The conceptual landscape of iSchools: Examining current research interests of

faculty members

Kim Holmberg1, Andrew Tsou2 and Cassidy Sugimoto2

1 k.holmberg@wlv.ac.uk, http://kimholmberg.fi School of Technology, University of Wolverhampton, UK

2School of Information and Library Science, Indiana University Bloomington, USA

Colis 2013, Copenhagen, Denmark, 20/8/13

Cascades, Islands, or Streams? Time, Topic, and Scholarly Activities in

Humanities and Social Science Research

Indiana University, Bloomington, USAUniversity of Wolverhampton, UKUniversité de Montréal, Canada

Cascades, Islands, or Streams? Integrate several datasets representing a broad range of scholarly activities

Use methodological and data triangulation to explore the lifecycle of topics within and across a range of scholarly activities

Develop transparent tools and techniques to enable future predictive analyses

RQThis project seeks to answer two research questions: 1. What is the current research landscape of the

iSchools?2. What types of shared research interests are

there amongst the iSchools?

DATAThe self-described research interests of iSchool faculty members were collected manually by visiting their university webpages, personal websites or other online profiles.

By studying the current research interests as described by the scholars themselves, we are able to elucidate an intellectual landscape and relationship among iSchools that is at once contemporary and highly granular.

A total of 6,760 keywords were collected from 1,168 researchers’ online profiles from 39 iSchools.

Keywords were available for more than 90% of the faculty at 30 iSchools, and 13 schools had a 100% keyword location rate.

DATA

METHODSCo-word analysis was used to 1) find the core research interests of the iSchools and 2) to discover the shared research interests between the iSchools. BibExcel and Gephi (Force Atlas) were used for the analysis.

Cosine similarities were calculated between the iSchools . Only those research interests that were shared by at least two researchers from same university were included. University of Amsterdam and University of Glasgow did not meet this criterion and were not included in this part.

RESULTSTable 1. The 20 most frequently mentioned research interests

RQ1

# of researchers Research interest # of researchers Research interest85 human-computer interaction 35 software engineering72 information retrieval 35 information management71 digital libraries 32 information behaviour54 information technology 30 privacy52 information systems 29 technology52 data mining 28 information policy49 social media 28 learning48 knowledge management 28 evaluation42 information seeking 27 machine learning36 education 25 artificial intelligence

Figure 1. Co-word map of the research

interests at iSchools

RQ1

RQ1

Community detection in Gephi discovered 7

communities

The 7 communities detected represent the very core of the current research interests at the iSchools, thus addressing this study’s first research question.

RQ1

Computing information

Social media and information systems

Education and information technology

Information seeking and digital libraries

Libraries and library services

Information retrieval and data mining

Data analytics and computing

Cosine similarityUniversity of California, Irvine Georgia Tech 0.617University of North Texas University of North Carolina 0.593Wuhan University Nanjing University 0.578Florida State University Rutgers 0.489University of California, LA University of Kentucky 0.471University of California, Berkeley Pennsylvania State University 0.469Rutgers University of North Texas 0.458Rutgers Drexel University 0.453

Table 2. The strongest cosine similarities between the iSchools

RQ2The relations between the iSchools’ current research profiles were also compared by calculating the cosine similarity between them.

Cosine similarities between the iSchools

RQ2

Human-computer interaction Information retrieval Digital libraries

Information technology Information systems Data mining

The six most frequently shared research interests, along with the iSchools from which at least two researchers had mentioned the respective research interest

RQ2

LIMITATIONS

• Data collection• Different coverage of the faculty at the iSchools

(e.g. University of Kentucky with 67% coverage compared with 100% for 13 other iSchools).

• Wuhan University; keywords translated via Google Translate

• Finding the perfect level between general and specific keywords

CONCLUSIONS

The iSchools still contain many dominant themes from LIS, but have an expanded conceptual landscape with the introduction of new iSchools reflecting the iSchools’ interdisciplinary composition.

The methods used for data collection guaranteed the most current data available (in contrast to using publications) and the methods used for analyses gave multiple perspectives to the research landscape of the iSchools.

FUTURE

Are the core research interests built by the first iSchools, with the newcomers occupying the less shared research interests?

Have the discovered shared research interests manifested in collaboration in the form of co-authored research articles or are the iSchools in that sense isolated islands?

Do the research profiles based on self-described research interests differ from the ones obtained by using keywords from publications?

Kim HolmbergStatistical Cybermetrics Research GroupUniversity of Wolverhampton, UKK.Holmberg@wlv.ac.uk http://kimholmberg.fi @kholmber

AcknowledgementsThis manuscript is based upon work supported by the international funding initiative Digging into Data. Specifically, funding comes from the National Science Foundation in the United States (Grant No. 1208804), JISC in the United Kingdom, and the Social Sciences and Humanities Research Council of Canada.

Thank you for your attention