Show simple item record

dc.contributor.advisorHeidorn, Patrick Bryan
dc.contributor.authorAtkinson, Brian Lewis
dc.creatorAtkinson, Brian Lewis
dc.date.accessioned2019-01-08T01:52:38Z
dc.date.available2019-01-08T01:52:38Z
dc.date.issued2018
dc.identifier.urihttp://hdl.handle.net/10150/631354
dc.description.abstractPosition postings and professional literature represent opportunity for insights into the expectations of aspiring position applicants as well as describe appropriate skills, experiences and management or leadership approaches that can inform incumbent information executive (e.g. Chief Information Officers and Library Deans) on effective leadership of their respective organizations within colleges and universities across North America. This study also lays the foundation of a model that leverages massive quantities of anecdotal literature in quantitative and empirical manners. Such a model, when fully developed can be used to examine all sorts of qualitative data of sufficient size. Analysis of the data can then be performed using a variety of Natural Language processes and tools as well as statistical tests to evaluate correlations, commonalities and frequency of the terms in relation to the roles that individuals play in their institutions. Individual tokens are used rather than nGrams or co-located tokens in an effort to minimize potential bias that can be found in phrases or word combinations such as ‘creative leadership’ or ‘strong management’. Instead, terms are used as a way to identify the tokens ‘creative’ and ‘strong’ and correlate their frequency with roles, literature and position postings. In contrast to other common techniques, such as Latent Semantic Analysis, this approach values the impact of individual responses and perspectives by using qualitative questions to cull initial tokens as well as quantitative surveys to reinforce outcomes. With few adjustments to accommodate specific disciplines, this study successfully forms a foundation requiring future development that analyzes large quantities of textual data that otherwise might have been painstakingly analyzed using time and labor-intensive qualitative processes such as content analysis. This study also identifies a list of tokens that are commonly used in position postings and leadership literature for library deans and Chief Information Officers. Using these tokens in a reflective manner, the initial postings and literature can later be more granularity analyzed the potential for inclusion in future writing. Outcomes of this investigation found 8 tokens that both Library Deans and CIOs found to be ‘extremely important’ as well as a longer list of tokens that while not found in common were considered important to particular roles. Considering these tokens can be important for authors of position postings and professional literature alike.
dc.language.isoen
dc.publisherThe University of Arizona.
dc.rightsCopyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction, presentation (such as public display or performance) of protected items is prohibited except with permission of the author.
dc.subjectData Mining
dc.subjectLatent Semantic Analysis
dc.subjectLeadership
dc.subjectLemmatization
dc.subjectNatural Language Processing
dc.subjectText Mining
dc.titleInformation Leadership: A Quantitative Analysis of Language Across Literature, Position Postings and the Roles that Leaders Play
dc.typetext
dc.typeElectronic Dissertation
thesis.degree.grantorUniversity of Arizona
thesis.degree.leveldoctoral
dc.contributor.committeememberStoffle, Carla
dc.contributor.committeememberHammond, Mike
dc.contributor.committeememberVaillancourt, Alison
dc.contributor.committeememberBonito, Joe
thesis.degree.disciplineGraduate College
thesis.degree.disciplineInformation
thesis.degree.namePh.D.
refterms.dateFOA2019-01-08T01:52:38Z


Files in this item

Thumbnail
Name:
azu_etd_16760_sip1_m.pdf
Size:
7.584Mb
Format:
PDF

This item appears in the following Collection(s)

Show simple item record