Troped Philip J, Cromley Ellen K, Fragala Maren S, Melly Steven J, Hasbrouck Hope H, Gortmaker Steven L, Brownson Ross C
J Phys Act Health. 2006 Feb;3(s1):S158-S175. doi: 10.1123/jpah.3.s1.s158.
To determine how trail characteristics may influence use, reliable and valid audit tools are needed.
The Path Environment Audit Tool (PEAT) was developed with design, amenity, and aesthetics/maintenance items. Two observers independently audited 185 trail segments at 6 Massachusetts facilities. GPS-derived items were used as a "gold standard." Kappa (k) statistics, observed agreement and intraclass correlation coefficients (ICCs) were calculated to assess inter-observer reliability and validity.
Fifteen of 16 primary amenity items had k-values ≥ 0.49 ("moderate") and all had observed agreement ≥ 81%. Seven binary design items had k-values ranging from 0.19 to 0.71 and three of 5 ordinal items had ICCs ≥ 0.52. Only two aesthetics/maintenance items (n = 7) had moderate ICCs. Observed agreement between PEAT and GPS items was ≥ 0.77; k-values were ≥ 0.57 for 7 out of 10 comparisons.
PEAT has acceptable reliability for most of its primary items and appears ready for use by researchers and practitioners.
为了确定步道特征如何影响其使用情况，需要可靠且有效的审计工具。
路径环境审计工具（PEAT）是根据设计、便利设施以及美学/维护项目开发的。两名观察员独立对马萨诸塞州6个设施的185个步道段进行了审计。利用全球定位系统得出的项目作为“金标准”。计算卡帕（k）统计量、观察一致性和组内相关系数（ICC）以评估观察员间的可靠性和有效性。
16个主要便利设施项目中有15个的k值≥0.49（“中等”），且所有项目的观察一致性≥81%。7个二元设计项目的k值在0.19至0.71之间，5个有序项目中有3个的ICC≥0.52。只有两个美学/维护项目（n = 7）具有中等ICC。PEAT与全球定位系统项目之间的观察一致性≥0.77；10次比较中有7次的k值≥0.57。
PEAT的大多数主要项目具有可接受的可靠性，似乎可供研究人员和从业人员使用。
