MPLUS结构方程模型应用：[6]模型语句举例

1、带协变量的CFA模型（MIMIC模型）TITLE: this is an example of a MIMIC modelwith two factors, six continuous factor indicators, and three covariates！CFA中包含两个因子，留个连续因子指标，以及三个协变量DATA: FILE IS mimic.dat; ！数据文件VARIABLE: NAMES ARE y1-y6 x1-x3; ！变量名称，y1-y6为测量指标，x1-x3为3个协变量MODEL: f1 BY y1-y3;f2 BY y4-y6; f1 f2 ON x1-x3; ！y1-y3负载在f1因子上,y4-y6负载在f2因子上，同时均有f1 f2在x1-x3上的回归。

2、带时间不变性协筐毙险裆变量的增长模型TITLE: this is an example of a linear growt茑霁酌绡hmodel for a continuous outcome at fourtime points with the intercept and slopegrowth factors regressed on two timeinvariant covariates！四个时间点上的连续变量，同时截距和斜率因子在两个时间不变性协变量上的回归DATA: FILE IS growth.dat;VARIABLE: NAMES ARE y1-y4 x1 x2; ！y1-y4表示4个点的变量数据，x1 x2是两个协变量MODEL: i s |y1@0 y2@1 y3@2 y4@3;i s ON x1 x2;！i表示截距，s表示斜率，|表示定义随机效应变量，i s在y1-y4上的负载均分别为0,1,2,3。i s ON x1 x2; 表示x1,x2对i s的回归。

3、带协变量和直接效应的潜类别分析模型（LCA）TITLE: this is an example of a latent classanalysis with two classes, one covariate,and a direct effect ！潜类别分析包括两类，一个协变量和一个直接效应DATA: FILE IS lcax.dat;VARIABLE: NAMES ARE u1-u4 x;CLASSES = c (2); ！定义潜类别变量c,包括两个类别 CATEGORICAL = u1-u4; ！潜类别变量的指示变量u1-u4为分类变量ANALYSIS: TYPE = MIXTURE; ！分析方法为混合模型MODEL: %OVERALL% ！混合模型中必须的语句c ON x; ！潜类别c在协变量x上的回归 u4 ON x; ！指示变量u4在x上的回归

4、带随机截距和随机斜率的多水平回归模型TITLE: this is an example of a multilevelregression analysis with one individual- level outcome variable regressed on an individual-level background variable where the intercept and slope are regressed on a cluster-level variable！多水平回归相当于我们平时才用HLM分析多水平线性模型一样，一个个体水平上的变量（一级水平），一个簇水平上的变量（二级水平）DATA: FILE IS reg.dat;VARIABLE: NAMES ARE clus y x w;CLUSTER = clus; WITHIN = x; BETWEEN = w; MISSING = .; ！clus为二级分组变量，x为被试内变量（一级变量），y为被试间变量（随着clus而变化的二级变量）。缺失值为“."。DEFINE: CENTER x (GRANDMEAN);！定义对x变量进行整体中心化。ANALYSIS: TYPE = TWOLEVEL RANDOM; ！两水平的随机模型MODEL:%WITHIN% s | y ON x; ！被试内的y随x变化的斜率%BETWEEN% y s ON w; ！被时间y和斜率随w变化