TM Project web site: Bruce Thompson Colleen Cook Martha Kyrillidou Glasgow January 19-20, 2004 Quantitative Grounding “22 Items and a Box” Quantitative Grounding “22 Items and a Box” :
Score Integrity uReliability uValidity
By Language ServiceInfo.Lib as Group n AffectControlPlaceTOTAL American (all)59, British (all) 6, French (all) alpha By Language
By University Type ServiceInfo.Lib as Group n AffectControlPlaceTOTAL Comm Colleges 4, yr Not ARL36, yr, ARL14, Acad Health 3, alpha by University Type
By Consortium ServiceInfo.Lib as Group n AffectControlPlaceTOTAL AAHSL 3, OhioLINK 7, Oberlin 2, Merlin NY3R Coll/Uni13, SCONUL 6, Alabama 1, alpha by Consortium
n = 59,318; α =.9593 "Corrected" Item ‑ alpha Totalif Item ItemCorrelationDeleted SA01APER SA04APER SA07APER SA10APER SA12APER SA15APER SA17APER SA20APER SA23APER
n = 59,318; α =.9593 "Corrected" Item ‑ alpha Totalif Item ItemCorrelationDeleted PC02APER PC11APER PC16APER PC21APER PC25APER IA03APER IA14APER IA18APER
n = 59,318; α =.9593 "Corrected" Item ‑ alpha Totalif Item ItemCorrelationDeleted LP05APER LP09APER LP13APER LP19APER LP24APER
Service Affect (n = 71,170 English) SA20APER SA07APER SA17APER SA04APER SA15APER SA23APER SA01APER SA12APER SA10APER Service Affect
Library as Place Library as Place (n = 71,170 English) LP13APER LP05APER LP09APER LP24APER LP19APER
Information Control Information Control (n = 71,170 English) IA18APER PC11APER IA03APER PC25APER PC21APER PC02APER PC16APER IA14APER
2. Validity Correlations Validity Correlations Serv_AffInfo_ConLibPlaceTOTALper Serv_Aff Info_Con LibPlace TOTALper ESAT_TOT EOUT_TOT
Mean Perceived Scores 2001/2002 Trend (n=34)
The Challenge of Score Use u Securing information that contributes meaningfully to planning and improvement efforts at a local level u Providing analytical frameworks that institutional staff can apply without extensive training or assistance u Helping decision-makers understand success of investments u Finding useful inter-institutional comparisons
Two Interpretation Frameworks uZone of Tolerance uScore Norms
1. Zone of Tolerance uThe distance between minimally acceptable and desired service quality ratings uPerception ratings ideally fall within the Zone of Tolerance
Faculty Dimension Summary Note: LibQUAL+ Spring 2002 Survey Results - TAMU. (2002). Vol. 2, p. 42 Average Rating Access to Information Affect of Service Library as Place Personal Control Perceived Range of Minimum to Desired Range of Minimum to Perceived (“Gap”)
7 Institution Types 1.Community Colleges 2.Health Sciences 3.Military 4.Colleges/Universities 5.State/Publics 6.United Kingdom 7.French Canadian
LibQUAL+™ 2003 Summary Colleges or Universities
LibQUAL+™ 2003 Colleges or Universities – Dimension Means
LibQUAL+™ 2003 Colleges or Universities – Library Use
LibQUAL+™ 2003 Colleges or Universities – Core Questions
LibQUAL+™ 2003 Summary Community Colleges
LibQUAL+™ 2003 Community Colleges – Dimension Means
LibQUAL+™ 2003 Community Colleges – Library Use
LibQUAL+™ 2003 Summary Academic Health Sciences
LibQUAL+™ 2003 Health Sciences – Dimension Means
LibQUAL+™ 2003 Health Sciences – Library Use
LibQUAL+™ 2003 Summary Military Institutions
LibQUAL+™ 2003 Military Institutions – Dimension Means
LibQUAL+™ 2003 Military Institutions – Library Use
LibQUAL+™ 2003 Summary State and Public Institutions
LibQUAL+™ 2003 State and Public – Dimension Means
LibQUAL+™ 2003 State and Public – Library Use
LibQUAL+™ 2003 Summary United Kingdom
LibQUAL+™ 2003 United Kingdom – Dimension Means
LibQUAL+™ 2003 United Kingdom – Library Use
LibQUAL+™ 2003 Summary French Canadian
LibQUAL+™ 2003 French Canadian – Dimension Means
LibQUAL+™ 2003 French Canadian – Library Use
2. Score Norms uNorm Conversion Tables facilitate the interpretation of observed scores using norms created for a large and representative sample. uLibQUAL+ norms have been created at both the individual and institutional level
Institutional Norms for Perceived Means on 25 Core Questions Note: Thompson, B. LibQUAL+ Spring 2002 Selected Norms, (2002).
Response Rates 1. It’s representativesness that counts more than the response rate!
Response Rates 2. We only know the lower bound (or minimum) estimate of the response rate, because we don’t know the correct denominator.
Response Rate Contexts 1. Response rates are higher for action-oriented surveys, which are not relevant or practical for total market surveys.
Response Rate Contexts 2. Survey saturation is tending to lower typical response rates over time.
2005 and Beyond 1.Matrix Sampling / Short Forms 2.Software Used for Local/Consortial Assessments 3.“Action Snapshots”
LibQUAL+ Related Documents LibQUAL+ LibQUAL+ Web Site LibQUAL+ LibQUAL+ Bibliography Survey Participants Procedures Manual