Walk Score®作為台灣地區宜走性環境測量工具之效度檢驗

Validation of Walk Score® for estimating neighborhood walkability in Taiwan

目標：即使在美國、加拿大、與日本地區，已有研究證實Walk Score®作為宜走性環境測量工具之效度，但對於台灣地區仍屬未知，因此本研究檢驗Walk Score®與使用地理資訊系統（geographic information system, GI）所取得之客觀宜走性環境指標之相關性，以驗證Walk Score®應用於台灣地區之效度。方法：以台灣地區之65歲以上高齡者為目標族群，進行電腦輔助電訪調查，蒐集自陳資料（社會人口學變項與居住之村里），有效樣本為1,040人；使用GIS計算五項客觀宜走性環境指標（人口密度、十字路口密度、總目的地數、人行道面積與大眾運輸站點密度），接著使用Walk Score®網站（https://www.walkscore.com/）取得各村里Walk Score分數，使用皮爾森相關係數（Pearson’s correlation coefficient）檢視Walk Score分數與客觀宜走性環境指標（GIS）的相關程度。結果：統計顯示Walk Score分數與客觀宜走性環境指標具有顯著正相關，其中Walk Score分數與十字路口密度（r＝0.7）及總目的地數量（r＝0.7）之相關性最高。結論：本研究結果與西方國家及日本一致，WalkScore®作為宜走性環境測量工具，具有易操作與低成本的特性，且在台灣地區有良好的效度，但該工具在東部及離島的應用須較為謹慎。建議未來政策制定者與公共衛生研究者能利用WalkScore®以取得台灣各鄰里之宜走性。 Objectives: Although some studies have evaluated the concurrent validity of the Walk Score® as a measure of neighborhood walkability in the United States, Canada, and Japan, whether the Walk Score® is a valid measure in Taiwan has not yet been determined. Accordingly, this study examined how Walk Score® is correlated with geographic information system (GIS)- derived characteristics of neighborhood walkability for residential neighborhoods in Taiwan. Methods: A computer-assisted, telephone-based survey was conducted on older Taiwanese adults. Self-reported data (sociodemographic and residential neighborhood) were collected from 1040 Taiwanese residents aged 65 years or older. Five environmental characteristics (population density, intersection density, local destinations, sidewalk availability, and access to public transportation) were calculated for each neighborhood by using a GIS. Consequently, Walk Scores were calculated on the Walk Score® website (https://www.walkscore.com/) by inputting residential neighborhood information. The Pearson correlation coefficients between Walk Score® data and objective measures of environmental attributes (GIS data) were determined. Results: Significant positive correlations were discovered between the Walk Score® data and five GIS-derived environmental characteristics. Walk Score® was most strongly associated with intersection density (r = 0.70) and number of local destinations (r = 0.70). Conclusions: Consistent with the results obtained in Western countries and Japan, Walk Score® , a simple and low-cost tool, appears to also be a valid measure of neighborhood walkability in Taiwan. However, the tool should be applied with caution in the eastern regions and offshore islands of Taiwan. Walk Score® data can enable policy makers and public health practitioners to identify the walkability of local areas in Taiwan.