Files in this item



application/pdfKyrillidou_Martha.pdf (889kB)
(no description provided)PDF


Title:Item sampling in service quality assessment surveys to improve response rates and reduce respondent burden: The “LibQUAL+® Lite” randomized control trial (RCT)
Author(s):Kyrillidou, Martha
Director of Research:Smith, Linda C.
Doctoral Committee Chair(s):Smith, Linda C.
Doctoral Committee Member(s):Deltas, George; Palmer, Carole L.; Thompson, Bruce
Department / Program:Library & Information Science
Discipline:Library & Information Science
Degree Granting Institution:University of Illinois at Urbana-Champaign
Subject(s):Library assessment
Measuring service quality
Randomized Control Trial
Matrix sampling
Survey research
Experimental research design
Survey length
Survey response time
Abstract:The purpose of the study was twofold: (a) identifying whether item sampling using matrix sampling methods improved the well-known survey protocol, LibQUAL+®, and produced a viable alternative, LibQUAL+® Lite; in particular, improvements regarding participation rates, completion time for the survey, and results comparisons are examined in the Lite version of the protocol within different institutional settings through a series of randomized control trials; (b) identifying whether there are differences in the total, subscale, and linking item scores between the long and the Lite protocol overall as well as within the three main user groups: undergraduate students, graduate students and faculty. For the purposes of this study data from more than 10,000 library users from 14 institutions that implemented randomized control trials during the spring 2008, fall 2008, and spring 2009 survey cycles were analyzed. Findings indicate that LibQUAL+® Lite is a viable and preferred alternative to the long form of 22 core items that has been established since 2003. LibQUAL+® Lite uses item sampling methods to: (a) gather data on all 22 LibQUAL+® core items, while (b) each individual participant responds to only a subset of items. Every Lite user responds to one “linking” item from each of the subscales, and to a randomly-selected subset of five items from the remaining 19 (22-3) core LibQUAL+® items. As a consequence, survey response times are roughly cut in half, while the library still receives data on every survey question. The matrix sampling method, the randomized control trial framework, and the statistical analysis methods outlined in the current study are useful heuristic methods for other high stakes library survey implementations whether for a physical as well as a digital library environment. These methodological approaches add rigor and thoughtful perspectives as they inform ways libraries shape their services and “touch” their users through improvements and innovations in the years to come.
Issue Date:2010-01-06
Rights Information:Copyright 2009 Martha Kyrillidou
Date Available in IDEALS:2010-01-06
Date Deposited:December 2

This item appears in the following Collection(s)

Item Statistics