Water intensity benchmarks for sustainable retail stores

The table present retrieved data of the highest revenue retailers for the research paper entitled "Water Intensity benchmarks for sustainable retail stores", according to the variables: "Water Intensity (WI)", “Company”, “Country of origin”, ”Dominant operational category”, ”Store typology”, ”Number of countries of operation”, “Retail revenue”, “Number of stores”, ”Average store sales area”, ”Total store sales area”, “Water intensity”, ”Average number of workers per store”, “Total number of workers” and “Revenue per store sales area”. Based on this data, a WI benchmark was performed, identifying average, minimum and maximum values for each food and non-food retail sub-type, and outliers were identified with the interquartile range and removed so as to reduce error in each category. A linear regression analysis was also performed in retail sub-types that had data from three or more retailers, in order to estimate the relationship between WI as a dependent variable and other independent variables. R-squared values (R²) were calculated for each independent variable, those scoring higher than 0.7 were considered to have a strong effect size on the prediction of WI.本表格呈现了题为《可持续零售门店用水强度基准》的研究论文中，针对营收最高零售商所检索获取的数据，涉及变量包括：用水强度（Water Intensity, WI）、公司（Company）、起源国家（Country of origin）、核心运营品类（Dominant operational category）、门店类型（Store typology）、运营国家数量（Number of countries of operation）、零售营收（Retail revenue）、门店总数（Number of stores）、单店平均销售面积（Average store sales area）、总销售面积（Total store sales area）、用水强度、单店平均员工数（Average number of workers per store）、员工总数（Total number of workers）以及单位销售面积营收（Revenue per store sales area）。基于该数据，研究开展了用水强度基准分析，分别针对食品零售与非食品零售的各子品类计算其平均值、最小值与最大值；同时采用四分位距（interquartile range）法识别并剔除异常值，以降低各品类的分析误差。针对拥有三家及以上零售商数据的零售子品类，研究还开展了线性回归分析（linear regression analysis），以估算以用水强度为因变量（dependent variable）、其余变量为自变量（independent variable）的相关关系。研究计算了各自变量的决定系数（R-squared, R²），其中数值大于0.7的变量被认为对用水强度的预测具备较强的效应量（effect size）。