Predictors, standardised, and unstandardised coefficients for the discriminant analysis model and logistic regression model.

Note:1The standardised discriminant function coefficients serve the same purpose as beta weights in multiple regressions (partial coefficient): they indicate the relative importance of the independent variable in predicting disability status within the study population. They allow you to compare variables measured on different scales. Coefficients with large absolute values correspond to variables with greater discriminating ability2The structure matrix table shows the correlations of each variable with each discriminant function; the correlations then act similarly to factor loadings in factor analysisagreater score = worse healthbSDQ—peer problems was not a significant predictor in the logistic regression modelSPPA = Self—Perception Profile for Adolescents.Predictors, standardised, and unstandardised coefficients for the discriminant analysis model and logistic regression model.注释1：标准化判别函数系数（standardised discriminant function coefficients）的功能等同于多元回归中的β权重（beta weights，偏回归系数）：其可反映自变量在研究人群中预测残疾状态时的相对重要性，可用于比较不同量纲下测量的变量。绝对值较大的系数，对应的变量区分能力更强。 注释2：结构矩阵表展示了各变量与各判别函数之间的相关关系，此类相关的作用与因子分析中的因子载荷（factor loadings）相仿。 a 得分越高，健康状况越差 b SDQ—同伴问题未在逻辑回归模型中成为显著预测变量 SPPA即青少年自我知觉量表（Self-Perception Profile for Adolescents）。 本部分呈现了判别分析模型与逻辑回归模型的预测变量、标准化系数及非标准化系数。