of the most common but time consuming methods for discovering social ties between people is to ask questions about their perceived social ties via a survey. However, such a survey is difficult to collect due to the high cost associated with data collection and the sensitive nature of the types of questions that must be asked. To overcome these limitations, the paper presents a new, content-based method for automated discovery of social networks from threaded discussions dubbed name networks. When fully developed, name networks can be used as a real time diagnostic tool for educators to evaluate and improve teaching models and to identify students who might need additional help or students who may provide such help to others. 

1 The Use of Social Network Analysis in E-Learning Assessment

Social Network Analysis (SNA) is a common method to study social interactions and collaborative learning of online groups. Some examples of studies that relied on SNA to evaluate individual learning based on the position of individuals in a network and group cohesion based on general properties of a network include Cho et al. (2007), Reyes & Tchounikine (2005) and Willging (2005). From the social network perspective, individual behavior is defined by others. Thus, to understand individual behavior, we need to "describe patterns of relationships between actors, analyze the structure of these patterns, and seek to uncover their effect on individual behavior" (Nurmela et al., 1999, n.p.). In any social network, there are nodes which represent group members, and edges (often referred to as ties) that connect people by means of various types of relations. The strength of the relations is usually conveyed via a weight assigned to each tie.

A traditional way to collect information about social networks within communities is to ask group members themselves via a survey. However, this method is very time consuming and prone to a high rate of non-responses. Dillman (2000) posited two main reasons for the high rate of non-responses: the questions are highly personal, and the procedures for answering the questions are too burdensome. As a result of these inherent flaws with surveys data, many researchers are focusing on cheaper and more objective, automated methods for collecting data on social networks. Some of these automated methods include using movement tracking devices (e.g., Matsuo et al., 2006), log analysis (e.g., Nurmela et al., 1999), and co-citation analysis (e.g., White et al., 2004).

The most common automated method to collect information on social networks in online communities is to gather ‘who replies to whom’ data which counts the number of messages exchanged between individuals from their recorded interactions. Higher number of messages exchanged is usually interpreted as stronger ties between people. This method is often used with email-types data. In online communities that use threaded discussions, researchers usually rely on information in posting headers about the chain of people who had previously posted to the thread (further referred to as reference chain) to gather ‘who replies to whom’ data. For logical and practical reasons, in the past researchers generally assumed that the reference chain may reveal the addressee(s). More specifically, it is usually assumed that a poster is replying to the previous poster in the reference chain. (For the remainder of this discussion, I will refer to any social network that is built using information in the reference chain as a chain network.) Unfortunately, this assumption is not always 100% true in highly argumentative and collaborative communities such as online classes. A previous poster is not always an addressee of the posting and vice versa. A poster may address or reference other posters from earlier in the thread, from another thread, or even from other channels of communication (e.g., emails, chats, face to face meetings, etc). So, while the use of reference chains provides some mechanism to approximate ‘who replies to whom’ data for threaded discussions, such approximation is not very accurate and is likely to cause an undercounting of possible connections. To overcome the inherent flaws associated with gathering ‘who replies to whom’ data from threaded discussions, I propose a hybrid approach called name networks for inferring social networks using both the posting headers and the actual content of postings. The next section will briefly describe the procedure for building name networks.

2 Building Name Networks

In general, to build name networks, the method starts by automatically finding all mentions of personal names in the postings and uses them as nodes in the name networks. Next the method proceeds to discover ties between all the nodes by connecting a poster to all names found in his/her postings. Finally, to disambiguate name aliases, the algorithm adopts a simple but effective approach that relies on associating names in the postings with email addresses in the corresponding posting headers. For more detailed description of the method, see Gruzd & Haythornthwaite (2008b).

Personal names were chosen as the main input into building name networks because they have been shown to be good indicators of social ties. Linguistically speaking, the use of personal names performs two main communicative functions as identified by Leech (1999): (1) addressee-identifying and attention-getting, (2) social bond-maintaining function. The first function is self-explanatory, when calling somebody by his/her name, a person identifies somebody among others to talk to and at the same time tries to get that person’s attention. As for the social bond-maintaining function, its main purpose is to maintain and reinforce social relationships. For example, when someone uses formal names and titles, it might be to indicate subordination in the relationships. While at the opposite end, someone might use informal names or nicknames to show the same social status or emphasize friendship. The social bond-maintaining function of naming is especially important in online groups. Since names are “one of the few textual carriers of identity” in discussions on the web (Doherty, 2004, p. 3), their use is crucial to create and maintain a sense of community (Ubon, 2005) and social presence (Rourke, 2001). As Ubon (2005) put it, by addressing each other by name, participates “build and sustain a sense of belonging and commitment to the community” (p.122). In sum, by focusing on personal names, the “name network” method can quickly identify addressees of each message and thus automatically discover “who talks to whom” in one-to-many types of online communication such as threaded discussions and chats. Furthermore, the social bond-maintaining function of personal names suggests that the discovered ties between people will not just reflect communication patterns, but also likely reflect real social relationships between people.

To evaluate the proposed method of building social networks and identify what exactly will be gained from using this new method, social networks derived using the ‘name network’ method will be compared against those derived from other means, specifically chain (reply-to) networks and students’ self-reported (perceived) social networks. For the purpose of this work, chain networks will be built by connecting a sender to the most recent poster in the thread, while self-reported social networks will be built based on the data collected via an online questionnaire.

3 Data Collection and Self-Reported Networks

The dataset for this study consists of bulletin board postings and students’ responses to an online questionnaire from six different graduate level online classes at the Graduate School of Library and Information Science, University of Illinois at Urbana-Champaign. The data was collected in Spring 2008 as part of a larger study on online learning in collaboration with Caroline Haythornthwaite. Prior to the beginning of the data collection, Institutional Review Board permission was obtained for this work. All students’ names in the study were ammonized to protect their privacy.

Instructors in these classes primarily relied on Moodle (an open source course management system) to make announcements, distribute class materials and facilitate weekly discussions using bulletin boards. Once a week, students met online using LEEP, a home-built online environment. During these live sessions, the instructor delivered the lecture via a live audio feed. During the lecture, students could ask questions or answer instructor’s questions by typing in the chat room. During some live sessions, the instructor divided students into smaller discussion groups (each group would use a separate chat room for their discussions).

Students’ self-reported social networks were collected via an online questionnaire administered at the end of the semester. The first group of questions asked students to indicate the frequency of their associations with each classmate on a scale from 1 to 5 (with [5] indicating a more frequent association) with respect to three different relations: learning something new about the subject matter from another student, working together, and friendship. The second group of questions asked students to nominate 5 to 8 prominent students that best fit the following four criteria: “influential in one’s learning”, “important in promoting discussion”, “help with understanding a topic or assignment” and “made class fun”. The response rate for the questionnaire was 64% (a total of 81 responses). Each question was designed to discover one of the many possible social relations (e.g., learn, work, help, etc) that might exist between the students.

A self-reported network was then built using the following procedure. First, the procedure added a tie between each respondent and his or her nominees. For the questions from the first group, only nominations with an association level of 3 or higher were considered. The next step was to assign weights to each tie. The weights were assigned based on how many times each nominee was selected by the same respondent. To better reflect actual social relationships between students, the procedure removed all “weak” ties with a weight of less than 3. Since the procedure only kept so-called “strong” ties, it is very likely that they will be symmetric. To help restore some ties missing due to the non-respondents, the resulting network was symmetrized.

Open source software called phpESP (http://phpesp.sourceforge.net) was used to conduct the survey. A Social Network Analysis tool called ORA v.1.9.5.2.6 (http://www.casos.cs.cmu.edu/projects/ora) was used for storage and basic manipulations of the network data. Internet Community Text Analyzer (ICTA) (http://textanalytics.net) described in Gruzd & Haythornthwaite (2008a) was used to automatically build name and chain networks.

4 Chain Networks versus Name Networks

First, the analysis began with comparing name and chain networks using QAP correlations (Krackhardt, 1987). This was done to determine the level of overlap between these two types of networks. QAP correlation relies on Pearson’s correlation coefficient to compare relational data. It was chosen as the method of measurement for this work because “it presumes neither random sampling of cases from a population […] nor independence of observations” (White et al., 2004, p. 116).

Software called ORA was used to compute the QAP correlations. The results of the comparison are presented in Table 1 below. All tests were significant (p<=0.05). In all classes, pairs of name and chain networks demonstrated moderate correlations between 0.45 and 0.69 (See the “QAP” column in Table 1). As expected, there is some overlap between posting behavior as represented by the chain networks and “naming” behavior as represented by the name networks. However, there are also substantial differences in what is revealed by each of these networks. To better understand these differences and assess the accuracy of chain networks, the next section will compare all connections that make up each tie from the name network with those from the chain networks. More specifically, the next step in the analysis will determine how many connections discovered by the ‘name network’ method were not discovered by the ‘chain network’ method.

Chain networks are built based on the information in the reference chain; as a result, they will fail to connect a poster to poster’s addressee whose email is not yet in the reference chain. This situation can arise in one of two ways: (1) when it is a first posting of a new thread or (2) when an addressee has not posted anything to an existing thread. Since all of the names extracted for building name networks were manually inspected for accuracy in the study, it is fair to use these names as actual addressees of postings or people who are somehow connected to the poster. Using an automated script, I counted the number of instances for each of the two situations described above. The counts revealed some pleasantly unexpected results (See Table 2). On average, a chain network misses about 33% of the potentially important connections as compared to the name network. Of the 33% missed connections, 23% came from postings that were the thread starters (Column A) and about 10% came from subsequent messages in a particular thread (Column B). Additionally, there were another 7% of missed connections that occurred when an actual addressee or a ‘reference’ person was the author of a previous posting in the thread, but not the most recent one (Column D).

To determine the exact nature of connections that were missed by the chain networks, I analyzed all postings that correspond to columns A and B in Table 2 for all six classes.

Situation 1: First Posting of a Thread. The semi-automated content analysis of postings using ICTA revealed that among the most commonly used names in the first posting of a new thread was the instructor’s name. Specifically, instructor’s name was used to

§  Ask the instructor about something (e.g., “[Instructor’s name] if you see this posting would you please clarify for us”),

§  Ask peers to clarify something that the instructor said during the lectures (e.g., “I remember [Instructor’s name] asking us to email her with topics [...] I wonder if that is in replacement of our bb question?”), or

§  Share information with classmates obtained from the instructor via some other personal communication such as email. (e.g., “I just got a reply from [Instructor’s name], and she said that […]”)

This type of postings and the ties derived from them is very important in the context of learning. This is because “student-instructor” ties derived from these messages can be used to identify students who are repeatedly asking for instructor’s help. For example, a high weight for a tie between a student and the instructor may suggest that a student is uncertain about something in the class and might need extra attention from the instructor. However, if many students are connected to the instructor via these types of messages, then it may indicate that lectures or other class materials are unclear to not just one student and thus either the materials or a delivery method might need to be reconsidered by the instructor.

Another common category of messages was when an instructor mentioned a student. These were usually announcements from the instructor containing names of students responsible for leading a class discussion. For example, “Dan, [...] Since you have studied [Topic], would you get our discussion going on the forum for this week”. Sometimes an instructor would also mention a student praising him or her for some good work in the class. This suggests that if there is a tie from an instructor to a student based on this kind of postings, then it is very likely that this student is doing well in the class. Identifying reliable and successful students in a class is an important task for any instructor or school’s administration, especially when formal grading information is unavailable. For example, an instructor can use such information to assign students into more effective groups in which at least one of the students is doing above average in the class.

Another common type of messages in this category was when an instructor listed groups with their individual members for smaller group discussions. After examining these postings, I concluded that the ties derived from them do not necessarily reflect relationships between the instructor and a student. Instead, these postings can be used to automatically identify students who were assigned to work together, thus potentially creating “work” ties. “Work” ties are especially important for studying online groups since they are often precursors of even closer ties between online participates (See, for example, Haythornthwaite, 2002). This was confirmed by several students in the comment section of the online survey. They viewed the break down into smaller groups during live sessions as a good way to get to know their peers.

The last category of messages was when a student mentioned other student(s). In these cases, the poster often took a leadership role in a group, for example, by summarizing other group members’ postings or assigning roles for a project as demonstrated in the following excerpt: 

“Some quick poking around shows that Steve and myself are here in Champaign, [...] and Nicole is in Chicago. [...] does anyone have a strong desire to be our contact person to the administrators”

This type of messages is useful in identifying active group members and group leaders and would be very useful when studying collaborative learning. However, a lot of messages like this from the same person may be perceived negatively by other group members. For example, in a related study, when analyzing a large collection of Usenet newsgroup messages, Fiore et al. (2002) found that online participates who dominated the conversations were often viewed unfavorably. Nevertheless, a more detailed analysis is needed to study the influence of this type of messages/connections in the online learning environment.

Situation 2: Subsequent Posting in a Thread. The detail examination of this type of messages revealed three main types of references/connections:

• A reference to an event or interaction that happened outside the bulleting board (e.g., “Dan and I have been corresponding via e-mail and he reminded me that we should be having discussion here"). This type of messages is likely to connect people who work together.  It is also suggestive of stronger personal ties. This is because according to the idea of media multiplexity, stronger ties tend to communicate via more communication channels (See, for example, Haythornthwaite & Wellman, 1998; Haythornthwaite, 2001).
• A reference to someone as part of a group when providing a feedback to the whole group or posting on behalf of the whole group and signing the names of all group members (e.g., “Angela and Natasha, I couldn't wait to see your site. I knew it was going to [be] awesome!”). This is another type of messages that will likely indicate “work”-related ties.
• A reference to somebody who presented or posted something awhile ago or via different communication channel (e.g., “[…] it made me think of the faceted catalogs' display that Susan posted”). These postings are likely to identify “learning” ties. This is because they show that a poster was not just commenting on the previous post, but rather on something that was said awhile ago. This means that the poster was following the class discussion. And a student mentioned in the posting made some significant contribution to the discussion that resonated with the current poster. All these activities can be categorized as evidence of learning.

5 Chain networks and Name networks versus Self-Reported

The final part of the study was to compare chain and name networks with self-reported networks and to determine which of the two networks is a better approximation of self-reported (perceived) social networks (if any). Traditionally it is presumed that observed social networks such as chain networks can more accurately reflect actual relations between group members as opposed to their individually perceived perspectives and thus provide a better representation of what is really going on in an online community. But for online learning environments, due to the individualized nature of learning itself, it may be more important to identify and understand perceived social networks in the context of studying collaborative learning. This is because what is deemed as important or relevant to one student may only be marginally valued by another student. Until now, the only reliable way to collect perceived data has been through resource-demanding surveys. Therefore, it would be a methodological breakthrough if an automated method for mimicking perceived social networks is devised.

For this analysis, I conducted a pair wise comparison of the three types of networks using statistical network models and specifically Exponential Random Graph models or just p* models (Robins, In press). To build p* models, I used XPNET software (Wang et al., 2006). There are a few important reasons why p* models were selected to conduct this comparison and not other statistical models or QAP correlations. First, since some students did not participate in the survey, some possible ties were probably missing in the self-reported networks. As a result, QAP correlations would likely produce inadequately lower results. Second, parameters estimated by p* model is easy to interpret and compare across different pairs of networks. Finally, p* model is the only statistical model that is capable of modeling different network structures as well as individual characteristics of the group members (Snijders, 2008).

Using p* models, for each class I estimated the parameter EdgeAB for a pair of the chain network and self-reported network first and then for a pair of the name network and self-reported network. The parameter EdgeAB indicates the likelihood of two networks sharing ties not by a chance alone. The results are shown in Table 3. The model was converged (t–statistics<0.1 for all estimated parameters) and the model was found to be significant (the goodness of fit for EdgeAB was less than 0.1 and between 1 and 3 for all other parameters) for all classes, except the case of a pair of the name and self-reported networks for Class6.

The results show that for four out of six classes, the name networks are consistently more likely to share ties with the self-reported networks than the chain networks (more than just by chance alone). This supports my general expectation that the name networks are more reflective of students’ perceived relationships. However, for two smaller classes, Class5 and Class6, the name networks were less likely to match the self-reported networks. (For Class6, the model was not significant.) This was a very puzzling but intriguing result. It led to a separate investigation that is currently on the way. Below are some preliminary results that suggest some concrete steps on how to further improve the “name network” method in the future to better reflect self-reported (perceived) social networks. 

To find out why the name networks for Class5 and Class6 were less likely to share ties with the self-reported networks than the chain networks, I decided to analyze the network signatures for each student in both classes to discover specific differences between these two types of networks. In the current paper, only preliminary results for two students, Nick and Anna from Class5 are reported below. These two students were selected because their network signatures were the most different in each of the two types of networks. One student, Nick, had several ties in the self-reported network that were missing in the name network. The second student, Anna, had a couple of ties in the name network that were missing in the self-reported network. For these two students, I examined all of their ties that exist in the self-reported network but not in the name network and vice versa. One of the main goals of this analysis is to identify what caused the “name network” method to miss some self-reported ties and to include some ties that are not in the self-reported network. Furthermore, the analysis will help to identify any additional clues from the content of postings that can be used to improve the “name network” extractor. For this examination, I used ICTA.

5.1 Why did the Class5 name network miss some self-reported ties?

A student named Nick from Class5 was selected by seven other students in the self-reported survey, but strangely in the name network, Nick was not connected to any of these seven individuals. After a brief investigation, it was determined that Nick only posted three messages to the bulletin board for the whole semester. There was simply not enough evidence on the bulletin board for the name network to discover ties to other individuals. So, on the surface, it is not clear what was the basis for these 7 nominations from his fellow students. A posting from the instructor can shed some light on this mystery. The instructor mentioned Nick on the bulletin board once, when assigning students into smaller discussion groups for the chat sessions. It turned out that the other two students who were assigned to work with Nick were among those who nominated Nick in the survey. This suggested an important future improvement to the ‘name network’ method. In addition, to connecting a poster with all people who are mentioned in the body of his or her posting, the ‘name network’ method should also connect any people whose names co-occur in close proximity in the same messages. With such a modification, Nick would gain two more additional ties in the name network to the two students who nominated him in the survey. As a proof of the concept, I re-built the name network for this class using co-occurrence of names in the text as an additional indicator of personal ties and re-run the comparison analysis between the name network and self-reported network for Class5. This time the likelihood of sharing ties between these two networks increased to 1.50 (t-statistics=0.067) which is higher than the corresponding value from the chain network.

5.2 Why did the Class5 name network include some ties that were not in the self-reported network?

Anna is a well connected student in the self-reported network. However, she only had three strong ties in the name networks. For the purposes of this section, I will only focus on two of the three ties from the name network that are missing in the self-reported network. (The third tie was reported in the self reported network and thus is not relevant to this part of the discussion.) The two ties in questions are with fellow students Rick and Mark.

The tie between Anna and Rick resulted from Rick posting three different messages to Anna thanking her for “insights”, “thoughtful comments”, and “all the wonderful posts and information”. However, surprisingly there was no tie between these two students in the self-reported network. After a detail investigation, it turned out that Rick did select Anna in the survey as a person who influenced his learning and helped the most in the class. (Rick was not nominated by Anna.) But because all ties with a weight less than 3 were removed (See Section 3), a tie of 2 between Anna and Rick also disappeared. As an experiment, I built a “learning” network based on the students’ responses to only one of the question in the survey about “learning”. In this learning network, there was a tie between Anna and Rick. Next I compared this “learning” network with the original name network (without using co-occurrences). The resulting likelihood has slightly increased from 0.96 to 1.17 (t-statistics = -0.062). This suggests that the name network was a bit more similar to the “learning” network than to the overall self-reported network for this particular class. Therefore, the continuation of this study will be to compare name networks with each type of self-reported networks to determine if name networks are better in predicting “learning” ties than others. However, it is possible that for some other class, depending on the prevalence of one type of interactions over the other, a name network can better reflect other types of self-reported networks such as “friendship” or “work” networks. Therefore, as a future improvement, the “name network” method should be able to not just discover ties but also categorize them into different relations. This can be done by using information about roles of participants (e.g., student, guest speaker, instructor, etc), a position of a message in the thread as suggested in Section 4, and/or the context words where particular names are mentioned in a posting. For example, words like “thank you”, “help”, “assistance” may indicate that a student helping another student, thus they are connected via the “help” relation. With such an algorithm in hand, it will be possible to build name networks that reflect only “help” relations, only “learning” relations, only friendship or some other relation that is important to members of a certain online community.

The tie between Anna and Mark resulted from Mark posting two messages with Anna’s name in them. The first posting from Mark was a question directed at Anna, “Anna -- what did you mean by [word] in paragraph 3 of your reply?” The second message was a thank you message from Mark to Ann for posting an interesting article to the bulletin board. (There were no messages from Ann mentioning Mark’s name.) But regardless, this may be enough to suggest a tie between Mark and Anna. Unfortunately, because Mark did not participate in the survey, the self-reported network did not include a tie between them. In such case, a researcher can rely on tools like ICTA to conduct a semi-automated content analysis of messages to make the final decision about the accuracy of the “name network” method.

6 Conclusions

The “name network” method as proposed and evaluated in this work provides one more option for understanding and extracting social networks from online discussion boards. Section 4 demonstrated that name networks provide on average 33-40% more information about social ties in a group as compared to chain networks. This additional information is available because name networks can account for instances when a poster addresses or references somebody who has not previously posted to a particular thread. Furthermore, the results of the study demonstrated that name networks are also very adept at detecting social relations such as work and help which are considered by many researchers to be crucial in shared knowledge construction and community building. Based on the discussion in Section 5, there is an overall tendency of name networks to better reflect self-reported ties than chain networks. These characteristics make name networks a useful real time diagnostic tool for educators to evaluate and improve teaching models, and to identify students who might need additional help or students who may provide such help to others. 

Finally, the semi-automated content analysis of postings from Class5 and Class6 classes using ICTA (http://textanalytics.net) suggested two important improvements to the “name network” method to increase the accuracy of tie discovery. The suggested improvements include (1) using names co-occurrence as an indicator of the “work” relation and (2) identifying types of different relations based on the context words used in the postings.

References

Bernard, H.R., P. Killworth, L. Sailer. 1981. Summary of Research on Informant Accuracy in Network Data and the Reverse Small World Problem. Connections 4(2): 11-25.

Cho, H., Gay, G., Davidson, B. and Ingraffea, A. (2007). Social Networks, Communication Styles, and Learning Performance in a Cscl Community. Computers & Education 49(2): 309-329.

Dillman, D.A. (2000). Mail and Internet Surveys: The Tailored Design Method. New York, Wiley.

Doherty, C. (2004). Naming the Trouble with Default Settings. In the Proceedings of “SFL Ripples in the 21st Century” Australian Systemic Functional Linguistics Association Conference, Brisbane, Australia.

Fiore, A.T., Tiernan, S.L. and Smith, M.A. (2002). Observed Behavior and Perceived Value of Authors in Usenet Newsgroups: Bridging the Gap. In the Proceedings of the SIGCHI conference on Human Factors in Computing Systems: Changing our world, changing ourselves, Minneapolis, Minnesota, USA, ACM.

Gruzd, A. and Haythornthwaite, C. (2008a). The Analysis of Online Communities Using Interactive Content-Based Social Networks (Extended Abstract). In the Proceedings of the ASIS&T 2008 conference, pp. 523-527.

Gruzd, A. and Haythornthwaite, C. (2008b). Automated Discovery and Analysis of Social Networks from Threaded Discussions. Paper presented at the International Network of Social Network Analysts.

Haythornthwaite, C. (2002). Strong, Weak, and Latent Ties and the Impact of New Media. The Information Society 18(5): 385-401.

Haythornthwaite, C. (2001). Exploring multiplexity: Social network structures in a computer-supported distance learning class. The Information Society 17(3): 211-226.

Haythornthwaite, C. & Wellman, B. (1998). Work, friendship and media use for information exchange in a networked organization. Journal of the American Society for Information Science 49(12): 1101-1114.

Leech, G. (1999). The Distribution and Function of Vocatives in American and British English Conversation. In H. Hasselggård and S. Oksefjell (Eds.), Out of Corpora: Studies in Honour of Stig Johansson, Rodopi.

Matsuo, Y., Hamasaki, M., Nakamura, Y., Nishimura, T., Hasida, K., Takeda, H., Mori, J., Bollegala, D. and Ishizuka, M. (2006). Spinning Multiple Social Networks for Semantic Web. In the Proceedings of the Twenty-First AAAI Conference on Artificial Intelligence, Boston, Massachusetts.

Nurmela, K., Lehtinen, E. and Palonen, T. (1999). Evaluating Cscl Log Files by Social Network Analysis. In the Proceedings of the 1999 Conference on Computer Support for Collaborative Learning, Palo Alto, California, International Society of the Learning Sciences.

Reyes, P. and Tchounikine, P. (2005). Mining Learning Groups' Activities in Forum-Type Tools. In the Proceedings of the 2005 Conference on Computer Support for Collaborative Learning: Learning 2005: the Next 10 years!, Taipei, Taiwan, International Society of the Learning Sciences.

Robins, G. (In press). Exponential Random Graph (P*) Models for Social Networks. In R. Myers (Ed.), Encyclopaedia of Complexity and System Science, Springer.

Rourke, L., Anderson, T., Garrison, D. R., & Archer, W. (2001). Methodological Issues in the Content Analysis of Computer Conference Transcripts. International Journal of Artificial Intelligence in Education 12: 8-22.

Snijders, T.A.B. (2008). Models for Social Networks. Course Lecture, University of Oxford.

Ubon, A.N. (2005). Social Presence in Asynchronous Text-Based Online Learning Communities: A Longitudinal Case Study Using Content Analysis. Department of Computer Science, The University of York. Doctor of Philosophy.

Wang, P., Robins, G. and Pattison, P. (2006). PNet: Program for the Estimation and Simulation of P* Exponential Random Graph Models, User Manual. Department of Psychology, University of Melbourne.

White, H.D., Wellman, B. and Nazer, N. (2004). Does Citation Reflect Social Structure?: Longitudinal Evidence from the Globenet Interdisciplinary Research Group. Journal of the American Society for Information Science and Technology 55(2): 111-126.

Willging, P.A. (2005). Using Social Network Analysis Techniques to Examine Online Interactions. US-China Education Review 2(9).

Over the past several decades, the LIS community has adopted theories, methodologies, philosophical bases and assumptions from other academic disciplines to solidify its domain. Many of the ideas from cognitive science, psychology, management science, systems science, communication science, organizational science, and computer science have positively contributed to the emergence of new research areas in library and information science. This has resulted in a broadening of the library and information science curriculum. Recently, museum studies has been tendered more attention. Several archives programs in LIS (e.g., Pittsburgh’s museum archives, FSU’s museum informatics, etc.) try to incorporate the subjects (or contents) of museum studies. Conferences called “Museums and the Web” and “International cultural heritage informatics meetings” are attended every year by myriad museum studies researchers and LIS participants. This synthesis of studies has produced a new domain called “museum informatics.” It is an emerging, interdisciplinary field of study, which studies the sociotechnical issues that arise when people, information, and technology interact in museums (Marty, Rayward, & Twidale, 2003).

This is the challenge: How do we promote cooperative relationship between LIS and museum studies? This might be answered by recognizing similarities and differences between LIS and museum studies’ curriculum. So, this study analyzed the curriculum offered by museology or museum studies degree programs in the United States to ascertain the nature of museum studies education. In addition, it analyzed the museum informatics curricular practice of LIS graduate program in the United States to identify the current state of museum education in LIS schools.

The study began by identifying museum studies programs that offer undergraduate and graduate programs. Schools that offer a graduate certificate program are also included in this study, but they will be analyzed separately and later. However, training programs provided by a museum institution or society were excluded in this study. Programs were identified from:  U. S. News, Worlds Report and Smithsonian’s Museum Studies Training Program Directory (http://museumstudies.si.edu/resources.html), Committee on Museum Professional Training’s member program (http://www.comptaam. org/aboutcompt/memberprograms.html). First, the program descriptions in each program were viewed on the open web to characterize the museum studies education: the number of credit-hours, the number of required courses, the area of interests, and other requirements were collected. In addition, course titles and short catalog descriptions were collected to identify curricular trend and content.

Then this study examined museum informatics courses in Library and Information Science programs accredited by American Library Association. Course titles, course descriptions, and syllabi were collected. The courses on the topic of museum studies were identified based on their titles and descriptions where museum was mentioned as a term (e.g., museum informatics, museum archives, etc.).

The results of this study will be expected to provide implications for museum education in LIS schools. This will inform how best to design interdisciplinary and multidisciplinary curricula approach in LIS to support the education of museum professionals.

References

Marty, P.F., Rayward, W.B., & Twidale, M.B. (2003). Museum Informatics. Annual Review of Information Science and Technology, 37, 259-294.

Abstract

While the term “communities of practice” (CoP) dates to the 1990s (Lave & Wenger, 1991; Wenger, 1998), the phenomenon it describes is the timeless work of learning through mentoring and apprenticeships. Contrast this with our typical methods of professional development today -formalized courses or one-time workshops, often conducted as lectures that focus on individual learning and answers rather than inquiry. Theories of constructivist and social learning and an awareness of multiple intelligences and diverse learning styles have positively influenced how we teach, but much work still needs to be done to improve teaching and learning, particularly in online and blended (combined face-to-face and online) environments. The CoP framework provides an integrated model and guide for social learning that is applicable to a broad range of contexts -from classrooms and professional communities to project management and organizational development.

As formal education and library services increasingly move online, the integration of library services into the curriculum using information and communication technologies has moved beyond information literacy across the K-12 curriculum to campus-wide, web- based information literacy initiatives in higher education. The term “blended librarian” dates to 2004 when Stephen J. Bell and John Shank proposed this definition: “the blended librarian [is] an academic librarian who combines the traditional skill set of librarianship with the information technologist's hardware/software skills, and the instructional or educational designer's ability to apply technology appropriately in the teaching-learning process” (2004, p. 374).

This paper summarizes the theoretical and research foundations in inquiry learning, project-based learning, computer-mediated communication, blended learning, and communities of practice that support a CoP-based approach to blended librarianship (Bell & Shank, 2007). A CoP model that incorporates a creative tension between face-to-face and online communication (Barab, MaKinster, & Scheckler, 2004) is presented. This is followed by a description of the application of the model, with the use of social software and low threshold applications, to two courses (Reference & Online Services, and

Internet Fundamentals and Design) taught by the author in an accredited MLIS program. The main advantages are: (1) The CoP delineation of social learning processes provides coherence for project-based learning approaches that build on the interdisciplinary expertise of students and also provides mechanisms for sharing that expertise and building collective knowledge on an ongoing basis. (2) The dualities, or creative tensions, posited by the CoP framework provide a means to clarify the choices faced by instructors in curriculum design. (3) The use of a CoP-based approach that incorporates social software and low threshold applications in LIS curricula can serve as a model for blended librarianship for future librarians. Student evaluations of these courses were highly positive. Finally, the disadvantages of a CoP-based approach to blended learning are discussed, with recommendations for further research and application. Teaching, learning, and librarianship, particularly reference services, can succeed or fail based on the quality of the human interactions. The CoP-based approach can positively influence human interactions in blended learning and librarianship contexts. uthor in an accredited MLIS program. The main advantages are: (1) The CoP delineation of social learning processes provides coherence for project-based learning approaches that build on the interdisciplinary expertise of students and also provides mechanisms for sharing that expertise and building collective knowledge on an ongoing basis. (2) The dualities, or creative tensions, posited by the CoP framework provide a means to clarify the choices faced by instructors in curriculum design. (3) The use of a CoP-based approach that incorporates social software and low threshold applications in LIS curricula can serve as a model for blended librarianship for future librarians. Student evaluations of these courses were highly positive. Finally, the disadvantages of a CoP-based approach to blended learning are discussed, with recommendations for further research and application. Teaching, learning, and librarianship, particularly reference services, can succeed or fail based on the quality of the human interactions. The CoP-based approach can positively influence human interactions in blended learning and librarianship contexts.

References

Barab, S.A., MaKinster, J.G., & Scheckler, R. (2004). Designing system dualities: Characterizing an online professional development community. In S.A. Barab, R. Kling, & J.A. Gray (Eds.), Designing virtual communities in the service of learning (pp. 53-90). Cambridge: Cambridge University Press.

Bell, S.J., & Shank, J.D. (2004). The blended librarian: A blueprint for redefining the teaching and learning role of academic librarians. College and Research Libraries News, 65(7), 372-375.

Bell, S.J., & Shank, J.D. (2007). Academic librarianship by design: A blended librarian's guide to the tools and techniques. Chicago: ALA.

Lave, J., & Wenger, E. (1991). Situated learning: Legitimate peripheral participation. Cambridge, MA: Cambridge University Press.

Wenger, E. (1998). Communities of practice: Learning, meaning, and identity. Cambridge, MA: Cambridge University Press.
“We Create”: Blended Learning in LIS Courses
Using the Communities of Practice Framework
Joyce Yukawa
Assistant Professor, MLIS Program, College of St. Catherine
Email: jyukawa@stkate.edu

Introduction

While activities combining face-to-face and online activities in libraries predates the use of the term, “blended librarian” dates to 2004 when Stephen Bell and John Shank proposed this definition: “the blended librarian [is] an academic librarian who combines the traditional skill set of librarianship with the information technologist’s hardware/software skills, and the instructional or educational designer’s ability to apply technology appropriately in the teaching-learning process” (2004, p. 374).
Librarianship, information technology skills, and instructional design using appropriate technology for learning are the three pillars of expertise needed for effective information literacy instruction (Bell & Shank, 2007). How can LIS curricula provide students with opportunities to develop this diverse expertise? A potentially valuable approach is blended learning in a classroom community of practice (CoP). The CoP framework provides an integrated model for inquiry learning and social learning within the context of professional principles, practices, values, and leadership. Using the framework in a blended learning context presents opportunities and challenges, due to the affordances of online communication. The paper presents a CoP framework for blended learning, describes its implementation in LIS courses, and evaluates the advantages and disadvantages of its use.

Challenges of Blended Learning

Blended learning brings benefits and challenges on many levels. A frequently mentioned benefit of online learning is flexibility in time and learning environments (Vaughan, 2007). Multimedia environments can provide visualizations that simulate the face-to-face environment and provide engaging stimuli for learner involvement. Interactive tutorials scaffold instruction with timely feedback that encourages active involvement and individual control over the pace of learning, appropriate for problem solving of well-structured tasks.
The challenge, however, is the appropriate use of technology. Larreamendy-Joerns and Leinhardt (2006, p. 586-90) note that multimedia experiences are often isolated from other important learning factors and that online tutorials are generally not available for domains that require deep conceptual understanding. They argue that a view of learning that proceeds incrementally from simple to complex cannot approximate the multilayered, creative responses provided by expert teachers, leaving students with a superficial understanding of complex subjects.
Moreover, social constructivist approaches such as communities of practice stress that learning is not merely knowledge acquisition but more fundamentally a process of identity formation and empowerment through participation in learning communities. New concepts, models, and theories are constructed from a base of previous experiences and worldviews. Styles, discourses, and patterns of practice emerge through mutual participation. Dialogue and social interaction are “essential aspects of knowing a domain” (Larreamendy-Joerns & Leinhardt, 2006, p. 590).
While social software provides tools for intellectual and social discourse that support cognitive development, social networking, and community building (Kim, 2000; Paloff & Pratt, 1999; Preece, 2000), the primary means of online communication is typed text, which lacks the expressive features of nonverbal communication and voice tone, as well as a shared awareness of context. Thus, some studies stress the importance of face-to-face encounters to promote bonding as a basis for continued interaction and participation online (Haythornthwaite, Kazmer, Robins, & Shoemaker, 2000).

Constructivist Learning and Adult Learners

Departing from the early behaviorist orientation of instruction, contemporary learning theories lean heavily toward constructivist and social constructivist ideas (Salomon & Perkins, 1998). Constructivism in education has its roots in John Dewey’s (1916) ideas of progressive education based on learning through inquiry and experience, combined with insights on cognitive development put forth by Piaget (1932), Vygotsky (1962; 1978), Bruner (1990), and others. Learners interact with the physical and social world to actively create rather than passively receive knowledge. Some see dialogue and guided conversation as the primary means of inquiry learning (e.g., Edwards, 1991; Mercer, 1995; Wells, 1999).
In LIS education where many students are returning to school to pursue second careers, theories of adult learning provide additional guidance. Mezirow’s (1978, 2000) study of older women returning to community colleges to continue their education found that adults learn by negotiating contested meanings, using contextual understanding, and critically reflecting on assumptions to reinterpret the meaning of their experiences as a guide for future action (2000, p. 5). Adult learning is active, problem-based, dialogic, and focused on critical reflection for more effective action. Such approaches as problem-based learning, project-based learning, and case-based learning emphasize solving authentic problems to develop professional expertise (Prince & Felder, 2006; Larreamendy-Joerns & Leinhardt, 2006).
Two community-based approaches have received popular and research attention in the last decade: communities of inquiry (COI) and communities of practice (CoP). The COI model focuses on using text-based computer-mediated communication for computer conferencing in higher education. It posits that cognitive, social, and teaching presence are essential to the online learning experience (Garrison, Anderson & Archer, 2000). Research on COI has focused on the effects of writing on critical thinking (e.g., Rourke, Anderson, Garrison & Archer, 1999, 2001) and has recently addressed issues in blended learning (Garrison & Vaughan, 2007).
Lave and Wenger (1991) introduced the term, “communities of practice.” Using different apprenticeship situations, they describe how learning begins with observing experts, being mentored, and other forms of legitimate peripheral participation before an apprentice becomes a full-fledged member of the community. The work on CoP is broader in scope, audience, and influence than COI, particularly in professional education such as the health sciences and teacher education. The CoP framework is richly complex but has been criticized for its lack of rigor as a testable theory (Storberg-Walker, 2008).
These theories of constructivist and social constructivist learning have contributed much to our understanding of how adults learn, but LIS education lacks a holistic, integrated professional development model for blended environments. The CoP framework has the potential to fill this gap for LIS curricula aimed at educating practitioners.

Communities of Practice

Communities of practice differ from other types of communities because they involve a domain and a practice as well as a community (Wenger, 1998). According to Snyder, Wenger, and de Sousa Briggs (2004):

•    “Community” includes its member relationships and the nature of their interactions—levels of trust, belonging, and reciprocity.
•    “Domain” refers to the community’s focal issues and the sense of members’ identity with the topic.
•    “Practice” consists of a repertoire of tools, methods, and skills, as well as members’ learning and innovation activities.

CoP learning is based on mutuality and shared learning through the dynamics of dualities (Wenger, 1998). A duality is a pair of elements that is always present in different forms and degrees, not a spectrum that indicates movement from one pole to another. A duality can be characterized as a creative tension. Wenger (1998, p. 231) identifies four dualities that address “the fundamental issues of meaning, time, space, and power” and present their own resources and constraints.
The duality of participation and reification signifies the interaction between the social production of meaning and the concrete forms (termed “reifications”) that reflect that meaning. Participation means action, involvement, and connection. It results in reification – an outcome such as an idea, an artifact, increased knowledge or competence. Reification provides a means to share practice while also, to a degree, restraining it. Wenger (1998, p. 52) states that the interaction between participation and reification is essentially the negotiation of meaning.
The duality of the designed and the emergent signifies the interaction between designed structures and those that emerge through practice over time. Wenger’s basic premise is that community structures arise organically through practice. Communities may also intentionally design structures to direct their practices, but these designed structures are also subject to change. Thus, practice is not the result of design but a response to it. I characterize this duality as the negotiation of models of practice.
The duality of the local and the global refers to how CoPs relate to other CoPs through crossing boundaries and bridging differences. The differences are bridged through: (1) the use of boundary objects, such as artifacts, standards, or codes of ethics; (2) brokering to make connections, enable coordination, and facilitate meaning making; and (3) boundary encounters, opportunities for dialogue that allow individuals to “deepen their relationships and discover … collective ways of thinking, approaching a problem, and developing a solution” (Wenger, McDermott, & Snyder, 2002, p. 84). I characterize the interaction between the local and the global spaces of knowledge and practice as the negotiation of expertise.
The duality of identification and negotiability focuses on how individuals form identities through participation in community practices, multiple memberships in local and global communities, their learning trajectories within a CoP, and modes of belonging. An individual takes ownership of collective meaning and shapes that meaning through negotiation. Negotiability is the use of power, not driven by conflict or domination, but “primarily as the ability to act in line with the enterprises we pursue” (Wenger, 1998, p. 189). I characterize this duality as the negotiation of identity and leadership within the community.
Negotiability is made possible through three modes of belonging: engagement, imagination, and alignment. Engagement is active involvement in meaning making, based on mutuality, competence, and continuity (Wenger, 1998, p. 237-8). Through engagement, community members participate in and shape practice, contribute to the community’s shared history, and gain competence and respect.
Imagination is seeing connections through time and space by extrapolating from experience through orientation, reflection, and exploration (Wenger, 1998, p. 238). Imagination involves new insights, awareness of broader systems, standing in others’ shoes, creating models and visions, and reinterpreting histories.
Alignment is coordinating activities and resources to fit within broader structures and achieve results, involving convergence, coordination, and jurisdiction (Wenger, 1998, p. 238-9). Alignment involves finding common ground, defining visions, establishing procedures and structures, and exercising power to focus efforts.    Wenger (1998, p. 240) proposes a design framework for CoP learning that incorporates the four dualities and three modes of belonging (see Table 1).

Source: Wenger, 1998, p. 240, 263-270. Wenger’s descriptions of the modes of belonging for each duality are quoted from Figure 10.3, “Articulating components and dimensions,” with my clarifications derived from other parts of his text in brackets.

Face-to-Face and Online Communication

Barab, MaKinster, and Scheckler (2004) propose a fifth duality: face-to-face and online communication. Herbert Clark's (Clark & Brennan, 1991) theory of common ground provides a basis for understanding the tensions of this duality through the inhibiting and enabling effects of different communication media. The theory posits that common ground must be established for effective communication to occur, and that this “grounding” is based on least collaborative effort. Clark identifies eight “constraints” (i.e., enabling factors) that affect communication efficiency: copresence, visibility, audibility, cotemporality, simultaneity, sequentiality, reviewability, and revisability (see Table 1). Face-to-face communication allows the greatest communication efficiency because it provides the richest array of enablers.

Establishing common ground entails costs, including: (1) effort needed to formulate utterances, (2) effort needed to receive and process communication received, (3) effort needed to understand communication, (4) effort needed to start up new communication, (5) adverse effects of delaying one's communication, (6) effort needed for turn taking, (7) effort needed to gesture and indicate, and (8) adverse effects of committing a communication mistake. The type of medium influences the techniques used to compensate for these costs. Four media that I use in my blended courses offer different configurations of enablers and costs (see Table 3).

Some observers believe that text-based online communication may be superior to face-to-face for supporting reflection because it initially requires more care taken in writing (higher formulation costs) and has reviewability (e.g., Garrison, Anderson & Archer, 2000; Larreamendy-Joerns & Leinhardt, 2006; Yukawa, 2005). The influence of writing on critical thinking has also been examined (Applebee, 1984; Ong, 1988). However, writing cannot be viewed in isolation from other factors in the learning environment (Mimirinis & Bhattacharya, 2007; Tolmie & Boyle, 2000; Ziegler, Paulus, & Woodside, 2006), nor is the reviewability of online communication sufficient encouragement in and of itself for critical reflective thinking (Seale & Cann, 2000).

A Model for CoP Blended Learning

While a full description of the process of model building is beyond the scope of this paper, it involved developing a preliminary model based on Wenger’s design framework (Table 1) and related concepts from constructivist learning approaches and adult learning theories, with attention to the fifth duality – face-to-face/online communication. This was used to redesign two blended LIS courses and iteratively tested and refined during implementation.
The model (Figure 1) focuses on the four CoP dualities – negotiating the meaning of core concepts, negotiating models of practice, negotiating expertise, and negotiating identity-leadership – which I believe students should ideally experience in a blended learning LIS classroom. These dualities are articulated through the three modes of belonging (engagement, imagination, alignment) and enacted through face-to-face and online communication (the fifth duality), requiring a negotiation of the benefits and costs of each medium to meet learning needs. The desired result of negotiating all five dualities is the personal realization of professional identity, principles, practices, values, and leadership skills that help students become accepted and contributing members of the profession.

Purpose of the Study

The purpose of this study is to test the CoP framework for its efficacy in designing blended learning environments that ground students in the core concepts, practices, and values of library and information science. These questions guided the study:

1.    How effective is the CoP framework for supporting student growth toward knowledge of professional principles, skills of practice, appreciation of the values of the profession, leadership skills, and professional identity?
2.    How effective is the CoP framework for guiding the development of effective blended learning environments?
3.    What refinements were made to the CoP model during implementation?
4.    What changes from previous teaching practice were required to implement the CoP model?
5.    Do the advantages outweigh the disadvantages of using the CoP framework for blended learning to achieve student learning objectives?

Methodology

The approach taken in this study is design-based research (DBR), a young methodology that has been undergoing renewed scrutiny and revitalization in the past five years. Introduced by Brown (1992) and Collins (1992) as “design experiments,” it was a response to laboratory research in education that lacked ecological validity due to its isolation from complex educational settings. DBR involves the integration of design, theory, research, and practice in naturalistic settings (P. Bell, 2004). The focus of DBR is generally innovation in educational interventions, often but not always involving technology development.
According to Edelson (2002, p. 108), “design research explicitly exploits the design process as an opportunity to advance the researchers’ understanding of teaching, learning, and educational systems.” The design and the hypotheses upon which it is based are continually refined during implementation until a more coherent theory emerges, embodied in the design. Sandoval (2004) calls these hypotheses “embodied conjectures.” The embodied conjectures in this study are the CoP learning processes (Figure 1).
Data collected include a comprehensive record of the design process, how learning was organized and supported, and data about the learning processes and outcomes (Borko, Liston, & Whitcomb, 2007, p. 7). This study is being conducted in two phases: formative and summative. This paper reports on the formative analyses conducted during the implementation of the model in fall 2008. These analyses focused on how the model was implemented and how student learning was supported through modifying conjectures and refining the model. A summative analysis will be conducted to determine how the features of the model affected student learning and outcomes. Data will be collected from students on perceptions of gains made in knowledge, skills, professional identity, and values of the profession using questionnaires and analyzing communication in the course related to the dualities. The summative analysis is expected to yield more clearly defined constructs, a refined coding scheme, and modifications to the model and the teaching and learning strategies and tools. This will initiate another cycle of implementation, analysis, and refinement.
The strengths and weaknesses of this study are those that characterize DBR in general (Hoadley, 2004). Its strengths are that: (1) it presents a realistic view of learning as the result of a complex interaction between educational intervention, learner psychology, personal histories, and situational contexts; (2) it recognizes that the educational intervention, the blended CoP learning model, is itself an outcome (i.e., dependent not independent variable); and (3) as in qualitative research, it provides detailed reports from the insider’s intimate knowledge of the design process, the students, and teaching and learning in context. DBR may in fact be more rigorous than experimental research at aligning theory, treatments, and measurement in complex classroom situations (Hoadley, 2004, p. 204). Its major weaknesses are: (1) the threat to rigor when the designer is also the researcher and the implementer, (2) the difficulty of transferring a locally sensitive design to other contexts, and (3) the difficulty of ascertaining whether outcomes can be attributed to the design rather than other complex factors operating within the learning environment.

Implementation of the Model

Guided by the model, I refined course content, structure, strategies, and tools in two courses (Reference & Online Services and Internet Fundamentals & Design). The subject of the course determined the focus of the dualities. I also identified face-to-face and online strategies to facilitate learning. This process of “embodying” the model in course design is a process of defining and refining the model’s constructs as classroom practices in response to a particular course with particular students.

Negotiating Identity and Leadership

Of the four substantive dualities, negotiating identity and leadership is the most encompassing and perhaps the most challenging, one that is a lifelong effort that begins with a decision to educate oneself in the profession. This duality is a creative tension between taking ownership of community meanings and taking leadership in the community based on expertise, influence, and relationships built over time. I believed that engaging students to find personal meaning in the focus of the course and to commit to collaborative learning were the most critical early tasks. My strategies from the first class session included building an atmosphere of trust to encourage confidence building, risk taking, and relationship building. I expressed enthusiasm for the subject and the learning process, in order to stimulate the acquisition of basic knowledge, curiosity, exploration, and intrinsic motivation (Bain, 2004). I shared my commitment to the goals and values of the profession at opportune moments in discussions and encouraged students to do the same. I also built into discussions and assignments student reflections on personal philosophy related to the subject. I viewed myself as a co-learner, described my learning processes related to the subject, and encouraged students to share their insights and challenges related to learning. I assigned participation in online discussion forums and student blogs to foster identification with the class community and to provide additional opportunities to reflect on and influence emerging meanings. Group work and team projects such as website design provided opportunities to lead, inspire, share expertise, persuade, and advocate for action.

Negotiating the Meaning of Core Concepts

The core concepts of the subject are the basic building blocks of knowledge. Negotiating their meaning involves a creative tension between action and the concepts that represent that action, the proverbial tension between theory and practice. It also involves a creative tension between previously held assumptions and new understandings. Students initially engaged with core concepts through the syllabus, which describes the knowledge, skills, and processes expected as learning outcomes. The syllabus structured the main course websites created with Moodle, MediaWiki, or PBwiki , which I encouraged students to contribute to. Other engagement strategies included readings to scaffold the learning of core concepts; lectures with Socratic questioning to ascertain what students know and to build upon it; and structured and open discussions based on dialogic inquiry (Wells, 1999) to encourage collaborative questioning and synthesis. In addition, the students and I exchanged ideas via wiki page creation and comments, email exchanges, feedback on assignments, and responses to student blogs.
Face-to-face group discussions and problem solving provided opportunities for students to orient to, explore, and collaboratively reflect on core concepts. Online group exercises, student reports, and summaries of in-class group work posted on the class wiki were additional venues for sharing, reflection, and creative thinking.
Assignments such as search exercises, pathfinders, and web page construction required that students put core concepts into practice. Group presentations that required critical thinking about issues, problems, and values were part of the process of aligning ideas and vocabulary to the discourse of practice. Clear rubrics for assessment were boundary objects that helped me negotiate meaning with students. As well as providing individual feedback on assignments, I also posted on the class wiki summary comments and exemplary student work (posted anonymously). This provided advice on how to solve professional problems that built upon student work and created a community knowledge base.

Negotiating Models of Practice

Negotiating models of practice involves a creative tension between one’s own mental models of an area of practice and those valued by LIS professionals and experts. This process focuses the learning on developing serviceable models of practice, achieved through an iterative process of exposure to models of practice, constructing one’s own models from existing knowledge and practical experience, reflection on practice, and model building and refinement. One example is how the views that students held of information seeking behavior in their own lives changed with exposure to Kuhlthau’s (2004) model of information seeking and their own practical experience of seeking information for assignments and observing the information seeking of others.
Concept mapping allowed students to explore the content and structure of practice. Concept maps posted on the class wiki supported exploration and reflection. Face-to-face simulations such as reference interview roles plays gave students practical experience. Online simulations (e.g., chat reference simulations) were also used to explore and reflect on models of practice. Project-based and case-based learning were used to help students refine models of practice, coordinate group activity, self-assess, and align to standards of good practice. My wiki postings of general feedback and student assignments posted voluntarily on the class wiki supported the development of shared standards and best practices and added to the community knowledge base.

Negotiating Expertise

Negotiating expertise involves a creative tension between one’s existing expertise and that of professionals and experts in the field. Engagement strategies included sharing stories from my own professional life and inviting stories from students. Students often commented that hearing from their peers was especially meaningful and instructive. The students and I also linked class discussions to real world events whenever possible. Blogs, personal pages, and spontaneously created new pages on the class wiki supported online sharing of experiences and expertise and added to the community knowledge base. Exposure to professionals in action is essential for negotiating expertise. Guest speakers, informational interviews, and observations exposed students to real world practices, while multimedia presentations provided vicarious experience. These experiences were discussed in class and online to probe and deepen understanding related to the discourse and patterns of practice. Students monitored LIS listservs and posted summaries and responses on issues of personal interest on the class discussion forum. Through collaboration on group projects, students developed shared standards, practices, and values around centers of expertise within the group.

Discussion

This section summarizes the main points of the formative analyses of the effectiveness of the CoP model for supporting student learning and guiding the development of blended learning environments, model refinement during implementation, and changes in teaching practices. It concludes with an assessment of the model’s advantages and disadvantages.
The CoP model is more abstract and broader in scope than the learning objectives specific to a given course (denoted by the center circle in Figure 1). As Figure 1 suggests, the features of the model serve as a systematic yet flexible set of principles that surround the specific student learning outcomes for a given course and guide the design of learning environments toward professional principles, practices, values, and leadership skills. Supporting student learning requires balancing each duality with its three modes of belonging, and embodying these in the course content, structure, strategies, and tools. In implementation, each course requires an adaptation of the model based on the subject matter and needs of the students. Assessing the effectiveness of the model requires not only determining how well it helps each student achieve course learning outcomes but also how well it supports his/her growth as a member of an LIS community of practice. These are challenging issues that the summative analysis will grapple with.
Assessment of the model’s effectiveness for guiding the development of blended learning environments faces similarly challenging issues. I believe assessment should be considered at three levels: micro, meso, and macro. I define micro-level assessment as an evaluation of ease of use of the technology, similar to the usability testing advocated by Jakob Nielsen.  Usability issues that result in a failure to achieve common ground will inhibit communication and possibly derail the blended learning process at the outset. At the micro level, the formative analysis identified difficulties students had using multiple tools. While those more experienced with online tools and communication reported few difficulties, those with less experience reported confusion about what tool to use for what purpose and where to find information online. New users expressed trepidation but also confidence and a sense of accomplishment after use.
Meso- and macro-level assessments will be addressed in the summative analysis. I define meso-level assessments as those that address appropriate uses of technology to support social constructivist pedagogy, with attention to issues such as establishing norms of communication, online facilitation strategies, and psychological barriers to online communication. I consider the guidelines provided by Preece (2000) and Paloff and Pratt (1999, 2003), for example, to be meso-level treatments. Macro-level assessments are the most challenging and address the capacity of blended learning environments to support the CoP dualities. This extends beyond an assessment of the communication function or educational technology function of the online tools. Larreamendy-Joerns and Leinhardt, (2006, p. 591-2) describe the challenges well:

Although online learning environments that allow for social interaction constitute a remarkable advance, they should not be construed as inevitably conducive to learning solely because student-student and student-instructor exchanges take place. Nor should they be understood as obviously consistent with … efforts to nurture communities of practice … The design of online environments should be primarily dictated by an understanding of the epistemic and discursive practices that constitute disciplinary communities, and not by pedagogical considerations and technologies that short-circuit the engagement of students.

Regarding refinements made to the model during implementation, this was an ongoing effort to better understand Wenger’s ideas and interpret them accurately and meaningfully using the frameworks of social constructivism and adult learning theories. While Wenger does deal briefly with design for education (1999, p. 263-270), his advice is broadly suggestive rather than definitive and given in the form of questions to stimulate thinking.
The changes in my teaching practice during implementation were primarily a matter of self-training and disciplined commitment to a more refined teaching-learning model. I began to be influenced by communities of practice in early 2003, drawn to it for its resonance with my ideas of practice-based community learning. However, my teaching practices regarding CoP learning had been more instinctive and opportunistic than systematic. As I used the model, I continually referred to it, refined it by clarifying ambiguities, and examined the connection between it and my practice. I kept a reflective journal and used Schon’s (1983) guidance for reflection-in-action and reflection-on-action to record and respond to the complex, multilayered consideration of my observations of teaching and learning in my classes.
In particular, I realized that I needed to become more skilled at reflection-in-action when facilitating dialogic inquiry in order to be responsive in the moment to conversational contributions. A contribution may be relevant to subject knowledge (core concepts), patterns of practice, identity formation, communal knowledge creation, leadership development, instructional goals, and/or ascertaining the student’s ongoing understanding. This is a highly complex process and set of factors to consider. Students contribute much to the learning of their peers, but primary responsibility lies with the instructor for appraising the importance of each contribution and facilitating the making of connections to support learning. The instructor should also provide periodic summaries of the learning to promote imagination, reflection, and alignment and to mark progress and celebrate achievements.
I provisionally conclude that the main advantages of using the model are as follows: (1) The CoP approach of situating learning within a community, a domain, and a practice provides coherence for social constructivist learning approaches that build on the interdisciplinary expertise of students; the model also suggests mechanisms for sharing that expertise and building collective knowledge on an ongoing basis. (2) The CoP dualities are a guide for prioritizing blended learning design choices and have the potential to guide curriculum planning, implementation, and assessment in practitioner education programs. (3) The use of the CoP approach with social software can serve as a model for blended learning and librarianship for future librarians. The main disadvantages to faculty of this CoP-based approach to blended learning are: (1) its complexity, (2) time needed to implement the model, (3) lack of support for course redesign, (4) difficulty of acquiring new teaching and technology skills, and (5) risk factors associated with this type of course (cf. Vaughan, 2007).

Conclusion

The CoP model serves to focus attention on broader professional development goals, much as a program’s mission and goals attempt to do. A typical mission statement might include wording such as “educates students in the principles, practices, and ethics of library and information science, imbues them with a sense of service to diverse populations, and prepares them to be life-long learners and active leaders in a rapidly changing information society. ” While the specific learning outcomes for each course differ, aggregated course learning outcomes should be the means of achieving the curricular goals and mission of a program.
In practice, this is a difficult challenge. Within a local community (e.g., the faculty of a LIS program), perspectives on how a mission should be accomplished will differ. The LIS field itself is undergoing redefinition, so global LIS CoPs may also have differing characteristics and emphases. While online tools overcome many physical and temporal barriers to communication, they also create barriers to deep learning and meaningful community building. As blended and online learning environments grow increasingly attractive to busy students and professionals, we may also find it increasingly difficult to build community on common ground. Under the pressure of these dispersive forces, the communities of practice model may prove valuable as an integrated, holistic guide to course, curriculum, and program planning that maintains a clear focus on professional principles, practices, values, and leadership.

References

Applebee, A. N. (1984). Writing and reasoning. Review of Educational Research, 54(4), 577-596.

Bain, K. (2004). What the best college teachers do. Cambridge, MA: Harvard University Press.

Bandura, A. (1977). Social learning theory. Englewood Cliffs, NJ: Prentice Hall.

Bandura, Albert. 1986. Social foundations of thought and actions. Englewood Cliffs, NJ: Prentice-Hall.

Barab, S.A., MaKinster, J.G., & Scheckler, R. (2004). Designing system dualities: Characterizing an online professional development community. In S.A. Barab, R. Kling, & J.A. Gray, J.A. (Eds.), Designing virtual communities in the service of learning (pp. 53-90). Cambridge: Cambridge University Press.

Bell, P. (2004). On the theoretical breadth of design-based research in education. Educational Psychologist, 39(4), 243–253.

Bell, S. J. & Shank, J. D. (2007). Academic librarianship by design: A blended librarian’s guide to the tools and techniques. Chicago: American Library Association.

Borko, H., Liston, D., & Whitcomb, J. A. (2007). Genres of empirical research in teacher education. Journal of Teacher Education, 58(1), 3-11. DOI: 10.1177/0022487106296220.

Bruner, J. (1990). Acts of meaning. Cambridge, MA: Harvard University Press.

Brown, A. L. (1992). Design experiments: Theoretical and methodological challenges in creating complex interventions in classroom settings. Journal of the Learning Sciences, 2(2), 141–178.

Collins, A. (1992). Toward a design science of education. In E. Scanlon & T. O’Shea (Eds.), New directions in educational technology (pp. 15–22). Berlin: Springer-Verlag.

Dewey, J. (1916). Democracy and education. New York: The Free Press.

Edelson, D. C. (2002). Design research: What we learn when we engage in design. Journal of the Learning Sciences, 11(1), 105-121.

Edwards, D. (1991). Discourse and the development of understanding in the classroom. In O. Boyd-Barrett & E. Scanlon (Eds.), Computers and learning (pp. 186-204). Wokingham, England: Addison-Wesley.

Garrison, D. R., & Vaughan, N. (2007). Blended learning in higher education: Framework, principles, and guidelines. San Francisco: Jossey-Bass.

Garrison, D. R., Anderson, T., & Archer, W. (2000). Critical inquiry in a text-based environment: Computer conferencing in higher education. Internet and Higher Education, 2(2-3), 87 - 105.

Haythornthwaite, C. et al. (2000). Community development among distance learners: Temporal and technological dimensions. Journal of Computer-Mediated Communication, 6(1). Retrieved September 23, 2008, from http://www3.interscience.wiley.com/cgi-bin/fulltext/120837807/HTMLSTART.

Hoadley, C. (2004). Methodological alignment in design-based research. Educational Psychologist, 39(4), 203–212.

Kim, A. J. (2000). Community building on the web. Berkeley, CA: Peachpit Press

Kuhlthau, C. 2004. Seeking meaning: A process approach to library and information services. 2nd ed. Westport, CT: Libraries Unlimited/Greenwood Press.

Lampert, M. (2002). Appreciating the complexity of teaching and learning in school: A commentary on Cobb; Forman and Answell; McClain; Saxe; Schliemann; and Sfard. Journal of the Learning Sciences, 11(2-3), 365-368.

Larreamendy-Joerns, J. & Leinhardt, G. (2006). Going the distance with online education. Review of Educational Research, 76(4), 567-605.

Lave, J., & Wenger, E. (1991). Situated learning: Legitimate peripheral participation. Cambridge, MA: Cambridge University Press.

Mercer, N. (1995). The guided construction of knowledge. Tonawanda, NY: Multilingual Matters.

Mezirow, J. (1978). Education for perspective transformation: Women’s re-entry programs in community colleges. New York: Teachers College Press.

Mezirow, J. (2000). Learning to think like an adult: Core concepts of transformation theory. In J. Mezirow (Ed.), Learning as transformation (pp. 3-34). San Francisco, CA: Jossey-Bass.

Mimirinis. M. & Bhattacharya, M. (2007). Design of virtual learning environments for deep learning. Journal of Interactive Learning Research, 18(1), 55-64.

Ong, W. (1988). Orality and literacy: The technologizing of the word. London; New
York: Routledge.

Paloff, R. M. & Pratt, K. (1999). Building learning communities in cyberspace: Effective strategies for the online classroom. San Francisco, CA: Jossey-Bass.

Paloff, R. M. & Pratt, K. (2003). The virtual student: A profile and guide to working with online learners. San Francisco, CA: Jossey-Bass.

Piaget, J. (1932). The language and thought of the child. London: Routledge.

Preece, J. (2000). Online communities: Designing usability and supporting sociability. New York: John Wiley.

Rourke, L., Anderson, T. Garrison, D. R., & Archer, W. (1999). Assessing social presence in asynchronous, text-based computer conferencing. Journal of Distance Education, 14(3), 51-70.

Rourke, L., Anderson, T., Garrison, D. R., & Archer, W. (2001). Methodological issues in the content analysis of computer conference transcripts. International Journal of Artificial Intelligence in Education, 12. Retrieved Oct. 13, 2008, from http://aied.inf.ed.ac.uk/members01/archive/vol_12/rourke/paper.pdf.

Salomon, G. and Perkins, D. (1998). Individual and social aspects of learning. Review of Research in Education, 23(1), 1-24.

Sandoval, W. A. (2004). Developing learning theory by refining conjectures
embodied in educational designs. Educational Psychologist, 39(4), 213–223.

Schon, D. A. (1983). The reflective practitioner. New York: Basic Books.

Seale, J. K. & Cann, A. J. (2000). Reflection on-line or off-line: The role of learning technologies in encouraging students to reflect. Computers & Education, 34(3-4), 309-320

Shea, P., Li, C. S., & Pickett, A. (2006). A study of teaching presence and student sense of learning community in fully online and web-enhanced college courses. The Internet and Higher Education, 9(3), 175-190.

Snyder, W.M., Wenger, E., & de Sousa Briggs, X. (2004). Communities of practice in government: Leveraging knowledge for performance. The Public Manager 32(4), 17-21.

Storberg-Walker, J. (2008). Wenger’s communities of practice revisited: A (failed?) exercise in applied communities of practice theory-building. Research Advances in Developing Human Resources, 10(4), 555-577.

Tolmie, A., & Boyle, J. (2000). Factors influencing the success of computer-mediated communication (CMC) environments in university teaching: A review and case study. Computers and Education, 34, 119-140.

Vaughan, N. (2007). Perspectives on blended learning in higher education.
International Journal on ELearning, 6(1), 81-94.

Vygotsky, L.S. (1978). Mind and society: The development of higher mental processes. Cambridge, MA: Harvard University Press

Wells, Gordon. 1999. Dialogic inquiry: Toward a sociocultural practice and theory of Education. Cambridge: Cambridge University Press.

Wenger, E. (1998). Communities of practice: Learning, meaning, and identity. Cambridge, MA: Cambridge University Press.

Wenger, E., McDermott, R., & Snyder, W.M. (2002). Cultivating communities of practice: A guide to managing knowledge. Cambridge, MA: Harvard Business School Press.

Yukawa, J. (2007). Factors influencing online communication style in LIS problem-based learning. Journal of Education for Library and Information Science, 48(1), 52-63.

Yukawa, J. (2006). Co-reflection in online learning: Collaborative critical thinking as narrative. International Journal of Computer-Supported Collaborative Learning, 1(2), 203-228.

Ziegler, M., Paulus, T., & Woodside, M. (2006). Creating a climate of engagement in a blended learning environment. Journal of Interactive Learning Research, 17(3), 295-318.

lililuo@slis.sjsu.edu

Peer communications and establishment of a peer network are a crucial component of educational experiences for graduate students. However, in distance education programs, this component is significantly weakened because students take classes mostly in an isolated online environment and lack the opportunities to engage in real-time interactions with their peers (Nicolson, 2005). As more and more LIS schools are delivering education via online programs, it becomes ever more important to take notice of this weakness of online environments and design strategies to compensate for the rather “removed” nature of distance education and increase the sense of “presence” and “community”. One of the solutions is to use social networking Websites to help graduate students strengthen their peer communications, build peer networks and maintain peer connections in distance LIS programs.

In Spring 2008, School of Library and Information Science at San Jose State University (SLIS at SJSU) launched a school-wide social networking website “SLISLife” to help students in the online program to become more connected and engaged in their educational experience. The proposed study will examine how SLISLife and other general-purpose social networking Websites such as MySpace and Facebook are being used by SLIS students to strengthen peer communications in the distance education program.

A self-administered survey questionnaire will be distributed to each student enrolled in the distance education program at SLIS, eliciting information about 1) how they currently communicate/interact with their peers for both course-related purposes and professional networking purposes; 2) the perceived difficulty or inconvenience of peer communications/interactions in the distance education program; 3) their awareness of social networking Websites; 4) the purposes of their current use of social networking Websites; 5) the perceived advantages of using social networking Websites for professional networking among peers; 6) their anticipated/actual use of social networking Websites in establishing a community and strengthening peer connections in the distance education program. The study will be a Web-based survey study. The questionnaire will be pilot-tested and then sent to each student via a listserv to which all the graduate students subscribe. Participation in the study is completely voluntary and anonymous.

This study will provide a new perspective to examine distance education, furthering the understanding of how new technologies can be used to enhance learning experiences for graduate students in LIS programs. Social networking Websites have been proven to be effective in supporting both professional networking and casual mingling (Cachia, Compano, & Da Costa, 2007), and this study will find out how they can be tailored to serve the needs of distance education students. Thus, results from this study will help distance education programs in the field of LIS and beyond obtain a clear idea of students’ needs for peer interactions and community building as well as how their needs can be addressed by the use of social networking Websites. Hence, distance education programs can devise strategies to appropriately incorporate Social networking Websites in the curriculum and create a more engaging, interactive, and community-driven learning environment for graduate students.

References

Nicholson, S. (2005). A framework for technology selection in a Web-based distance education environment: Supporting community-building through richer interaction opportunities. Journal of Education for Library and Information Science, 46(3), 217-233.

Cachia, R., & Compañó, R., & Da Costa, O. (2007). Grasping the potential of online social networks for foresight. Technological Forecasting and Social Change, 74(8), 1179-1203.

Library and information service clients’ needs change suddenly during community wide disasters. Library research and literature about disasters emphasizes staff, client and collection physical protection as well as return to normal services, but not the special services librarians can provide during community wide disasters. A few librarians have been information first responders with much needed professional services. Just like health care professionals who understand that they have special skills useful in a disaster, these librarians practiced their profession even without their normal working environment and resources.

The authors are in the third year of a four year project, “Investigating Library and Information Services During Community-Based Disasters: Preparing Information Professionals to Plan for the Worst” supported by a National Leadership Grant from the Institute for Museum and Library Services. The end product of the project will be case based modules to train students and practitioners how to improvise such services in a disaster.

The research phase of the project sought evidence of how individual librarians in a variety of libraries and a variety of natural and accidental disasters actually have designed and provided such unanticipated services. The research method of grounded theory development was based on lessons learned in preliminary studies information services improvised in Louisiana and Mississippi after hurricanes Katrina and Rita.

Through various library associations the researchers invited librarians to participate in a survey designed to identify possible study participants. Through searches of news reports and personal contacts, they identified additional librarians with similar experiences resulting in a contact pool of approximately two hundred librarians who had recently experienced community wide disasters. Most had carried out collection and staff protection procedures and returned to normal services. Many had performed commendable volunteer services. A minority had improvised new services for the suddenly changed information needs of their client community.

After the researchers had developed and tested the interview protocol, a representative sample of twenty librarians providing services in different disasters and different kinds of libraries participated in extensive recorded interviews. The participants also provided documentary evidence of their services. The interviews were transcribed and member-checked for accuracy. The resulting multiple case study data was analyzed using open and axial coding (with NVIVO software) to discover common themes in a variety of cases.

The researchers will develop case study materials to exemplify the best practices identified during the data collection and analysis phase. The material developed will be a combination of generic case studies and individual explanatory case studies. Each case study will include a general description of the particular community-based disaster, the responses taken by the information professional(s) involved, and a guide for discussing the issues involved in the situation.

This paper will report preliminary findings from the research phase and progress on the educational module building phase.