Paper
Monday, November 14, 2005
The Development and Psychometric Evaluation of the Personal Respiratory Protection Scale
M. Susan Jones, PhD, RN1, Donna Gates, EdD, RN2, and Linda Baas, PhD, CS-ACNP2. (1) Department of Nursing, Western Kentucky University, Bowling Green, KY, USA, (2) College of Nursing, University of Cincinnati, Cincinnati, OH, USA
| Learning Objective #1: Describe the steps in the conceptual development of an new measurement instrument |
| Learning Objective #2: Discuss the psychometric evaluation of a new mesurement instrument |
Workers in swine confinement buildings (SCBs) are frequently exposed to environmental dusts and gases; however, many of these workers do not use personal respiratory protection (PRP). The purpose of this presentation is to describe the conceptual development and psychometric evaluation of a Personal Respiratory Protection Scale (PRPS) designed to identify the variables influencing SCB workers to wear PRP. Based on a literature review and findings from two focus group interviews with 22 workers in SCBs, 37 items were identified and grouped into five conceptual constructs (barriers, benefits, severity, susceptibility, and norms) theorized to influence the frequency of PRP use by SCB workers. The establishment of content validity involved a critique by a panel of eight professionals with expertise in health risks associated with working in SCBs. Information obtained from these experts was used to delete non-relevant items, make editorial changes to improve the clarity and readability of the items, and to clarify the definition of one theorized construct. Next, sections were added to the instrument to elicit information about the SCB workers' demographics, occupational and respiratory health histories, and frequency of PRP use before the instrument was pilot tested with SCB workers. The instrument was used to collect data from 503 SCB workers attending the 2003 World Pork Expo in Des Moines, Iowa in June 2003. Principle component analysis (PCA) and varimax rotation were used to establish construct validity and resulted in the identification of eight factors (i.e., benefits, norms, severity, susceptibility, personal barriers, knowledge barriers, external barriers, and habit barriers). Cronbach alpha values for the factors ranged from .58 to .91 with all but two values exceeding .77. Suggestions for improving the alpha values for these two factors will be addressed.