Selecting the Right Research Instrument: Going Beyond the Title and Reported Psychometrics

Wednesday, 24 July 2013: 9:10 AM

Lillie M. Shortridge-Baggett, RN, EdD, FAAN, FNAP
Graduate Studies in Nursing, Pace University, Pleasantville, NY
Mary Courtney, PhD, MHP, BAdmin (Acc), RN
School of Nursing, Midwifery and Paramedicine, Australian Catholic University, Brisbane, Australia
Carol Reid, PhD, MAppSc, GradCertHlth (Sexual Health), BHSc(N), RN
Research and Development Unit, Centre for Clinical Nursing, Royal Brisbane and Women's Hospital, Brisbane, Australia

Learning Objective 1: The learner will be able to describe how the principles of instrumentation are to be followed in development and selection of the instrument.

Learning Objective 2: The learner will be able to identify key aspects of measurement that must be addressed in selecting an instrument.

Purpose:  

The aims of this presentation are to examine the critical instrumentation and measurement issues in selecting a research instrument for use in research and to apply to two scales used in a recently completed study.  Often the title and reported psychometrics are used as the main guide for choosing an instrument for use in a study.   Additional criteria will be discussed.

Methods:

Two instruments used in a recent study in which these presenters were involved as a consultant, supervisor, and investigator will be discussed as exemplars.  These instruments, both scales based on the same theoretical perspective, will be examined from the titles of the scales, adherence to the theory and assumptions, appropriateness for the research questions and hypotheses, completeness in covering the essential aspects of the construct, consistency with variable definitions, and use of the recommended format for measuring the construct. An overview of the psychometrics of these scales in previous studies and the current one as well as the outcomes of the above review will be presented.

 Results:  

The results of the critical appraisal of two research instruments show that both scales had strong psychometrics in previous studies in which they were used.  They did not adhere, however, to all dimensions of the theory they were developed to measure.  In addition, they did not adhere consistently to the recommended format for measuring the construct. 

Conclusion:

These results indicate careful appraisal of the instrumentation and measurement properties using additional criteria is important.  Both of these areas are essential to the quality of the study.