Introduction

This  is  a  year  two  report  of  a  3-­year  longitudinal  design  that  tracks  the  Journal   of   Education   for   Library   and   Information   Science   (JELIS)   subscribers/readers’   satisfaction  with  the  scholarly,  double-­blind,   peer-­reviewed  publication  of  the  Association   for   Library   and   Information   Science   Education   (ALISE).   According   to   Park   and   Capps,   last   year’s   2009   JELIS   survey   indicated   “that   relevance   and   quality   of   content   are   major   factors   contributing   to   the   dissatisfaction   with   JELIS”   (2010,   p.  7).  This  year’s  JELIS  2010  Satisfaction   Survey   addresses   these   factors   and   highlights   the   changes   noted   in   user   satisfaction   as   indicated   by   the   feedback   of   the   anonymous  participants  (n  =  76).
As  part  of  their  stated  editorial  goals  to   make   JELIS   a   communication   dissemination  hub,  Burnett  and  Kazmer  demonstrated  a  concerted  effort  over  the  past  year  to   raise  the  acceptance  standards  of  the  manuscript   publications.   They   also   initiated   the  process  of  activating  the  evaluation  of   JELIS   for   inclusion   in   ISI.   In   addition   to   keeping   the   journal   publication   on   schedule,   the   Editors   continued   to   improve   the   journal’s  presentation  format;  e.g.,  reducing  the  table  size  and  adding  the  month.
The   web   presence   role   of   the   jelis.org   community   was   expanded   to   meet   the   needs   of   a   dynamic   research   community.   Operating   within   copyright   boundaries,   the   site   provides   high-­quality   content   online   and   communication   tools   that   may   be   used  to  initiate  scholarly  debate  on  current   issues.   The   jelis.org   site   is   designed   to   address  the  information  needs  of  JELIS  readers,  reviewers,  and  future  contributors.

Research  Design

The   JELIS   Satisfaction   survey   was   designed  to  assess  the  satisfaction  of  readers   and   to   inform   the   editorial   decision   processes.  The five major categories devised for   the   2009   survey   were   carried   over   to   the  2010  administered  online  survey.  The   areas   were:   (1)   JELIS   use,   (2)   degree   of   satisfaction,  (3)  topics  on  demand,  (4)  loyalty,   and   (5)   demographic   questions.   The   questionnaire   scope   covered   the   JELIS   subscribers/readers’  satisfaction  on  the  issues  published  during  the  2009  golden  anniversary  year,  50(1–4),  and  on  the  newly   revised  JELIS  website.
There  were  minor  2010  survey  revisions  to reflect participant comments received from  the  previous  year’s  questionnaire  administration.  To  investigate  the  sharing  of   JELIS  copies  with  others,  a  Yes/No  option   was  adopted  over  the  previous  year’s  multiple-­choice  response  format.  The  instructions   for   the   question   related   to   journal   ranking  were  revised  to  improve  readability plus a grammatical error was fixed. In addition   to   these   basic   readability   changes,   questions   on   the   recently   redesigned   JELIS  website  were  added  to  capture  user   satisfaction  with  the  website,  the  purposes   of  site  visitations,  and  future  site  enhancement  recommendations  that  would  address   currently  unmet  JELIS  community  needs.
The  survey  was  announced  at  the  2010   ALISE   conference   and   on   the   jESSE   listserv.  A  recruitment  blurb  was  also  posted on   the   JELIS   website.   The   respondents   were  asked  to  indicate  their  reading  and   use   patterns   plus   their   overall   satisfaction   with  JELIS.  The  survey  was  available  online   between   January   13th   and   February   12th,  2010.

Results  and  Discussion

The   questionnaire   data   was   exported   to   an   Excel   spreadsheet   from   the   online   survey   system   and   analyzed   with   SPSS.   Open-­ended   question   responses   were   manually   coded.   In   compliance   with   the   Internal  Review  Board  to  obtain  informed   consent,   the   one   negative   consent   response  participant  data  was  removed  from   the  study  prior  to  analysis.

Demographic  Characteristics

The   demographic   questions   gathered   occupational   position,   organizational   affiliation, and geographic region information.  The  characteristics  of  respondents  to   the   2010   survey   were   similar   to   those   in   the   2009   survey.   As   shown   in   Appendix   A,  70%  of  this  year’s  respondents  reported   to   be   faculty   members.   The   faculty   classification included professors and dean/department   chairs.   The   remaining   participants were self-identified as students (17%)  or  librarians  (6%).   Most  of  the  respondents  were  from
doctorate-­granting   universities   or   master’s   college   and   universities,   67%   and   22%   respectively.   Eighty-­three   percent   of   the   re-­ spondents   lived   in   the   United   States   at   the   time  of  the  survey;  the  remaining  11%  per-­ cent  of  respondents  were  distributed  across   two   categories,   Canada   (7%)   and   Asia   and   Pacific (4%). The demographic characteristics were used to obtain readership profile information.   Similar   to   the   2009   results,   Appendix   A  illustrates  that  the  majority  of   respondents   in   the   2010   respondents’   profile were from an academic setting.

JELIS  Use

Respondents   were   asked   how   they   use   JELIS  in  terms  of  reading,  citation,  publication,  length  of  subscription,  sharing,  and  retention of the journal. Ninety-five percent  of   respondents   this   year   reported   to   have  either  read  or  looked  through  an  issue   of   JELIS.   This   was   a   substantial   increase   over  last  year’s  89%;  5%  reported  spending  more  than  an  hour  on  an  issue  upon  receipt  while  72%  spent  30  minutes  or  less.   In   addition,   68%   of   the   respondents   indicated  that  they  skim  the  titles  to  be  aware   of  the  contents  and  read  the  articles  of  interest  (see  Figure  1).
46%   of   respondents   reported   that   they had  cited  an  article  from  JELIS.  This  was   a   substantial   gain   over   the   30%   reported   in   the   previous   year.   When   asked   how   often   they   cite   an   article   in   JELIS,   there   was   little   change   over   last   year.   Twenty-­ one   percent   still   responded   ‘never,’   40%   responded  ‘less  than  once  a  year,’  24%  responded  ‘once  a  year,’  and  16%  responded   ‘more  often  than  once  a  year.’  There  was   a  slight  downward  shift  from  34%  to  30%   of  the  respondents  that  indicated  they  had   previously   published   in   JELIS.   Fifty-­one   percent  reported  that  not  sharing  a  copy  of   JELIS  with  others.  When  asked  how  long   they  keep  a  copy  of  JELIS,  this  year’s  57%   was   in   keeping   with   last   year’s   58%   that   responded ‘indefinitely.’ Appendix B contains   additional   2009–2010   comparison   results  related  to  JELIS  use.

JELIS  Satisfaction

Figure   2   shows   the   changes   of   JELIS   subscribers/readers’   satisfaction   on   six   constructs   as   well   as   overall   satisfaction.   Based  on  these  numbers,  the  respondents’   satisfaction   rates   on   all   measurements   in   2010  are  higher  than  in  2009  and  they  may   be   interpreted   as   a   positive   indication   of   JELIS  subscribers/readers  satisfaction  rate   increases.
To test the statistical significance of rate increase  between  2009  and  2010,  T-­tests   were  conducted,  and  the  results  reveal  that  there was a statistically significant increase in  overall  satisfaction  from  2009  to  2010  (t   =  –2.681,  p  <  0.05).  In  terms  of  constructs,   in  particular,  the  mean  value  of  JELIS  subscribers’   satisfaction   for   three   constructs,   including   relevance,   timeliness,   and   layout, was significantly increased. However, there was no statistically significant change  in  2009  and  2010  in  terms  of  quality   and   breadth   of   topics,   which   suggests   editors   and   board   members   need   to   continue  their  effort  to  enhance  the  quality  of   the   journal   and   to   improve   the   breadth   of   journal  topics.

JELIS  Website

The  JELIS  editors  redesigned  the  web-­ site  to  include  blog  communication  tools   to   facilitate   an   interactive   relationships   between   readers-­authors   and   readers-­ readers.   The   redesign   of   the   JELIS   website  was  announced  in  the  jESSE  listserv   on   January,   2010.   Due   to   the   short   time   period   between   the   launching   of   the   revised  jelis.org  site  and  the  survey  administration   period,   the   readers’   ability   to   access   the   website,   only   36.8   %   of   total   respondents   had   visited   the   JELIS   web-­ site.   Overall,   the   majority   of   these   respondents were satisfied with the website contents  and  design.
The   most   frequently   stated   purpose   of   those   visiting   the   JELIS   website   was   to   ‘Find  author  guidelines’  (n  =  16).  This  was   followed   by   ‘Read   full-­text   articles   of   in-­ terest’  (n  =  12),  ‘Find  reviewer  guidelines’   (n   =   11),   ‘Find/Verify   citation   information’  (n  =  11),  and  ‘Review  recent  trends   in  LIS  education  research’  (n  =  10).
The   questionnaire   included   an   open-­ ended   question   to   identify   needs   of   website  users.  Even  though  there  are  only  seven   comments,   three   of   them   were   related   to  open  access  and  full-­text  service.  As  the   two  results  indicate,  open  access  is  a  topic   of  interest  to  JELIS  readers,  and  an  objective  of  the  editors.

Conclusion

This   article   reported   on   survey   results  that reflected respondent perceptions prior to  January  2010.  Satisfaction  is  a  dynamic   process.   The   JELIS   editors,   staff,   and   the   authors   continue   to   work   toward   the   production  goal  of  a  high  quality  journal  that   meets   the   needs   of   the   ALISE   members   and  the  research  community  at  large.  Your   participation   and   feedback   will   assist   in   improving   the   readability   and   reputation   of   JELIS.   If   you   missed   the   opportunity   to   share   your   thoughts   in   2010,   we   look   forward   to   hearing   from   you   in   the   2011   JELIS  Satisfaction  survey.

Acknowledgement

The  JELIS  co-­editors  and  editorial  staff   sincerely   thank   all   of   the   respondents   of   the   JELIS   2010   Satisfaction   survey.   Special   thanks   to   Lorna   Peterson   (ALISE   President)   and   Kathleen   Combs   (Executive  Director)  for  their  support  in  publicizing  the  survey.

References

Park,   S.,   &   Capps,   J.   (2010).   2009   JELIS   Satisfaction   Survey.   Journal   of   Education   for   Library   and  Information  Science, 51(1),  4–8.

Keywords: service-learning, equity of access, latinos, LIS curriculum, diversity, cultural competence, qualitative analysis

Two challenges facing 21st century librarians are rapid growth in information technology and the growing population of diverse students (U. S. Census Bureau, 2006), many of whom lack access to information and technology (Fairlie, 2005; Fairlie, London, Pastor, Rosner, 2006). Most LIS students are aware of disparities in society through classroom-based coursework on topics such as the digital divide, limited access to technology, and inadequate availability of linguistically and culturally relevant content (American Library Association, 2004). However, the reality of these social inequities is often beyond LIS students’ experiences, even though the likelihood is great that they will need to understand issues involving access for diverse populations during their professional career (Josey & Abdullahi, 2002).

The challenge is particularly significant for librarians who are culturally and ethnically distinct from the populations they    serve    (Lynch, 1998).    For    example, the majority of school librarians are older, monolingual, and do not generally come from the same ethnic background as their students (American Library Association, 2007). Efforts to prepare LIS students for    challenges of  today’ s society have been undertaken by LIS schools through incorporating courses on diverse populations into the general curriculum. However, the complexity of issues raised in these courses requires more than a theoretical understanding that comes from classroom instruction. Firsthand experience with diverse populations is often needed, particularly when those populations are from linguistic, cultural, ethnic, and socio-economic backgrounds that are different from that of the students.

Background

Service-Learning

Service-learning, a pedagogy which integrates professional learning experiences and academic curriculum (Billig, 2002; Eyler & Giles, 1999; Furco & Billig, 2002; Kraft, 1996; Pritchard, 2002), has been used to improve learners’ understanding of course content through service to communities. Real-life experiences become a learning and teaching tool to help students develop a deeper understanding of information (Rhoads, 1998). Students learn to work with rather than for communities (Weah, Simmons, & Hall, 2000) to develop interpersonal skills and social responsibility (Eyler, 2000; Eyler & Giles, 1999), two essential attributes required of LIS professionals to adequately provide service to communities (American Library Association, 1999).

Service-learning was introduced as a course requirement into an LIS graduate course taught by the author to narrow the gap between theory and practice (Bransford, Brown, & Cocking, 1999). The course, “Equity of Access for Diverse Populations,” examines the debate on the digital divide and other issues involving access to information and technology, such as availability of culturally diverse content for diverse populations including Latinos whose access is often limited. By providing students with greater awareness of sociocultural issues related to the course, a deeper under- standing of issues was anticipated (Valerius &   Hamilton, 2001). The requirement of service within a diverse community was seen as a way to provide LIS students firsthand knowledge and experience with populations different from themselves. Such experiences are seen as a way to help students develop cultural competence and respect for cultural differences (Montiel-Overall, 2009).

Equity of Access

Equity of access refers to equality in opportunities for diverse groups to access in- formation including access to library collections, library resources, and instruction. An aspect of discussions about equity of access involves the digital divide, which refers primarily to access to technology. Lack of access to resources, information and computer technology has been found to be high among minority and underserved populations (e.g., Native Americans, African Americans, Latinos, poor, elderly, and disabled), and is highest among growing populations of Latinos (Pew Hispanic Center, 2006). The increase in this population has resulted in an effort by LIS schools to prepare students to better understand Latinos and their information needs.

Two required components of the equity of access course are a written cultural heritage autobiography (CHA) and a required fieldtrip. The CHA is submitted electronically before the first class session, and provides an opportunity for students to discuss their cultural backgrounds. Rosaldo’ s (1989) definition of culture as daily activities shared by individuals and groups is used in conjunction with assigned course readings related to culture.

The second requirement is a fieldtrip at the beginning of the semester to a state-of-the-art elementary school library. The fieldtrip library is a two-story facility with 40,000 books, 25 computers, an electronic blackboard and document camera, and active participation of parents as library volunteers. The fieldtrip library is located in a geographic and socioeconomic area that differs from the service-learning communities.

Theoretical Framework for Experiential Learning

Service-learning is grounded in experiential learning and works of John Dewey (1938), David Kolb (1984) and others (e.g., Bruner, 1966), which provide a framework for incorporating experiential opportunities into instruction for LIS students. Kolb (1984) defines learning as a “process whereby knowledge is created through the transformation of experiences” (p. 38). Applying these theories to learning provides a lens to examine learning as “an intimate and necessary relation between the process of actual experience and education” (Dewey, 1938, p. 20), and “a continuous process grounded in experience” (Kolb, 1984, p. 27). Experiential learning also provides individuals with opportunities to contribute to society (Bruner, 1966, p. 167) and to build social capital within community groups and organizations. These multiple theoretical frameworks provide a broad background for studying the effect of service-learning with Latino and Spanish-speaking populations on library and information science (LIS) graduate students.

Problem Statement and Purpose

Equity of access is a major concern to LIS professionals inasmuch as populations of library users are increasingly those who are from underserved and underrepresented populations. Under- standing issues related to equity of access is critical for those graduating from LIS programs. This study examines graduate students’ understanding of equity of access issues by connecting content and service-learning to experiences in schools and public libraries serving large populations of Latinos. The research questions addressed include: (1) Does service-learning contribute to a better understanding of course related issues? (2) To what extent does service-learning contribute to depth of understanding of issues related to course content? (3) To what extent do service-learning experiences contribute to increased understanding of equity of access issues and cultural understanding? and (4) To what extent are social capital and civic involvement developed through service to community?

Procedures

Student Participants

Following approval from the Institutional Review Board, seven students who had enrolled in a 15 week course at a large university in a southwest metropolitan area consented to participate in the study in the fall of 2007. Students were at various stages in their graduate program in LIS and the course was one of several courses available to students to fulfill a diversity requirement for graduation. A total of four female and three male students participated in the study. One male student was a native Spanish speaker from South America. The remaining students spoke English only and had no previous experience working in libraries or with Latinos.

Procedure

Students agreed to complete 30 hours of service-learning with librarians at elementary schools or public libraries serving large populations of Latinos. Seven sites were selected by the instructor, two public libraries and five K–8 school libraries. Libraries were selected because the populations served were predominantly Latino. Schools selected had approximately 90% of children on free or reduced lunch.

Data Sources and Collection

LIS students met once a week during the semester for a two and a half hour face-to-face class where they participated in discussions of course readings and group activities. Students also engaged in online discussions and kept logs and fieldnotes of their service-learning activities. These were posted to a private space on an electronic course management tool, Desire to Learn (D2L). Students also submitted self-reflections and took pre- and postsurveys on service-learning. Questions on a 44-item service-learning survey were adapted from questionnaires developed by Simons and Cleary (2005) and Cram (1998). At the end of the semester, six students were interviewed. Various qualitative data were used to achieve triangulation. See Appendix A for sample pre-survey questions.

Data Analysis

Qualitative data for this study were grouped into four sets of items for analysis: (1) fieldnotes, (2) interviews, (3) self-reflections, and (4) pre- and postsurveys on service-learning. This article reports on data derived from these four sets. Analysis proceeded in the manner described by Miles and Huberman (1994). Coding of data was carried out by the researcher and two research assistants who identified broad a priori categories based on the research questions. Coders independently coded a student self-reflection based on categories to determine interrater reliability. A comparison of coding yielded high results. Each set of items was then coded independently; however, discussion between coders was ongoing. Preliminary categories identified included culture (CUL), equity of access (EQ), connection between SL and course content (CON), and change over time (COT), social capital (SC) and civic engagement (CE).

Member check occurred at weekly class meetings during discussions regarding the service-learning experiences and course readings. At that time issues and questions that appeared on online logs and fieldnotes were clarified. Small group discussions of course content also took place weekly with a focus on ongoing findings and conclusions involving assigned readings and service-learning experiences. Throughout this process, coders discussed stability of categories across data as findings emerged. Several categories were divided further into sub-categories and additional categories were added. This procedure was equivalent to “surfacing” and “filling-in” discussed in Lincoln and Guba (1985). Several iterations of coding occurred on each set of data to identify information that clustered together into categories. Recoding continued and tables and summaries were made for further analysis. When coders thought categories had reached saturation and no new categories could be identified, the four data sets were compared. Themes emerged from the final list of categories, from which further interpretation of data could be made. Results are reported in following section.

As with all studies, there are limitations to this study that must be considered in interpreting results. A limitation of this study was the length of time students were involved in service learning, which was limited by the duration of the course.

Results

From the extensive analysis discussed above, recurring themes emerged to address research questions regarding students’ understanding of course content (equity of access for diverse populations) and knowledge of the effect of Latino language and culture on equity of access, and the development of civic responsibility and social capital. Table 1 illustrates overarching themes and categories, including: technology and computers, economic disparities, language and literacy, culture, and social benefits. Themes are identified on the left side of the table. Categories appear in the center. Summary statements from multiple data sets appear on the right side.

Table 1: Themes and Categories Related to Equity of Access Based on Qualitative Data Analysis.

Technology

LIS students noted similar issues involving computers and technology at school and public libraries. These included insufficient computers, insufficient time on computers, inadequate computer literacy, and lack of maintenance of computers. For example, an LIS student assigned to a public library serving a predominantly Latino community noted that a request to repair computers submitted three weeks before he began his service-learning had still not been acted upon by the end of his service-learning several months later. “I understandably got a lot of [patrons’ ] questions (four the first hour) about when the computers would be up and running again, and if the library maintenance had been contacted.”

Another LIS student at a large midtown public library explained that although only 5% of the population in the neighborhood served by the library owned their own computers, the library had only eight computers for public use. These were in constant demand and required time limitations be placed on the computers. High demand resulted in many computers being available for only 15 minute periods, and a few computers being available for 45 minutes. An additional restriction on 45-minute computers was that they could only be used by a patron once a day. Patrons were often seen waiting up to one hour to use a computer. The student reported “. . . the neighborhood heavily uses the technology and could easily fill ten more computers in the evenings if they were available.”

At school libraries, systems were old and outdated, and when newer technology was made available, there were in- sufficient new applications to accommodate all students. Consequently, students simultaneously worked on old and new systems resulting in considerable wasted time for those using newer systems waiting for students to complete their work on older operating systems.

Another common situation was lack of basic and computer literacy skills demonstrated by Latinos in public and school libraries. Latino children at schools lacked basic keyboarding ability and seemed unfamiliar with “basic site for- mat.” For example, students were uncertain which box to put search information in (i.e., box for URL or search box). Catalog searches required a high level of spelling ability. A student looking for books on skateboards would be unsuccessful if the word were misspelled or entered incorrectly (e.g., skatebord). Consequently, the librarian spent much of her time spelling search terms for students.

Classes were offered at the public branch but were “rudimentary,” which surprised the LIS student who wondered if “illiteracy [made] using computer work difficult.” He noted that a library user took almost 20 minutes to write a sentence on a computer.

At schools, computer use was highly variable. During lunchtime, large numbers of students used computers to play games. At other times, computers were not used at all, suggesting the need for improved scheduling. Another reason computers were not used was that students were not allowed to use computers unless a teacher or librarian was present. At one site, computers were vacant much of the time for this reason.

Another commonly observed phenomenon at school libraries was inadequate space for computer use. School libraries, which were generally described as large open spaces, served multiple purposes including serving as a computer lab. However, libraries often served as multipurpose rooms. When libraries were used for other purposes such as staff meetings or testing (e.g., DIEBLES or speech therapy testing), computers could not be used. Libraries were also closed when they substituted as classrooms. An LIS student noted that a school library was closed for several months when a classroom air conditioning unit was being repaired and when a classroom was being painted. During that time, students had no access to computers or to the library. An LIS student noted differences between the library at her service-learning site and the fieldtrip school library, which had several rooms for different uses. It “had a room for just picture books and a room for all the other books, and a room upstairs where the older kids can feel separate. There wasn’t that kind of separation of space at [the service-learning school].”

Finally, adequate technical support was noted by LIS students as a challenge at both public and school libraries. While public libraries offered classes to patrons, lack of computer literacy of library users was daunting to volunteers who taught computer classes. At elementary schools, while librarians demonstrated a high level of technical ability, both students and teachers lacked the ability to use innovative technology. As an example, a teacher used an electronic board as a screen for an overhead projector (“She was projecting onto a SMART board ironically to use as the projector screen.”).

Economic Disparity

A second theme in Table 1 is economics, which was highlighted by disparities that emerged from the data. Students observed daily examples where equity of access was limited by the economic situation of library users. For example, LIS students noted that most students or patrons at their libraries did not have home computers although literature discussed in class indicated the digital divide had been bridged (U.S. Department of Commerce, 2002, 2004). Also noted was that patrons at public libraries were unable to pay even small fines, and replacement costs of $1.50 for lost library cards were also difficult for patrons. At a school, a child printed a homework assignment in the library because her family could not afford the cost of ink for a printer.

Another aspect of economic disparity identified was transiency of students. An LIS student noted that up to 20 students who were in one school at the beginning of the year couldn’t be found in that school two months later. LIS students observed that many students traveled back and forth from the United States to Mexico. Also, some students moved frequently. Others lived out of their cars.

The economic situation of Latino families affected the amount of time parents were able to volunteer in classrooms. At some schools, unless parents were in paid positions as cafeteria helpers or crossing guards, they rarely participated at school. A student commented “. . . in an inner city environment, the parents are busy with providing for their family, . . . they don’t have the time to invest into supporting the school as opposed to [the fieldtrip school] where [there were] 27 to 29 volunteers, almost all of them are parents . . .”

The result of economic disparities was noted by an LIS student who stated,

“Those without access are prevented from participating in social and political activities and events, such as finding voter information, witnessing debates, finding the location of their voting place, or being able to communicate with government members on a regular or daily basis.”

Cultural Awareness

A third theme identified in Table 1 is cultural awareness. This theme emerged from the data in statements by LIS students about cultural differences between themselves and Latinos. Cultural differences included practices LIS students identified as different from what they did (e.g., how Latino families used the library, attire of Latinos who used the library, and language and literacy needs of Latinos). Examples of cultural differences are discussed below. To preface the discussion, the term culture is defined.

Culture was defined as the daily activities that were evident in the lives of groups or organizations. Throughout the course, students were asked to explain what they thought culture meant and to provide examples of cultural differences. Through discussions in class and assigned readings, students who had initially equated culture with ethnicity began to separate these terms. Ethnicity was defined as the historical roots of groups of people from different geographic areas.

Examples of cultural differences were noted by LIS students in their fieldnotes. An entry noted that whole families came into the library together even though only one individual in the family used resources in the library. Several LIS students noted that families who had recently arrived from Mexico were not familiar with school or public libraries. They were not accustomed to checking out books from the library, and thought they needed to pay to check out books. Other cultural differences were noted. For example, an LIS student noted that “all [Latina] girls had ears pierced,” even baby girls, which she described as different from her own family’s practice. Another LIS student noted lack of direct eye contact, which caused some concern to the LIS student who stated, “The boy just wouldn’t speak up or look me in the eye, instead mumbling his request. I’m always disappointed when a patron walks away from a reference interview with such a downcast look . . .” Possible reasons for downcast eyes, other than misunderstanding, were discussed in class.

Three closely connected and overlap- ping aspects of the theme of Cultural Awareness, which were   prominent in the data, were language proficiency and literacy, and availability of Spanish language collections.

Language proficiency and literacy were identified as factors affecting access to information at school and public libraries. Lack of English language proficiency was noted as a factor affecting Latinos’ basic literacy in English including the ability to spell, read, and write. An LIS student noted that “the language divide was somewhat responsible for the digital divide.” Lack of English proficiency in adult Latinos was evident at school and public libraries, where adult Latinos who did not speak English used their children to translate written material for them. Children were also used as interpreters of conversations with English speaking librarians who did not speak Spanish. When public librarians were not able to communicate with Latino clients, their information requests could not be met and they were unable to access materials or information needed.

At school libraries, LIS students found similar literacy problems with students. For example, second graders had problems with alphabetical order, and few children were reading over grade level. An LIS student commented that perhaps Latino children who did not speak enough English to understand what was being taught “should be in a bilingual classroom.”

LIS students also commented on the effect of Spanish language use policies on access to information by parents at schools. For example, one elementary school used both English and Spanish during awards assemblies where parents were present. At another school a teacher translated for parents and instructions were provided in Spanish for parents. However, Spanish could not be used for instruction of Spanish speaking students. The misperception by some teachers and librarians at these schools was that using Spanish would delay children’ s acquisition of English.

At one school library students would not speak Spanish. “A student “…just wouldn’ t speak [Spanish] at   school because he thought he wasn’t supposed to because of the laws.” At the same time, other LIS students observed “Children spoke Spanish to each other sometimes and it seemed to make them more comfortable.” Children also seemed to feel more comfortable speaking Spanish with the Latino LIS student. “Being Latino . . . speaks to . . . the importance of having people with similar background [as the students] . . . especially within elementary schools and education.”

An LIS student noted that she had never worked with children who switched from one language to another. She commented, “Today, two girls playing a game would switch to Spanish for some words/phrases/sentences. I couldn’ t  discern a pattern    (they weren’ t  consistent in which words were spoken in which language), but they switched smoothly between English and Spanish without hesitation.”  The LIS  student’ s comment sparked a class discussion and the instructor used the opportunity to introduce research on first and second language acquisition and the phenomenon described by the student, which is called code-switching. Code-switching is a term used in linguistics to describe switching from one language to another language within the same sentence or during a conversation. It is a natural linguistic occurrence in areas where two or more languages are spoken, and requires a high level of linguistic competence in both languages contrary to the generally negative perception of code-switching held by many individuals. Some students admitted that they too had misconceptions about using Spanish and English together and had not considered code-switching an asset but a deficit in a person’ s    ability   to speak either language well.

At a public library, an LIS student who did not speak Spanish explained that he had gained a better understanding about what it felt like not to understand what others were talking about. He explained that he realized that “language barriers have a limiting effect on working with others [and] functioning productively.”

Spanish language library collections also emerged as a category, which included Spanish language books and other resources (e.g., newspapers) available to Latinos in school and public libraries. LIS students at public libraries noted adequate Spanish language materials in public libraries including books, magazines and newspapers. At one public library, older Latino men regularly used the public library to read newspapers. However, even though the public library had a large collection of Spanish language books, an LIS student wondered if some patrons could read at all. He said that for the first time in his life he dealt with semi-literate adults.

Access to library collections at schools was more complicated for Latino students. An LIS student commented that there were only three or four shelves of Spanish or bilingual books, despite a large Latino population. At another site with a large population of Spanish speaking students, only five percent of the collection was in Spanish. Also, there were only six chapter books in Spanish and these were targeted at girls.

Some school libraries had separate Spanish and English collections depending on the philosophy of the librarian or school. For example, at some schools, Spanish children’ s books were integrated into regular school library collections be- cause the librarian wanted Spanish language books to be more accessible, while at other schools Spanish language books were in a separate section. During a class discussion, an LIS student noted that when Spanish collections were separated from the regular collection, some children were embarrassed to check out At a public library, an LIS student who did not speak Spanish explained that he had gained a better understanding about what it felt like not to understand what others were talking about. He explained that he realized that “language barriers have a limiting effect on working with others [and] functioning productively.” Spanish language library collections also emerged as a category, which included Spanish language books and other resources (e.g., newspapers) available to Latinos in school and public libraries. LIS students at public libraries noted adequate Spanish language materials in public libraries including books, magazines and newspapers. At one public library, older Latino men regularly used the public library to read newspapers. However, even though the public library had a large collection of Spanish language books, an LIS student wondered if some patrons could read at all. He said that for the first time in his life he dealt with semi-literate adults.

Observations such as these were discussed in class, and misconceptions about language use in developing literacy were clarified during class discussions and course readings, which provided further information about language issues. An LIS student commented she thought that having Spanish language collections would be detrimental to developing English, but that “a real light bulb” had gone off during the discussion when she realized that “reading was reading in any language.”

Other issues discussed in this category included a wide range of issues that had presented themselves to LIS students during their service-learning. In one in- stance a parent would not allow her child to check out books in Spanish because she wanted her children to learn English. In another library, an LIS student noted that a first grade student checking out a book from the library  didn’ t know whether the book was in English or Spanish. Another LIS student commented that she often asked students to read the title of a book before checking it out and found that there were some students who couldn’ t speak English but   could read English. These situations were discussed in class and addressed in course readings.

Social Awareness, Civic Engagement and Social Capital

Civic involvement and social capital, which are considered essential elements of service-learning (Toole, 2002), emerged as themes and were labeled social awareness. LIS students felt they were helping to make a difference in the lives of the individuals they worked with in libraries whether in filling out job applications, or introducing children to literature. LIS students noted that they “developed empathy,” and that working with Latinos helped them develop a better understanding of their needs. Also, students noted that they became more sympathetic to those without access. Only one student stated that she “could see a lot of the problems . . . [but was] not particularly self motivated to go out and fix any of them.”

LIS students noted the need for librarians to be civic minded and contribute to underserved communities. These comments reflected an understanding of social capital discussed by Putnam (1995, 1996, 2000). Specific examples of students’ understanding of social capital included statements about providing clients with networking opportunities, building trust among community members and library patrons, and cooperating with individuals served (e.g., Latino children and adults).

Discussion

The service-learning experience appeared to provide LIS students with experiences that helped transform course content from theory to reality. Findings from pre- and post-surveys as well as other qualitative data provided evidence of changes in LIS students’ understanding of equity of access, and in students’ perceptions about diversity issues related to equity of access issues from the beginning of the semester to the end of the semester. Students indicated that the service-learning experience had helped them develop a deeper understanding of the type of access to information and technology Latinos had. LIS students also demonstrated a broader understanding of reasons for limited access (economic, language, lack of experience) and future needs of Latinos (having enough computers, time on computers, Spanish resources).

Students indicated they would “definitely recommend service-learning to other classmates” and that service-learning had been an ideal environment in which to experience first-hand what it will be like to be a librarian.” A student stated, “. . . I learned so much that I do not think I could have just ‘read’ about it.”

Service-learning was also valuable in providing first-hand knowledge of how disadvantaged Latinos were. A student commented,

“They needed library access to computers. The neighborhood was a bit scary in terms of crime and vandalism. The patrons were in lower income brackets. Lots of families who came together. Library used as safe place for children/teens after school. All ages used library.”

LIS students also gained experience with culturally different situations from service-learning as well as the fieldtrip. A student stated, “I have observed how people in other cultures see and experience things differently. I have more [of a] perspective about other perspectives.” Another student stated, “I had a “culture shock” visiting [the fieldtrip library]— that place was a palace in comparison with tons of technology, bright, well painted, and lots of books!”

Table 2 provides additional key experiences from service-learning during the course.

Table 2: Summary of Key Service-Learning Experiences.

Value Added

Finally, data indicate that service-learning indeed helped students gain an understanding of course content, which supports the growing body of literature on the benefit of service-learning (Billig, 2002; Billig & Eyler, 2003; Eyler & Giles, 1999; Eyler, Giles, Stenson, & Gray, 2001; Welch & Billig, 2004). As a student poignantly noted,

[We learned] what equity of access was. It’s not just a yes or no. It is the quality of access, the kind of access, how long you have it, how often you have it, where you have it, all those kinds of issues really play into what kind of access to information . . . technology that you have.

Students clearly became more aware of the community and all but one thought they had contributed to social capital of library and community. A student explained that she “thought it was helpful to talk about the inequalities in libraries— but see it played out in the two libraries [the fieldtrip library and my library] was priceless.” The fieldtrip clearly was an important part of the learning process that provided students with an example of equity of access from which they could compare their service-learning experience. A student summarized the experience by stating that he realized “there [was] a direct relationship between educational achievement and computer ownership . . .” and that it was a major factor in the digital divide. But other factors such as having a well-equipped library with WiFi, adequate space, enough functioning and current computers, adequate help at the desk, regular open library hours (not closed because of staff meetings, testing, special instruction, the library is serving as a backup classrooms) were essential to equal access to information.”

The service-learning experience appeared to support the notion of “value added” as introduced by Simons and Cleary (2005). Figure 1 illustrates two paths for teaching course content: a traditional path and a service-learning path.

The “value added” for the course is captured in a student’ s comment:

“[I]t gave a real-world example of how low funding affects disadvantaged populations” and “[o]bservation at the school reinforced what I was learning in class. I saw real life examples of the issues we were discussing as part of the course.”

Figure 1. Adapted from Simons and Cleary (2005) “Student and Community Perceptions of the ‘Value Added’ for Service-learning.” The diagram illustrates two possibilities related to student understanding of course content represented by arrows connected to the box on the far left. These arrows connect course content to the shaded area, which represents the learning en- vironment. The lower arrow illustrates course content without service-learning. The upper ar- row illustrates course content with the addition of service-learning, shown by the large box in the learning environment. The arrows on the right side of the box illustrate the benefits derived from service-learning. These include academic understanding (shown in the lower right hand circle), civic engagement and social capital, which come from working within communities (shown in the middle right hand circle), and increased knowledge of Latinos and greater aware- ness of the complexities of equity of access issues for Latinos (shown in the upper right hand circle). The added value that comes from the addition of service-learning to course content is the increase in student learning, illustrated by the three circles on the right.

Conclusions

This study adds to the body of knowledge about the effectiveness of service-learning to improve instruction for students (Bringle & Hatcher, 2005), about “facilitating cultural and racial understanding” (Eyler et al. 2001, para. 3), and about improving the depth of LIS students’ understanding of course content and its issues related to equity of access for Latino populations. The study also contributes to the growing body of information about service-learning in LIS studies (Roy, Jensen & Meyers, 2009).

Finally, this study supports findings of others that when service-learning is established as a vital and integral component of students’ education, students’ understanding of content is enhanced. This study provides evidence that service-learning does contribute to a better understanding of course related issues and leads students to a deeper level of examination of course readings. It is clear that students gained increased understanding of equity and cultural issues, which they will inevitably have to deal with as future library and information science professionals. Although not all students indicated that they would become more involved in civic endeavors in their community, an awareness of community issues among students was apparent.

For the LIS field, service-learning provides a bold new direction for instruction, which may better prepare students for their careers in librarianship by providing real-life experiences in settings with diverse populations where the themes identified in this study are the norm. Service-learning appears to be a strategy with tremendous potential for preparing future LIS professionals and should be considered as an essential component of the LIS curriculum.

Appendix A

Sample Questions for Presurvey:

Q1 What is service-learning?
Q2 What kinds of activities do you anticipate doing?
Q4 How do you think service-learning will help those you are going to work with?
Q5 How might service-learning help you understand or apply course content to “real life” experiences?
Q6 What experience do you have working with Latinos?

Sample Questions for Postsurvey:

Q1 Tell me about your service-learning experience?
Q7 What did you like least about your service-learning assignment?
Q15 Describe some of the needs you identified in the population where you did your service-learning.
Q16 Describe some of the resources you identified in the population where you did your service- learning.
Q23 Were there any changes in you as a service learner? If so, describe.
Q26 How can we improve the service-learning assignment?
Q28 Did service-learning give you a better understanding of the professional role of librarians? Explain.
Q29 Have your attitudes and perspectives toward diversity changed? If so, describe.
Q30 Do you feel more competent in relating to people with culturally different backgrounds, especially working with Latino students? Explain.
Q34 Has service-learning helped shape your perceptions of yourself and the community? If so, explain.
Q36 Do you feel more comfortable and competent in working with a population different from your own? Explain.
Q38 Explain if and how service-learning helped you understand course content.
Q39 Would you recommend your service-learning experience to any of your friends or classmates? Explain.
Q43 Describe what (if anything) you gained from participating in service-learning.
Q44 Do you feel more connected to the community compared to the beginning of the semester? If so, explain.

References

American Library Association. (1999). Libraries: An American value. Retrieved April from http://www.ala.org/ala/aboutala/offices/oif/ statementspols/americanvalue/librariesamerican. cfm

American Library Association. (2004). Library technical reports. Retrieved from http://www.ala.org/Template.cfm?Section= archive&template=/contentmanagement/ contentdisplay.cfm&ContentID=58395

American Library Association. (2007). Diversity counts. Office of Research & Statistics and Office of Diversity. Retrieved from http:// www.ala.org/ala/aboutala/offices/diversity/ diversitycounts/diversitycounts_rev0.pdf

Billig, S. (2002). Support for K–12 service-learning practice: A brief review of the research. Educational Horizons, 80(4), 184–189.

Billig, S. H., & Eyler, J. (Eds.). (2003). Deconstructing service-learning: Research exploring context, participation, and impacts. Greenwich, CN: Information Age Publishing.

Bransford, J., Brown, A., & Cocking, R. (Eds.). (1999). How people learn: Brain, mind, experience, and school. Washington, DC: National Academy Press.

Bringle, R., & Hatcher, J. (2005). Service-learning as scholarship: Why theory-based research is critical to service-learning. Acta Academica Supplementum 3, 24–44.

Bruner, J. S. (1966). Toward a theory of instruction (1st ed.). New York: W. W. Norton.

Cram, S. B. (1998). The impact of service-learning on moral development and self-esteem on community college ethics students. Dissertation University of Iowa. ED, 460-701.

Dewey, J. (1938/1963). Experience and education. New York: Macmillan.

Eyler, J., & Giles, D. E. (1999). Where’ s the Learning in Service-learning? San Francisco, CA: Jossey-Bass.

Eyler, J. S. (2000). What do we most need to know about the impact of service-learning on student learning? Michigan Journal of Community Service- learning, Special Issue, Research, 11–17.

Eyler, J. S., Giles, D. E., Stenson, C. M., & Gray, C. J. (2001). At A Glance: What We Know about The Effects of Service-learning on College Students, Faculty, Institutions and Communities, 1993-2000: Third Edition. Retrieved from http://servicelearning.org/filemanager/ download/4192_AtAGlance.pdf

Fairlie, R. W. (2005). Are we really a nation online? Ethnic and racial disparities in access to technology and their consequences. A report for the Leadership Conference on Civil Rights Education Fund. Retrieved from http://www.freepress.net/docs/lccrdigitaldivide. pdf

Fairlie, R., London, R., Pastor, M., & Rosner, R. (2006). Crossing the divide: Immigrant youth and digital disparity in California. Center for Justice, Tolerance & Community. Retrieved from http://cjtc.ucsc.edu/docs/digital.pdf

Furco, A. & Billig, S. H. (Eds.) (2020). Service-learning: The essence of the pedagogy. Greenwich, CN: Information Age Publishing.

Josey, J. E., & Abdullahi, I. (2002). Why diversity in American libraries. Library Management, 23(1/2), 10–16.

Kolb, D. A. (1984). Experiential learning: Experience as the source of learning and development. Englewood Cliffs, NJ: Prentice-Hall.

Kraft, R. J. (1996). An introduction to its theory, practice, and effects. Education and Urban Society, 28(2), 131–159.

Lincoln,Y.S.,& Guba, E.G. (1985).Naturalistic inquiry. Beverly Hills, CA: Sage.

Lynch, M. J. (1998). Racial and ethnic diversity among librarians: A status report. American Library Association. Retrieved from http://www.pla.org/ala/research/ librarystaffstats/diversity/racialethnic.cfm

Miles, M. B., & Huberman, A. M. (1994). Qualitative data analysis: An expanded sourcebook (2nd ed.). Thousand Oaks, CA: Sage.

Montiel Overall, P. (2009). Cultural competence: A conceptual framework for library and information science professionals. Library Quarterly, 79(2), 175–204.

National Service-Learning Clearinghouse (2001–2004). Retrieved from http://www.servicelearning.org/

Pew Hispanic Center. (2006). From 200 million to 300 million: The numbers behind population growth. Retrieved from http://pewhispanic.org/files/factsheets/25.pdf

Pritchard, I. A. (2002). Community service and service-learning in America: The state of the art. In Andrew Furco & Shelley H. Billig (Eds.), Service-learning: The essence of the pedagogy, pp. 3–21. Greenwich, CN: Information Age Publishing.

Putnam, R. D. (1995). Bowling alone: America’s declining social capital. Journal of Democracy, 6(1), 65–78.

Putnam, R. D. (1996). The strange disappearance of civic America. The American Prospect, 24(Winder). Retrieved from http://xroads.virginia.edu/~HYPER/detoc/assoc/ strange.html

Putnam, R. D. (2000). Bowling along: The collapse and revival of American community. New York: Simon & Schuster.

Rhoads, R. A. (1998). In the service of citizenship: A study of student involvement in community service. Journal of Higher Education, 69(3), 277–298.

Rosaldo, R. (1989). Culture and truth: A remaking of social analysis. Boston, MA: Beacon Press.

Roy, L., Jensen, K., Hershey Meyers, A. (2009). Service-learning: Linking library education and practice. Chicago: American Library Association.

Simons L., & Cleary, B. (2005). Student and community perceptions of the “value added” for service-learners. Journal of Experiential Education, 28(2), 164–188.

Stanton, T. K. (2000). Bringing reciprocity to service-learning research and practice. Michigan Journal of Community Service-learning Special Issue, 119–123.

Toole, J. C. (2002). Civil society, social trust and the implementation of service-learning. In Andrew Furco & Shelley H. Billig (Eds.), Service-learning: The essence of the pedagogy. Greenwich, CN: Information Age Publishing.

U. S. Department of Commerce. (2002). A Nation Online: How Americans are expanding their use of the Internet. National Telecommunications and Information Administration. Retrieved from http://www.ntia.doc.gov/ ntiahome/dn/nationonline_020502.htm

U. S. Department of Commerce. (2004). A Nation Online: Entering the broadband age. National Telecommunications and Information Administration. Retrieved from http://www.ntia.doc.gov/reports/anol/index.html

U. S. Census Bureau. (2006, May). Nation’s population one third minority. Retrieved from http://www.census.gov/Press-Release/ www/releases/archives/population/006808.html

Valerius, L., & Hamilton, M. L. (2001). The community classroom: serving to learn and learning to serve. College Student Journal, 35(3), 339–344.

Vygotsky, L. (1978). Mind in society: The development of higher psychological processes. Eds. Michael Cole, Vera John-Steiner, and Ellen Souberman. Cambridge: Harvard University Press.

Weah, W., Simmons, V. C., & Hall, M. (2000). Service-learning and multicultural/multiethnic perspectives. Phi Delta Kappan, 81, 673–675.

Welch, M., & Billig, S. H. (Eds.). (2004). New perspectives in service-learning: Research to advance the field. Greenwich, CN: Information Age Publishing.

Keywords: information retrieval, education, visualization, evaluation, user study

Introduction

Interactive visualization is a powerful educational tool. Visualization can provide a clear visual metaphor for un­derstanding complicated concepts and uncovering the dynamics of important processes that are usually hidden from the student’s eye (Gordin & Pea, 1995). Visualization has been used to enhance the teaching of various subjects ranging from chemistry (Evans, Yaron, & Leinhardt, 2008) to biology (McClean, et al., 2005) to physics (Perkins, et al., 2006). Computer science is one of the most active application areas for educa­tional visualization research (Naps, et al., 2003). In computer and information science (CIS) education, visualization is used almost exclusively in program­ming and data structure courses. We can name dozens of papers devoted to visu­alization of program execution on sev­eral levels from machine-level languages (Butler & Brockman, 2001) to high-level languages (Domingue & Mulholland, 1998; Haajanen, et al., 1997; Levy, Ben-Ari, & Uronen, 2003; Tung, 1998) to algorithms and data structures (Hundhausen, Douglas, & Stasko, 2002; Rößling, Schüer, & Freisleben, 2000). Our claim is that In­formation Science courses could benefit from this powerful technology.
This paper presents our research on us­ing interactive visualization in the non-traditional context of information re­trieval courses. Information retrieval has been in the curriculum of many computer, information, and library science depart­ments for more than 30 years. With the maturity of the World Wide Web, infor­mation retrieval became an important practical subject. Elements of information retrieval are now taught to students of many different specialties. We think that information retrieval provides an interest­ing and important application area for ex­ploring the power of interactive visualization. Over the last six years we developed and made publicly available a range of Web-based interactive visualiza­tion tools supporting various concepts taught in information retrieval courses. These tools were used for several years by faculty members at the University of Pitts­burgh to teach both graduate and under­graduate information retrieval courses. To evaluate the impact of our visualization tools and to solicit student feedback, we ran several classroom studies. The devel­oped set of tools is available on the project home page (http://ir.exp.sis.pitt.edu/res2/ resources.php). The tools, which are run­ning on our servers, could be used by any­one interested in teaching or learning information retrieval. This paper summa­rizes the results of our multi-year effort. To demonstrate our ideas, we present two examples of these interactive Web-based visualization tools for information re­trieval. We also report the results of class­room studies evaluating them. To conclude, we discuss the results and pros­pects for using interactive visualization in the context of information retrieval courses.

Interactive Visualization for Information Retrieval

One of the secrets of the power of inter­active visualization is its ability to un­cover and present in detail processes that are typically hidden from students’ eyes. Being interactive, visualization pro­grams allow the students to explore these processes step-by-step, with different pa­rameters, gaining as a result a deep under­standing of the processes and the con­cepts behind them. Every field has its critical processes and concepts, which could be better understood with the use of visualization.
The core of a traditional information retrieval (IR) course is a set of models, al­gorithms and technologies for process­ing, storing and retrieving textual information. Traditional presentation of this core usually starts with several IR models (such as the Boolean, vector, and probabilistic models and several varia­tions of them) and then follows by ex­plaining how the information is organized and retrieved in each of these models (Baeza-Yates & Ribeiro-Neto, 1999; Korfhage, 1997). The process by which information is retrieved in differ­ent models is one of the hardest topics in an IR course for students to grasp, despite being formalized and well understood by the IR research community. We have ob­served that even Boolean information re­trieval, the simplest of the models, is difficult for many students. At the same time, traditional educational tools—re­search or commercial IR systems—offer little educational help. The process of re­trieving information has several steps, from entering the query to matching the query to the documents to prioritizing the results. In an IR system (even an educa­tionally-oriented one) all these steps are hidden from a user: the only thing that a user can observe is the final results—a list of ordered documents. This is exactly a context that could benefit greatly from the use of interactive visualization and this realization formed the starting point for our research (Brusilovsky, 2002). Over the first years of our project we de­veloped and explored interactive visual­ization tools to visualize the process of retrieving information in several known models: Boolean, fuzzy, vector, and ex­tended Boolean (see Baeza-Yates & Ribeiro-Neto, 1999 and Korfhage, 1997 for the description of these classic mod­els). Since these visualization programs were the first to developed, we had more opportunities to improve them over the years and to evaluate them in the classroom. For the purpose of this paper we chose two of these model visualization tools to demonstrate the ideas of interactive visualization in the context of IR courses. The following subsections present the most recent versions of interactive visualization programs for the Boolean and vector IR models. The next section reports the results of classroom studies with these models.

Interactive Visualization of the Boolean Information Retrieval Model

The Boolean IR model is the oldest and the simplest of the IR models. In this model, a query is formed by a set of ele­mentary queries (usually keywords)con­nected by Boolean operators such as AND, OR, and NOT. The mechanism of this model is set theoretical. Every query is associated with a set of matching docu­ments. For an elementary query such as a keyword, the set of matching documents is simply all documents indexed by this keyword. To obtain the set of matching documents for two queries connected by a Boolean operator, one has simply to per­form the corresponding set operation on their matching sets (i.e., set intersection for AND, complement for NOT, etc.). Thus in several steps, a matching set for any complex Boolean query can be found.
While it all sounds quite simple and clear, we have found that many of our stu­dents have problems understanding how Boolean matching works. Our talks with students have indicated that one of the sources of their troubles is the failure to perceive Boolean operators as operations on sets of matching documents. This is a known problem for everyday users who confuse the informal use of AND, OR, and NOT in everyday language with their formal meaning in set theory and Boolean logic. Surprisingly, we also dis­covered that some students with good programming backgrounds (i.e., those who have routinely used Boolean opera­tors for writing conditional expressions in their programs) still have problems transferring their knowledge of these operators to the set theory context.
In developing an interactive visualiza­tion environment for the Boolean IR model we were trying to achieve two goals: to provide a helpful visual meta­phor and to visualize the process of Boolean matching step by step. Figure 1 presents an interface for our environ­ment. The core of this interface is a set of all documents visualized in a table (one document per row). For a sample document set in our system we choose textbook surrogates since this is the type of document most familiar to information science students. The goal of this visual representation is to help students to understand the core principle of this model—every query is associated with a particular subset of all documents. Showing the set of all documents on the screen makes it easy to demonstrate different subsets of the whole set as sets of differently colored rows of the table.
The students explore Boolean matching by writing and executing simple Boolean queries: pairs of elementary queries (terms) connected by a single Boolean operator (OR, AND), or a pair of operators (AND NOT), which simulate set difference. The results of an executed query are shown through row and cell coloring in the document table. The final set of documents is visible as a set of green rows (documents 2, 8, 12 in Figure 1). This is what a standard Boolean search engine would return. The visualization tool, however, attempts to show more by decomposing the standard “black box” matching process into two parts: (1) the process of matching an elementary query to the set of the documents and (2) the process of obtaining a new set from contributing sets by applying different Boolean operators. To achieve this goal, the tool extends the document table with three colored columns of cells. The first and the second column show the results of matching each of the documents to the first and the second query terms correspondingly. Documents matched to each term are marked as true in the corresponding column. To make the set of matched documents more visible, documents matched to the first term are marked by a red cell background and documents matched to the second term are marked with a blue background. The third column shows the results of matching the whole query using truth values and a green color for matching documents. The color-coding makes the two steps of the Boolean query matching process more transparent. For example, the student can see that only documents 2, 8, and 12, which matched to both elementary terms, are included in the final set. In contrast, documents 4, 10, and 11, which matched only to the first term and document 7, whichmatched only to the second term, are not included (see Figure 1).

Figure 1. BooleanModel Environment. Visualization of matching for a simple Boolean AND query. Documents matching the first elementary query are highlighted.

Beyond the term-based search shown on Figure 1, the Boolean IR visualization tool has several other functionalities. In particular, to help the student transfer the understanding of Boolean IR from classic IR to the database context, we have provided a very similar exploration interface where elementary queries are constructed not from keywords as in classic IR but from restrictions on various fields of a database record (i.e., year = 2000 and publisher ! = “O’Reilly”). Technically, the tool is implemented as a Java servlet working on a dedicated server.

Interactive Visualization of Vector Information Retrieval Model

The vector IR model is different from the Boolean model in many aspects. While the query in this model is also a set of terms(keywords), the terms can be weighted, stressing their relative importance. To perform the matching process, the query is converted into a weighted vector of terms. Similarly, all documents are represented as weighted vectors of terms. The weight of a specific term in a document vector represents the collection-adjusted importance of this term in the document content. The matching process is based on vector algebra. The goal of vector matching is to produce a relevance value for each document, which reflects how similar this document is to the query. The relevance value is produced by calculating the Euclidean or angular distance between the query and document vectors. Once relevance values are calculated, all documents can be ordered by their relevance to the query. Vector matching produces a ranked list, not just a subset of documents as Boolean matching does.

Figure 2. A sample document used by the vector model visualization tool.
On Figure 4 this document is marked as D2.

The hardest component of the vector matching process to understand is the calculation of the relevance value.While the geometrical nature of the relevance measure (Euclidean or angular distance) is relatively straightforward, the actual relevance values are produced by performing many operations with components of very large vectors. Not surprisingly, the final results of these calculations are much less evident to students than the results of Boolean operations with document sets. As we observed in the classroom, it is frequently hard for students to understand why a specific document was ranked particularly high or low in the resulting list.
As in the case of the Boolean visualization tool, the goal of the vector matching visualization was to uncover the steps of the matching process, which are hidden from the users of vector-based search systems. While these systems never go further than displaying the final relevance value of retrieved documents, we wanted to show how this relevance is calculated in the vector model. To uncover this process, our tool offers a small collection of documents formed from just 7 different terms (so that query and document vectors are very short). Figure 2 shows a sample document in this collection. The students explore vector matching by writing and executing simple vector queries, which are weighed sets of terms. A new query can be formed step by step by adding one term (selected from the same set of 7 terms) with its weight at a time. The example in Figure 3 shows a query formed by two terms: term Dog with weight 3 and term Fox with weight 1. For simplicity we use integer weights when forming a query; however, after the query is formed, its vector is normalized.

Figure 3. The process of forming a query in the vector model visualization tool.

The results of the query execution are shown as two tables (on the right side of Figure 4): one table presents ranking based on Euclidean distance and the second presents ranking based on the cosine measure. The more relevant a document is to the query, the smaller the distance measure and the larger the cosine measure. To dig deeper, a student can click on any relevance value (shown in blue) and see in a popup window (bottom of Figure 4) how this value is calculated from the components of the query and document vectors. (The example in Figure 4 uses Euclidean distance.) The vectors for the query and documents are shown on the left alongside the ranking results, so the students can see how the vector components are used in the calculations. By clicking on a document number (shown in blue), students can view the content of each document in a popup window. Thus the visualization tool makes all the traditionally hidden steps of the vector matching process visible: from documents and query to their vectors, from vectors to relevance values, and from relevance values to ranked lists.
In addition to the interactive simulation presented above, the tool also offers a brief tutorial on vector matching. It is implemented using a combination of Java servlets and client-side Javascript programs.

Figure 4. The results of query execution produced by the vector model visualization tool. The popup window at the bottom shows how the Euclidean distance value of 0.71 was produced for document D2 given the document and query vectors shown above.

Classroom Evaluation of Interactive Visualization

Study Design

To examine whether our visualization tools for teaching information retrieval are effective in a real educational process, we conducted several studies with students of information retrieval courses. In each of these studies we evaluated several visualization tools. This section reports results of five classroom studies performed in the context of a graduate course, “Information Storage and Retrieval,” offered annually at the School of Information Sciences, University of Pittsburgh. These studies were performed between Fall 2001 and Fall 2005. The Boolean and vector model visualization tools, which are the focus of this paper, were evaluated in each of the five studies. Table 1 shows the number of students involved. During 5 semesters, a total of 95 students completed the study, with a mean of 19 students per class (SD = 7.58).

Table 1: Classroom Studies of Boolean and Vector Model Visualization Tools.

All studies of our visualization tools have the same multi-stage design presented in Figure 5. The goal of this design was to assess both objectively and subjectively whether the visualization tools are actually working as useful aids to help the students better understand the underlying concepts. As an objective measure, we used knowledge gain from pre-test to post-test. As a subjective measure, we used student answers to a questionnaire about the systems and their features. The studies of different tools were performed during different weeks of the semester. In a given week, the students learned the concepts of the subject (i.e., Boolean or vector model) during a lecture and the instructor introduced the corresponding visualization tool, encouraging them to utilize it for their deeper understanding of the knowledge they acquired in the classroom.

Figure 5. The classroom study procedure.

At the end of the lectures devoted to the Boolean and vector space models, the students took a pre-test and were given homework assignments which, among other things, specifically required them to use the tools introduced in the lecture for actively solving hands-on problems. The role of this homework was to engage the students in active exploration of the visualization tools in the context of problem solving. Figure 6 shows an example of the homework on the vector space model.

Figure 6. Homework assignment example.

To evaluate the results of their learning, the students took a post-test before the next lecture (exactly one week from the first lecture on the topic), and filled in a questionnaire about the tools and their impact. The post-test was identical to the pre-test. The questionnaire included 14 questions, seven for each tool (Table 2). A five point Likert scale was used for the questions ranging from 1 (Strongly Agree) to 5 (Strongly Disagree).

Table 2: Survey Questions for the Subjective Feedback Analysis.

The role of the questionnaire was to evaluate student attitudes to the visualization tools and their features. However, the questionnaire alone was not sufficient for a reliable evaluation. To make sure that the student’s positive (or negative) attitude is grounded in the student learning experience, we had to check whether student knowledge increased after using this tool, and examine the correlation between knowledge gain and attitude. Knowledge increase (positive knowledge gain) after using the tool can support the student’s positive attitude and provide additional evidence in favor of the educational effectiveness of the tool. In contrast, the lack of knowledge increase (or knowledge decrease) would cause us to question the educational value of the tools even in the presence of positive feedback. Note that the format of a classroom study does not allow us to use knowledge gain alone as a reliable indicator of the effectiveness of the tools. While we attempted to minimize student learning from other sources by placing the pre-test after the lecture presentation of the corresponding topic, we were not able to prevent students from using other sources of knowledge during their work on the homework assignment. Only a controlled lab study could assure that the registered knowledge gain was the result of student work with the tools. Given that our main goal was to explore the value of the tools in a real educational process, we accepted the shortcomings of the knowledge gain measure and applied it as a secondary indicator.
To calculate knowledge gain, which measures an increase in the student’s knowledge of the topic, we used pre- and post test scores [Equation (1)]. Because of the minor differences in the number of pre/post-test questions over the study semesters, we used normalized pre- and post-test scores (both were normalized to range from 0 to 10) to calculate the knowledge gain. This allowed us to evaluate the effect of the tools over all five semesters.

Knowledge Gain = Normalized Post Test Score – Normalized Pre Test Score (1)

With two evaluation measures engaged, we can formulate two formal hypotheses to assess whether our Web-based interactive visualization tools are educationally effective.

H1: The student will feel that the visualization tools are helpful in understanding crucial concepts of information retrieval.

More specifically,

H1-1: The subjects will answer positively to the questions asking about the effectiveness of the visualization tools.

H2: Student knowledge will increase after their work with the tools.

That is,

H2-1: The knowledge gain calculated as a difference between post- and pre-test scores will be positive.

The following subsections report the results of the data analysis performed to evaluate these two hypotheses for each of the visualization tools.

Subjective Feedback Analysis

This section analyzes students’ subjective feedback on the visualization tools. Seven questions per model were asked each semester (some semesters also included one additional question). Students’ responses indicated a value from 1 to 5, where 1 is the most positive answer to each question. The distribution of student answers is shown in Tables 3 and 4 and Figures 7 and 8.

Table 3: User Feedback on the Boolean Model Tool.

Table 4: User Feedback on the Vector Space Model Tool.

Figure 7. User feedback for the the Boolean model visualization tool.

Figure 8. User feedback for the vector space visualization tool.

Q1 differs from the rest of the questions since it is focused on the students’ prior knowledge about the corresponding information retrieval model. As the data shows, about 80% of the subjects agreed or strongly agreed that they understood the Boolean model. In contrast, only 23% agreed or strongly agreed that they understood the vector model. This result is consistent with the objective analysis in the previous section, where the average pre-test score of the Boolean model was 7.87 out of 10.0whereas that of the vector model was only 4.58.
The next two questions (Q2 and Q3) assessed the usefulness of the visual and interactive nature of the tools in the classroom. About 80% to 90% of the students gave positive answers (Strongly Agree or Agree) for both the Boolean and the vector models. Almost no negative answers were submitted. Moreover, the interactive nature of the system was valued even higher than its visual nature. This is strong evidence in favor of interactive visualization.
Question 4 asked how easy itwas to understand the interface of the visualization tools. Overall, the students showed a high level of satisfactionwith theBoolean tool interface (83%) and slightly lower satisfaction with the interface of the vector model tool (72%). This indicates that the latter interface may need further improvement.
Questions 5 to 7 asked in different words about the importance and usefulness of the tools in the context of an information retrieval course. Students’ answers confirmed that the vector model tool, which helped them with a less well known topic, wasmost valuable for them. Over 80% of the students agreed that the vector tool should become one of the key course tools and over 90% wanted to recommend the system to students taking the course next semester. Corresponding numbers for the Boolean tool (over 60% and over 80%) are lower stressing that this tool was seen as less valuable. This may be caused by the simplicity of the Boolean model, which was relatively well understood by most the students before the start of the course. Yet, the feedback for both tools is overwhelmingly positive: both tools were considered important and useful by a clear majority of students. Moreover, about 90% of the students agreed or strongly agreed that both tools should be used in teaching information retrieval courses.

Table 5: A Comparison of Test Scores for the Two Models.

Knowledge Gain Analysis

As described earlier, we measured the pre- and post-test scores for two educational visualization tools for information retrieval: one for the Boolean and one for the vector model. Table shows the basic statistics comparing the scores. We can first observe that there are increases between pre- and post-test scores for both models. Test scores for the Boolean model increased by 0.59 and scores for vector model increased by 1.80. These differences were statistically significant (paired t-test) for both the Boolean and vector model tools (p = 0.03 and p < 0.01 respectively). It means there was actual knowledge gain after the students used the visualization tools.
It is interesting to observe that the starting knowledge of the Boolean model was relatively high. In this context it was a challenge for the tools to further increase the student knowledge. Indeed, the increase was quite moderate, although statistically significant. In contrast, the starting knowledge of vector model was relatively low—less than 50%. In this context, the visualization tools helped to achieve a more impressive knowledge gain, almost a 40% increase from the level of knowledge reported by the pre-test. This allows us to hypothesize that interactive visualization tools could be especially useful for teaching hard-to-understand topics, where regular educationalmeansmay be less effective.
It was also interesting that the standard deviation of pre-test scoreswere just 2.38 and 2.23 for the Boolean and vector model respectively. These values are relatively small, suggesting the starting levels of student knowledge within each topic were quite similar. Since this data was collected over five semesters of teaching the course, it could be used as a reliable estimation of student starting knowledge when preparing educational visualizations and other pedagogical material for the course.

Subjective Feedback versus Knowledge Gain

In the previous sections, we saw that there was an increase in students’ knowledge after they took the information retrieval classes, and that they supported the use of visualization tools in the classes. That is, there was an objective effect on their knowledge and the source of the effect was assumed to be the visualization tools, based on their positive response to the questions asking them about the usefulness of the tools. To confirm this assumption, we examined the relationship between the objective knowledge gain and the subjective answer from the students. We chose questions #2 and #3 from the questionnaires, which directly asked about the usefulness of the visual (Q#2) and interactive nature (Q #3) of the tools. The answers to the questions were compared with student knowledge gain scores for the Boolean and the vector model separately. Figures 9 and 10 show the results of these comparisons.

Figure 9. Comparison of subjective user feedback against knowledge gain for Boolean tools.

Figure 10. Comparison of subjective user feedback against knowledge gain for vector tools.

Here, three out of four graphs show a positive relationship between the subjective answers of the students and their actual knowledge gain. The students with higher knowledge gain scores tend to agree more with the statement on the usefulness of the visual nature of the Boolean tool (Figure 9 left). For the statement on the interactive nature of the Boolean tools, the effect is less pronounced and reversed (Figure 9 right). In terms of the vector visualization tools, the students with higher knowledge gain scores tend to be more positive about the visual and the interactive nature of the tools, although the effect is weaker than in the previous case (see Figure 10). Overall, it can be observed that students who learned more (higher knowledge gain) reacted more positively to the visualization tools for both models—Boolean and vector.

Discussion

This paper presents our experience in developing and using a set of interactive visualization tools for teaching and learning information retrieval. By using the power of interactive visualization, our tools allow a teacher to introduce important models of information retrieval in a way that cannot be replicated by traditional whiteboard and slides, and with almost no preparation time (simply by deciding which examples to show to cover the main set of ideas). The instructor can easily accommodate very different audiences by adjusting the number of examples to show, the speed, and the granularity of presentation.
Using these interactive visualizations as learning tools further increases their value. They let the students switch from passive learning-by-reading to active and interactive exploratory learning. By exploring a number of different examples with interactive visualization tools they get a chance to achieve a better understanding of complex IR topics. The results of our classroom studies indicate that use of interactive visualization in a homework context can result in significant growth of knowledge. The vast majority of the students recognize the value of interactive visualization and recommend its use in the context of information retrieval courses. The study also demonstrated that visualization focusing on less known and harder to understand topics causes a larger growth in knowledge and is perceived as more useful. This result suggests placing higher priority on the development of visualization tools for harder to understand topics.
Our results stressed the importance of developing, collecting and sharing tools for teaching information retrieval. This is consistent with the culture of sharing in IR research. A number of groups maintain Web pages of IR resources such as search software, text processing utilities, evaluation packages and topical bibliographies. We argue that similar collections of IR educational resources should be established and maintained. The Web makes it possible for researchers and educators to make their contributions to research and teaching available to others, making it much easier for others to implement. We have contributed to this process by developing a range of interactive visualization tools for teaching and learning information retrieval. We have demonstrated that these visualization tools have value for students learning the concepts behind specific information retrieval models. We have made these freely available to any IR educators or studentswishing to use them on our IR resources page (http://ir.exp.sis.pitt.edu/ res2/resources.php) along with similar educational resources developed by other teams. We welcome readers who wish to try these tools and hope that more educational repositories like ours will be established in the near future.
In our own future work we plan to continue exploration of interactive visualization for information retrieval. We hope to increase the number of visualization systems available for use in information retrieval courses and to perform additional formal studies of these tools.

Acknowledgements

Most of the tools in the original suite of learning environments were developed under the guidance of the first author by MSIS students of the School of Information Sciences at the University of Pittsburgh. Our School is fortunate to have many bright students and the author thanks them for their devotion and creativity. Special thanks go to Olena Scherbinina, Archana Sulebele, and Anand Banvasi, the developers of Boolean and vector model visualization tools presented in this paper.

References

Baeza-Yates, R., & Ribeiro-Neto, B. (1999). Modern information retrieval. NewYork, NY:ACMPress.

Brusilovsky, P. (2002, June 24–29, 2002). Web-based interactive visualization in an information retrieval course. Paper presented at the ED-MEDIA’2002—World Conference on Educational Multimedia, Hypermedia and Telecommunications, Denver, CO.

Butler, J. E., & Brockman, J. B. (2001). A Web-based learning tool that simulates a simple computer architecture. SIGCSE Bulletin—Inroads, 33(2), 47–50.

Domingue, J.,&Mulholland, P. (1998). An effective Web based software visualization learning environment. Journal of Visual Languages and Computing, 9(5), 485–508.

Evans, K., Yaron, D.,&Leinhardt, G. (2008). Learning stoichiometry: A comparison of text and multimedia formats. Chemistry Education: Research and Practice, 9, 208–218.

Gordin, D. N.,&Pea, R. D. (1995). Prospects for scientific visualization as an educational technology. Journal of the Learning Sciences, 4(3), 249–279.

Haajanen, J., Pesonius, M., Sutinen, E., Tarhio, J., Teräsvirta, T., & Vanninen, P. (1997). Animation of user algorithms on the Web. Paper presented at the VL ‘97, IEEE Symposium on Visual Languages.

Hundhausen, C. D., Douglas, S. A., & Stasko, J. T. (2002). A meta-study of algorithm visualization effectiveness. Journal of Visual Languages and Computing, 13(3), 259–290.

Korfhage, R. R. (1997). Information storage and retrieval.New York, NY: Wiley.

Levy, R. B.-B., Ben-Ari, M., & Uronen, P. A. (2003). The Jeliot 2000 program animation system. Computers and Education, 40(1), 1–15.

McClean, P., Johnson, C., Rogers, R., Daniels, L., Reber, J., Slator, B. M., et al. (2005). Molecular and cellular biology animations: Development and impact on student learning. Cell Biology Education, 4(2), 169–179.

Naps, T., Rößling, G., Anderson, J., Cooper, S., Dann, W., Fleischer, R., et al. (2003). Evaluating the educational impact of visualization. ACM SIGCSE Bulletin, 35(4), 124–136.

Perkins, K., Adams, W., Dubson, M., Finkelstein, N., Reid, S., Wieman, C., et al. (2006). PhET: Interactive
simulations for teaching and learning physics. The Physics Teacher, 44(1), 18–23.

Rößling, G., Schüer, M., & Freisleben, B. (2000). The ANIMAL algorithm animation tool. Paper presented at the 5th annual SIGCSE/SIGCUE ITiCSEconference on Innovation and technology in computer science education, Helsinki, Finland.

Tung, S.-H. S. (1998). Visualizing Evaluation in Scheme. LISP and Symbolic Computation, 10(3), 201–222.

Overall, the involvement of people of all colors at the doctoral level can be fairly characterized as minimal. The need for immediate response is acute (Turock, 2003, p. 493).

Doctoral fellows serve as the nucleus of energy for continued recruitment of a diverse doctoral student population. Attrition through graduation will extend the diversity to the LIS professoriate. Future generations of librarians are educated by the professoriate (Bonnici & Burnett, 2005, p. 125).

As is demonstrated in the larger field of librarianship, there is a serious dearth of minority scholars in the Library and Information Science (LIS) professoriate (Davis & Hall, 2006). The Spectrum Doctoral Fellowship Program is well on its way to producing 12 such candidates. This is an admirable start, but more needs to be done and more minority candidates need to be recruited, promoted, and funded through PhD programs.

During her tenure as the President of the American Library Association (ALA) (1995), Dr. Betty J. Turock, in collaboration with then ALA Executive Director Elizabeth Martinez, spurred the creation of the Spectrum Scholarship Initiative, which was designed to recruit and fund members of underrepresented minority populations through graduate programs in library science. Believing that the country was rapidly changing, Turock felt that the field of librarianship should be changing as well – libraries can only be responsive to their diverse clientele if their staffs are equally diverse. Minority populations in the United States are quickly becoming the majority, and librarianship, long known for being a primarily Caucasian and female field, has not done a good job of being reflective of and responsive to these changes. Turock credits friend and mentor E J Josey with the impetus for the Spectrum Initiative; expressing frustration and “disgust” for the lack of diversity in the library profession, Josey stated that ALA only recruits one minority librarian per year and thinks that’s progressive (personal correspondence with Dr. Turock, May 7, 2009). Determined to change this trend, Turock decided that ALA should recruit at least 50 minority librarians per year. And so Spectrum began.

Turock’s initiative was not a welcome one, and if not for her tireless effort, the Spectrum Initiative would not have come to fruition, as she faced significant resistance from the field and from members of ALA, whose support was needed to pass the initiative. Some library constituents were not interested in promoting Spectrum and its goals; this was perhaps more of a negative response to the discussion of diversity the initiative would generate, and not as much a response to the initiative itself. As Sandra Rios Balderamma (2000), the first director of ALA’s Office for Diversity, discusses, diversity inspires different reactions in different people, and instead of having difficult and revealing conversations, it is easier to stifle and ignore new ideas and initiatives, such as Spectrum.

Sometimes the definitions and visualizations are sharp and explicit: racism, white privilege, homophobia, heterosexual privilege, inequity of access, institutional racism, organizational barriers, apologies and reparation, “illegal” aliens, non-English speaking, non-white, non-user, old boys’ network, and old girls’ network. Sometimes the definitions and visualizations are easier on the senses and perhaps more elusive: celebration of difference, internationalism, intellectual diversity, global village, multiculturalism, organizational cultures, pluralism, diversity of work style, and diversity of learning styles. At times the term is simply empty and unfulfilling and has not earned its credibility (p. 195).

As long time LIS educator and University of Buffalo faculty Lorna Peterson discussed in her 1999 article, issues of diversity are often perceived as threatening by the majority in the field (in this case, White females and some males), and are conflated with issues of race and racism. All related issues, they tend to inspire passionate responses and feelings of exclusion. Peterson states,

If diversity were as non-threatening a concept as the rhetoric of difference would lead us to believe, then expressions of anger would not occur when practices to readdress past discrimination are enacted. “No one helped me to get here!” is their cry, but they don’t recognize that there was no barrier either. Expressed resentment means diversity may be about achieving equity (which means loss of privilege for some), but the scant evidence of progress may mean that diversity is not about equity at all. (p. 18)

Turock’s efforts were about “advancing social justice and human rights within organizations and the profession” and wanting minority librarians, and the populations they serve to not only survive, but thrive (personal correspondence, May 7, 2009). About Spectrum’s long journey to being, Elizabeth Martinez remembers,

It was a grand idea that we developed when I was Executive Director of ALA. At the time, I was frustrated that, after hearing for 20 years how much diversity was a priority for ALA and the profession, there still were no national scholarships for librarians of color. The ALA Council struggled with accepting the proposal, and past president Betty Turock shamed them to vote yes. It was later embraced and supported by library schools and the profession, and I am grateful that there are over 600 graduates. Today it is the largest and most prestigious ALA scholarship. (personal correspondence, May 20, 2009)

Begun with seed money from ALA, donations from every ALA division and personal donations, ALA’s Office for Diversity was created and the Spectrum Initiative launched, and now boasts a formidable number of alumni scholars who have gone through LIS masters programs and now work in the field. Ironically and sadly, despite having made a significant impact on the library profession (Whitwell, 1998; Roy, et. al, 2006; Kenney, 2005), and having benefited over 600 masters level scholars, and now 12 doctoral students, there is still resistance within in the field to support Spectrum and its work. In 2010 Turock launched a new campaign to raise $1 million dollars to continue Spectrum’s mission (to which she personally donated $100,000); when addressing ALA council members at a January 2010 meeting to announce the new initiative, she was met with lukewarm response. “It’s not clear that all Councilors will follow her example. After Councilor-at-large Pat Wand suggested that each Councilor make a donation, the applause was fairly weak” (Oder, 2010, para. 12). Despite these obstacles, the Spectrum Initiative remains “ ALA’s gift to the library education” and Turock’s bold promise to the profession continues to recruit professionals “who will act as bridges” between the understanding of cultures and knowledge and society at large (personal correspondence, May 7, 2009).

New bridges were initiated with the implementation of the Spectrum Doctoral Fellowship program in 2007. In 2007 and 2008, through the benefit of an IMLS grant, the University of Pittsburg and ALA’s Office for Diversity provided full fellowships for 12 Spectrum Doctoral Fellows to pursue advanced Library and Information Science degrees at accredited institutions around the country. The 12 Fellows represent the four underrepresented ethnic populations and are in various stages of study at Rutgers University, the University of Wisconsin-Madison, the University of Pittsburgh, Simmons College, UCLA, and the University of Tennessee at Knoxville. The ultimate goal of this program is to increase racial and ethnic diversity among the profession’s next generation of LIS faculty and leaders.

The dearth and attrition of minority PhD students has long been a topic of discussion in the higher education literature, all disciplines face this challenge (Meacham, 2002; Manzo, 1994; Pruitt & Isaac, 1985). And this issue is even more acute in library and information science (Brown-Syed, et. al, 2008; Franklin & Jaeger, 2007; Reeling, 1992). A derivative issue of diversity recruitment and retention in LIS education (typically addressed in relation to masters level degrees) (Jaeger, et. al, 2010; Dewey & Keally, 2008; Stringer-Stanback, 2008; Winston, 2008; Neely & Peterson, 2007; Barlow & Aversa, 2006; Honma, 2005; Wheeler, 2005; Alire, 2001; Gollop, 1999), minority recruitment and retention into PhD programs deserves fresh and dedicated attention and study. The latest statistics available from the Association for Library and Information Science Educators (ALISE) indicate that the LIS professoriate continues to lack diversity:

Only 3.7% of the fulltime faculty members are Latino, as compared to 14.5% of the total population, while African Americans comprise just 5.5% of the fulltime faculty as compared to 12.1% of the population. In 2002 – 2003, of the 82 LIS doctoral degrees awarded, only two were awarded to African Americans and one to a Latino (Sineath, 2005). As a result, the faculty population in LIS has remained fairly stable in its level of diversity, with the percentage of African Americans and Latinos in LIS faculties changing little since the passage of the Civil Rights Act in the 1960s (Sineath, 2005, as quoted in Jaeger & Franklin, 2007, p. 21).

These rates parallel the rates of minority librarians working as practitioners in libraries. Jaeger and Franklin (2007) propose that a cycle be strengthened and perpetuated; the rationale is that increased numbers of minorities in the LIS professoriate will shape and transform LIS graduate curricula and programs, which in turn will impact and inform the next generations of minority librarians, who will then adequately and appropriately serve the diverse communities that patronize libraries. And hopefully, these minority librarians will model and inspire up-and-coming students to pursue librarianship as a career. This “virtuous” cycle underscores the dire importance of recruiting and retaining minority LIS PhD students and emphasizes the need for librarianship (at all levels) to be representative of the communities being served.

In the circular, self cycling style of education and librarianship, minority school and public librarians who serve as role models for minority children may inspire the children to go to college. In college, minority academic librarians and library school faculty may inspire them to go to graduate school to become librarians and role models themselves.” (Totten, as quoted in Jaeger and Franklin, 2007, p. 23)

A direct response to Dr. Betty Turock’s call for action, to address the “acute need” for minority LIS PhD students (Turock, 2003, p. 493), the tagline of the Spectrum Scholarship Initiative is the future is overdue. The continued recruitment and retention of minority PhD candidates will have a direct and lasting impact on the LIS professoriate and the field of librarianship as a whole.

References

Alire, C. A. (2001). Diversity and leadership. Journal of Library Administration, 32(3), 99-114.

Balderrama, S. R. (2000). This trend called diversity. Library Trends, 49(1), 194-214.

Barlow, D. L., & Aversa, E. (2006). Library professionals for the 21st century academy. Advances in Librarianship, 30, 327-364.

Bonnici, L. & Burnett, K. (2005). A Web model of recruitment for LIS doctoral education: Weaving in diversity. In M. B. Wheeler (Ed.), Unfinished business: Race, Equity, and Diversity in Library and Information Science Education (pp. 199-130). Metuchen, NJ: Scarecrow Press.

Brown-Syed, C., Baker, L., & Wicks, D. A. (2008). Doctoral recruitment factors: Results of a survey of deans and directors. Journal of Education for Library and Information Science, 49(2), 107-115.

Davis, D.M. & Hall, T.D. (2006). Diversity Counts: Office for Research and Statistics & Office for Diversity (ALA). Retrieved March 20, 2010, from htp://www.ala.org/ala/ors/diversitycounts/DiversityCountsReport.pdf

Dewey, B. I., & Keally, J. (2008). Recruiting for diversity: strategies for 21st century research librarianship. Library Hi Tech, 26(4), 622-629.

Franklin, R. E. & Jaeger, P. T. (2007). A decade of doctorates: An examination of dissertations written by African American women in library and information studies, 1993-2003. Journal of Education for Library and Information Science, 48(3), 187-201.

Gollop, C. J. (1999). Library and information science education: Preparing librarians for a multicultural society. College & Research Libraries, 60(4), 385-395.

Honma, T. (2005). Trippin’ over the color line: The invisibility of race in library and information studies. InterActions: UCLA Journal of Education and Information Studies, 1(2), 1-26.

Jaeger, P. T., Bertot, J. C. & Franklin, R. E. (2010). Diversity, inclusion, and underrepresented populations in LIS research. The Library Quarterly, 80(2), 175-181.

Jaeger, P. T. & Franklin, R. E. (2007). The virtuous circle: Increasing diversity in LIS faculties to create more inclusive library services and outreach. Education Libraries, 30(1), 20-26.

Kenney, B. (2008). The spectrum initiative: Affirmative action in the library profession. Reason and Respect, 1(1), 1-3.

Manzo, K. K. (1994). Flaws in fellowships: Institutional support essential to boosting number of African American doctoral students. Black Issues in Higher Education, 11(10), 46-52.

Marton, F. (1981). Phenomenography: Describing conceptions of the world around us. Instructional Science, 10, 177-200.

Meacham, J. (2002). Our doctoral programs are failing our under graduate students. Liberal Education, 88(3), 22-28.

Neely, T. Y. & Peterson, L. (2007). Achieving racial and ethnic diversity among academic and research librarians: The recruitment, retention, and advancement of librarians of color. A White Paper by the ACRL Board of Directors Diversity Task Force. Chicago, IL: Association of College and Research Libraries.

Oder, N. (2010, January 17). ALA 2010 midwinter meeting: Spectrum scholarship fund gets $100,000 boost from Turock. Retrieved April 20, 2010, from Library Journal.com website: http://www.libraryjournal.com/article/CA6715423.html?industryid=47133

Peterson, L. (1999). The definition of diversity. Journal of Library Administration, 27(1), 17-26.

Pruitt, A. S., & Isaac, P. D. (1985). Discrimination in recruitment, admission, and retention of minority graduate students. Journal of Negro Education, 54(4), 526-536.

Reeling, P. G. (1992). Doctorate recipients in library science: How they compare with doctorate recipients in other disciplines. Journal of Education for Library and Information Science, 33(4), 311-329.

Roy, L., Johnson-Cooper, G., Tysick, C., & Waters, D. (2006). Bridging boundaries to create a new workforce: A survey of spectrum scholarship recipients, 1998-2003. Spectrum Survey Report. Chicago, IL: The American Library Association’s Office for Diversity.

Stringer-Stanback, K. (2008). Recruitment, retention & diversity in libraries & higher education: Why doing the right thing is easier said than done. North Carolina Libraries, 66(1), 25-27.

Turock, B. J. (2003). Developing diverse professional leaders. New Library World, 104(11/12), 491-498.

Wheeler, M. B. (Ed.). (2005). Unfinished Business: Race, Equity, and Diversity in Library and Information Science Education. Metuchen, NJ: Scarecrow Press.

Whitwell, S. (1998). News fronts-ALA-why we need the spectrum scholarship. American Libraries, 29(10), 6.

Winston, M. (2008). Diversity: the research and the lack of progress. New Library World, 109(3/4), 130-149.