Reliability Test of an Established Pedestrian Environment Audit in Rural Settings

• DOI: 10.4278/ajhp.081218-arb-311
• Publication Date: 2010-11-01T21:46:27Z
• Authors: Brian D. Fisher; Sandra Richardson; Akiko S. Hosler
• Keywords: Observer Variation; New York; Reproducibility of Results; Rural Health; Walking; 03 medical and health sciences; Cross-Sectional Studies; 11. Sustainability; Cluster Analysis; Humans; Environment Design; 0305 other medical science

To test inter-rater reliability and internal consistency of a reduced-item version an environmental audit tool for physical-activity resources in rural setting.A cross-sectional design.Five school districts western New York.A random sample 391 street segments (i.e., the length contained within census block.) Fifty were observed second time analysis.The Pedestrian Environmental Data Scan (PEDS), previously tested instrument conducting audits built environment supports, was used. Modifications made use settings.Four tests (Kappa [κ]; prevalence-adjusted, bias-adjusted κ [PABAK]; percent agreement; Spearman rho [ρ]) one (Cronbach α) used.Nineteen measures identified their utility setting. These had average .43, PABAK .72, agreement 86.2%, indicating good agreement. They organized into four subscale areas that ρ ranging from moderate (ρ = .38) to strong .66). Alpha (α) values scale overall (α .60) subscales (range, .25-.66) suggested items included likely be multi-dimensional.A simpler has been might more appropriate