Mplus model90 模型讲解

来自图书《MPlus中介调节模型》

Mplus 复杂中介调节模型教程

• 理论模型
• 数学模型
• 数学推导
• 代码解读

大家好，欢迎来到 Data Sense 的 Mplus 复杂中介调节模型教程。本教程旨在帮助大家快速掌握如何运用 Mplus 分析包含两个串行中介变量和一个调节变量的复杂模型。为了方便大家学习，我们将教程内容分为四个主要部分。首先，我们会深入探讨模型的理论基础，明确变量间的关系和研究假设。接下来，我们将呈现模型的数学表达形式，建立清晰的数学模型框架。然后，我们会详细推导模型的数学公式，帮助大家理解模型背后的计算逻辑。最后，我们会逐行解读 Mplus 代码，让大家能够实际操作并分析自己的数据。通过这四个部分的学习，相信大家可以熟练运用 Mplus 分析复杂的中介调节模型。本教程是您快速掌握此类模型分析的实用指南，无需从头学习所有知识，随时可以查阅。如果您需要更专业的统计咨询服务，欢迎联系 Data Sense。

理论模型

我们来看这个模型图。X通过箭头指向M1，表示X会影响M1；M1通过箭头指向M2，表示M1会影响M2；M2通过箭头指向Y，表示M2会影响Y。同时，X也通过箭头直接指向Y，表示X也会直接影响Y。W通过箭头指向Y，表示W会影响Y；W还通过箭头指向M2到Y的路径，表示W会调节M2对Y的影响。最后，X和W的交互作用也通过箭头指向Y，表示X和W的交互作用也会影响Y。

数学模型

这张图展示了一个比较复杂的模型。X通过两条路径影响Y。一条路径是，X先通过a1影响M1，然后M1通过d1影响M2，最终M2通过b1影响Y；另一条路径是，X直接通过a2影响M2，然后M2通过b1影响Y。同时，W会调节M2到Y的影响，具体来说，W通过b2这条路径影响Y。不仅如此，W还可以通过c2撇这条路径直接影响Y。X也能直接通过c1撇这条路径影响Y。此外，X和W的交互项XW通过c3撇这条路径影响Y。最后，M2和W的交互项M2W通过b3这条路径影响Y。总结来说，这个模型显示了X如何通过多个中介和调节路径影响Y。

数学公式1

模型公式:

1.  Y = b0 + b1M1 + b2M2 + b3M2W + c1'X + c2'W + c3'XW
2.  M1 = a01 + a1X
3.  M2 = a02 + a2X + d1M1

本页PPT展示了我们的模型公式，我们有三个回归方程，分别是Y，M1，和M2。我们的目标是要把Y表示成X的函数，从而理解X对Y的总效应。

数学公式2

模型公式:

1.  Y = b0 + b1M1 + b2M2 + b3M2W + c1'X + c2'W + c3'XW
2.  M1 = a01 + a1X
3.  M2 = a02 + a2X + d1M1
步骤1：替换

Y = b0 + b1(a01 + a1X) + b2(a02 + a2X + d1(a01 + a1X)) + b3(a02 + a2X + d1(a01 + a1X))W + c1'X + c2'W + c3'XW

第一步，我们将M1和M2的公式代入Y的公式中，得到新的Y的表达式，注意这里面既有X也有W。

数学公式3

模型公式:

1.  Y = b0 + b1M1 + b2M2 + b3M2W + c1'X + c2'W + c3'XW
2.  M1 = a01 + a1X
3.  M2 = a02 + a2X + d1M1
步骤1：替换

Y = b0 + b1(a01 + a1X) + b2(a02 + a2X + d1(a01 + a1X)) + b3(a02 + a2X + d1(a01 + a1X))W + c1'X + c2'W + c3'XW
步骤2：展开括号

Y = b0 + a01b1 + a1b1X + a02b2 + a2b2X + a01d1b2 + a1d1b2X + a02b3W + a2b3XW + a01d1b3W + a1d1b3XW + c1'X + c2'W + c3'XW

第二步，我们展开所有括号，将所有的乘法运算展开，以便后续合并同类项。

数学公式4

模型公式:

1.  Y = b0 + b1M1 + b2M2 + b3M2W + c1'X + c2'W + c3'XW
2.  M1 = a01 + a1X
3.  M2 = a02 + a2X + d1M1
步骤1：替换

Y = b0 + b1(a01 + a1X) + b2(a02 + a2X + d1(a01 + a1X)) + b3(a02 + a2X + d1(a01 + a1X))W + c1'X + c2'W + c3'XW
步骤2：展开括号

Y = b0 + a01b1 + a1b1X + a02b2 + a2b2X + a01d1b2 + a1d1b2X + a02b3W + a2b3XW + a01d1b3W + a1d1b3XW + c1'X + c2'W + c3'XW
步骤3：合并同类项

Y = (b0 + a01b1 + a02b2 + a01d1b2 + a02b3W + a01d1b3W + c2'W) + (a1b1 + a2b2 + a2b3W + a1d1b2 + a1d1b3W + c1' + c3'W)X

第三步，我们将不包含X的项合并在一起，包含X的项合并在一起，最后整理成Y=a+bX的形式，方便我们理解X对Y的总效应。

数学公式5

模型公式:

1.  Y = b0 + b1M1 + b2M2 + b3M2W + c1'X + c2'W + c3'XW
2.  M1 = a01 + a1X
3.  M2 = a02 + a2X + d1M1
步骤1：替换

Y = b0 + b1(a01 + a1X) + b2(a02 + a2X + d1(a01 + a1X)) + b3(a02 + a2X + d1(a01 + a1X))W + c1'X + c2'W + c3'XW
步骤2：展开括号

Y = b0 + a01b1 + a1b1X + a02b2 + a2b2X + a01d1b2 + a1d1b2X + a02b3W + a2b3XW + a01d1b3W + a1d1b3XW + c1'X + c2'W + c3'XW
步骤3：合并同类项

Y = (b0 + a01b1 + a02b2 + a01d1b2 + a02b3W + a01d1b3W + c2'W) + (a1b1 + a2b2 + a2b3W + a1d1b2 + a1d1b3W + c1' + c3'W)X
步骤4：提取效应

* 间接效应 (Indirect Effects):
    * a1b1
    * a2(b2 + b3W)
    * a1d1(b2 + b3W)

* 直接效应 (Direct Effect):
    * c1' + c3'W

最后一步，我们从公式中提取出X对Y的间接效应和直接效应。间接效应是X通过M1和M2对Y的影响，直接效应是X直接对Y的影响。同时，我们也能看到调节变量W是如何影响这些效应的。

代码解读1

! Predictor variable - X
! Mediator variable(s) - M1, M2
! Moderator variable(s) - W
! Outcome variable - Y

首先，我们看到的是注释部分，用感叹号开头。这里定义了模型中各个变量的角色：X 是预测变量，M1 和 M2 是中介变量，W 是调节变量，Y 是结果变量。

代码解读2

! Predictor variable - X
! Mediator variable(s) - M1, M2
! Moderator variable(s) - W
! Outcome variable - Y
USEVARIABLES = X M1 M2 W Y XW M2W;

接下来，USEVARIABLES 命令声明了模型中要使用的变量。除了 X、M1、M2、W 和 Y 这五个原始变量，还包括了 XW 和 M2W，它们分别表示 X 和 W 的交互项，以及 M2 和 W 的交互项。

代码解读3

! Predictor variable - X
! Mediator variable(s) - M1, M2
! Moderator variable(s) - W
! Outcome variable - Y
USEVARIABLES = X M1 M2 W Y XW M2W;
! Note that it has to be placed at end of USEVARIABLES subcommand above

这里是一个注释，提醒我们交互项的定义必须放在 USEVARIABLES 命令之后。

代码解读4

! Predictor variable - X
! Mediator variable(s) - M1, M2
! Moderator variable(s) - W
! Outcome variable - Y
USEVARIABLES = X M1 M2 W Y XW M2W;
! Note that it has to be placed at end of USEVARIABLES subcommand above
DEFINE:

XW = X*W;
M2W = M2*W;

DEFINE 命令块用于定义新的变量。这里，XW 被定义为 X 和 W 的乘积，M2W 被定义为 M2 和 W 的乘积，也就是我们需要的交互项。

代码解读5

! Predictor variable - X
! Mediator variable(s) - M1, M2
! Moderator variable(s) - W
! Outcome variable - Y
USEVARIABLES = X M1 M2 W Y XW M2W;
! Note that it has to be placed at end of USEVARIABLES subcommand above
DEFINE:

XW = X*W;
M2W = M2*W;
ANALYSIS:

TYPE = GENERAL;
ESTIMATOR = ML;
BOOTSTRAP = 10000;

ANALYSIS 命令块设置了分析的参数。TYPE = GENERAL 指定分析类型为一般模型。 ESTIMATOR = ML 表示使用最大似然法进行估计。BOOTSTRAP = 10000 表示使用 Bootstrap 方法进行估计，并设置抽样次数为10000。

代码解读6

! Predictor variable - X
! Mediator variable(s) - M1, M2
! Moderator variable(s) - W
! Outcome variable - Y
USEVARIABLES = X M1 M2 W Y XW M2W;
! Note that it has to be placed at end of USEVARIABLES subcommand above
DEFINE:

XW = X*W;
M2W = M2*W;
ANALYSIS:

TYPE = GENERAL;
ESTIMATOR = ML;
BOOTSTRAP = 10000;
! In model statement name each path using parentheses

这是一条注释，提醒我们可以在 MODEL 命令中使用括号来给路径命名，方便后续引用。

代码解读7

! Predictor variable - X
! Mediator variable(s) - M1, M2
! Moderator variable(s) - W
! Outcome variable - Y
USEVARIABLES = X M1 M2 W Y XW M2W;
! Note that it has to be placed at end of USEVARIABLES subcommand above
DEFINE:

XW = X*W;
M2W = M2*W;
ANALYSIS:

TYPE = GENERAL;
ESTIMATOR = ML;
BOOTSTRAP = 10000;
! In model statement name each path using parentheses
MODEL:

Y ON M1 (b1);
Y ON M2 (b2);
Y ON M2W (b3);
Y ON X (cdash1);
Y ON W (cdash2);
Y ON XW (cdash3);
M1 ON X (a1);
M2 ON X (a2);
M2 ON M1 (d1);

MODEL 命令块定义了模型中的路径关系。例如，Y ON M1 (b1) 表示 Y 受 M1 影响，路径系数命名为 b1。以此类推，定义了各个变量之间的影响关系，包括交互项的影响。

代码解读8

! Predictor variable - X
! Mediator variable(s) - M1, M2
! Moderator variable(s) - W
! Outcome variable - Y
USEVARIABLES = X M1 M2 W Y XW M2W;
! Note that it has to be placed at end of USEVARIABLES subcommand above
DEFINE:

XW = X*W;
M2W = M2*W;
ANALYSIS:

TYPE = GENERAL;
ESTIMATOR = ML;
BOOTSTRAP = 10000;
! In model statement name each path using parentheses
MODEL:

Y ON M1 (b1);
Y ON M2 (b2);
Y ON M2W (b3);
Y ON X (cdash1);
Y ON W (cdash2);
Y ON XW (cdash3);
M1 ON X (a1);
M2 ON X (a2);
M2 ON M1 (d1);
! Use model constraint subcommand to test simple slopes
! You need to pick low, medium and high moderator values,
! for example, of 1 SD below mean, mean, 1 SD above mean
! Also calc total effects at lo, med, hi values of moderator

这里是一些注释，解释了接下来的 MODEL CONSTRAINT 块的目的：检验简单斜率，需要在调节变量 W 的低、中、高值下进行计算，并计算总效应。我们通常选取均值加减一个标准差作为高低值，均值作为中值。

代码解读9

! Predictor variable - X
! Mediator variable(s) - M1, M2
! Moderator variable(s) - W
! Outcome variable - Y
USEVARIABLES = X M1 M2 W Y XW M2W;
! Note that it has to be placed at end of USEVARIABLES subcommand above
DEFINE:

XW = X*W;
M2W = M2*W;
ANALYSIS:

TYPE = GENERAL;
ESTIMATOR = ML;
BOOTSTRAP = 10000;
! In model statement name each path using parentheses
MODEL:

Y ON M1 (b1);
Y ON M2 (b2);
Y ON M2W (b3);
Y ON X (cdash1);
Y ON W (cdash2);
Y ON XW (cdash3);
M1 ON X (a1);
M2 ON X (a2);
M2 ON M1 (d1);
! Use model constraint subcommand to test simple slopes
! You need to pick low, medium and high moderator values,
! for example, of 1 SD below mean, mean, 1 SD above mean
! Also calc total effects at lo, med, hi values of moderator
MODEL CONSTRAINT:

NEW(LOW_W MED_W HIGH_W a1b1
LWa2b2 MWa2b2 HWa2b2
LWa1d1b2 MWa1d1b2 HWa1d1b2
IMM_A IMM_B
DIR_LW DIR_MW DIR_HW
TOT_LOWW TOT_MEDW TOT_HIW);

MODEL CONSTRAINT 命令块用于定义模型中的约束，例如计算特定效应。NEW 命令声明了一系列新的变量，如 LOW_W、 MED_W 和 HIGH_W，分别代表 W 的低、中、高值，以及其他用于计算的中间变量。

代码解读10

! Predictor variable - X
! Mediator variable(s) - M1, M2
! Moderator variable(s) - W
! Outcome variable - Y
USEVARIABLES = X M1 M2 W Y XW M2W;
! Note that it has to be placed at end of USEVARIABLES subcommand above
DEFINE:

XW = X*W;
M2W = M2*W;
ANALYSIS:

TYPE = GENERAL;
ESTIMATOR = ML;
BOOTSTRAP = 10000;
! In model statement name each path using parentheses
MODEL:

Y ON M1 (b1);
Y ON M2 (b2);
Y ON M2W (b3);
Y ON X (cdash1);
Y ON W (cdash2);
Y ON XW (cdash3);
M1 ON X (a1);
M2 ON X (a2);
M2 ON M1 (d1);
! Use model constraint subcommand to test simple slopes
! You need to pick low, medium and high moderator values,
! for example, of 1 SD below mean, mean, 1 SD above mean
! Also calc total effects at lo, med, hi values of moderator
MODEL CONSTRAINT:

NEW(LOW_W MED_W HIGH_W a1b1
LWa2b2 MWa2b2 HWa2b2
LWa1d1b2 MWa1d1b2 HWa1d1b2
IMM_A IMM_B
DIR_LW DIR_MW DIR_HW
TOT_LOWW TOT_MEDW TOT_HIW);
LOW_W = #LOWW;
! replace #LOWW in the code with your chosen low value of W

MED_W = #MEDW;
! replace #MEDW in the code with your chosen medium value of W

HIGH_W = #HIGHW;
! replace #HIGHW in the code with your chosen high value of W

这里是将 LOW_W、 MED_W 和 HIGH_W 赋值为预先选择的 W 的低、中、高值。 请注意，需要用户将代码中的 #LOWW、#MEDW 和 #HIGHW 替换成实际的数值。

代码解读11

! Predictor variable - X
! Mediator variable(s) - M1, M2
! Moderator variable(s) - W
! Outcome variable - Y
USEVARIABLES = X M1 M2 W Y XW M2W;
! Note that it has to be placed at end of USEVARIABLES subcommand above
DEFINE:

XW = X*W;
M2W = M2*W;
ANALYSIS:

TYPE = GENERAL;
ESTIMATOR = ML;
BOOTSTRAP = 10000;
! In model statement name each path using parentheses
MODEL:

Y ON M1 (b1);
Y ON M2 (b2);
Y ON M2W (b3);
Y ON X (cdash1);
Y ON W (cdash2);
Y ON XW (cdash3);
M1 ON X (a1);
M2 ON X (a2);
M2 ON M1 (d1);
! Use model constraint subcommand to test simple slopes
! You need to pick low, medium and high moderator values,
! for example, of 1 SD below mean, mean, 1 SD above mean
! Also calc total effects at lo, med, hi values of moderator
MODEL CONSTRAINT:

NEW(LOW_W MED_W HIGH_W a1b1
LWa2b2 MWa2b2 HWa2b2
LWa1d1b2 MWa1d1b2 HWa1d1b2
IMM_A IMM_B
DIR_LW DIR_MW DIR_HW
TOT_LOWW TOT_MEDW TOT_HIW);
LOW_W = #LOWW;
! replace #LOWW in the code with your chosen low value of W

MED_W = #MEDW;
! replace #MEDW in the code with your chosen medium value of W

HIGH_W = #HIGHW;
! replace #HIGHW in the code with your chosen high value of W
! Now calc indirect and total effects for each value of W

a1b1 = a1*b1;

这里开始计算不同 W 值下的间接效应和总效应。a1b1 = a1*b1 计算 X 通过 M1 对 Y 的间接效应。

代码解读12

! Predictor variable - X
! Mediator variable(s) - M1, M2
! Moderator variable(s) - W
! Outcome variable - Y
USEVARIABLES = X M1 M2 W Y XW M2W;
! Note that it has to be placed at end of USEVARIABLES subcommand above
DEFINE:

XW = X*W;
M2W = M2*W;
ANALYSIS:

TYPE = GENERAL;
ESTIMATOR = ML;
BOOTSTRAP = 10000;
! In model statement name each path using parentheses
MODEL:

Y ON M1 (b1);
Y ON M2 (b2);
Y ON M2W (b3);
Y ON X (cdash1);
Y ON W (cdash2);
Y ON XW (cdash3);
M1 ON X (a1);
M2 ON X (a2);
M2 ON M1 (d1);
! Use model constraint subcommand to test simple slopes
! You need to pick low, medium and high moderator values,
! for example, of 1 SD below mean, mean, 1 SD above mean
! Also calc total effects at lo, med, hi values of moderator
MODEL CONSTRAINT:

NEW(LOW_W MED_W HIGH_W a1b1
LWa2b2 MWa2b2 HWa2b2
LWa1d1b2 MWa1d1b2 HWa1d1b2
IMM_A IMM_B
DIR_LW DIR_MW DIR_HW
TOT_LOWW TOT_MEDW TOT_HIW);
LOW_W = #LOWW;
! replace #LOWW in the code with your chosen low value of W

MED_W = #MEDW;
! replace #MEDW in the code with your chosen medium value of W

HIGH_W = #HIGHW;
! replace #HIGHW in the code with your chosen high value of W
! Now calc indirect and total effects for each value of W

a1b1 = a1*b1;
LWa2b2 = a2*b2 + a2*b3*LOW_W;
MWa2b2 = a2*b2 + a2*b3*MED_W;
HWa2b2 = a2*b2 + a2*b3*HIGH_W;
! Conditional indirect effects of X on Y via M2 only given values of W

这部分计算在 W 的低、中、高值下，X 通过 M2 对 Y 的条件间接效应。

代码解读13

! Predictor variable - X
! Mediator variable(s) - M1, M2
! Moderator variable(s) - W
! Outcome variable - Y
USEVARIABLES = X M1 M2 W Y XW M2W;
! Note that it has to be placed at end of USEVARIABLES subcommand above
DEFINE:

XW = X*W;
M2W = M2*W;
ANALYSIS:

TYPE = GENERAL;
ESTIMATOR = ML;
BOOTSTRAP = 10000;
! In model statement name each path using parentheses
MODEL:

Y ON M1 (b1);
Y ON M2 (b2);
Y ON M2W (b3);
Y ON X (cdash1);
Y ON W (cdash2);
Y ON XW (cdash3);
M1 ON X (a1);
M2 ON X (a2);
M2 ON M1 (d1);
! Use model constraint subcommand to test simple slopes
! You need to pick low, medium and high moderator values,
! for example, of 1 SD below mean, mean, 1 SD above mean
! Also calc total effects at lo, med, hi values of moderator
MODEL CONSTRAINT:

NEW(LOW_W MED_W HIGH_W a1b1
LWa2b2 MWa2b2 HWa2b2
LWa1d1b2 MWa1d1b2 HWa1d1b2
IMM_A IMM_B
DIR_LW DIR_MW DIR_HW
TOT_LOWW TOT_MEDW TOT_HIW);
LOW_W = #LOWW;
! replace #LOWW in the code with your chosen low value of W

MED_W = #MEDW;
! replace #MEDW in the code with your chosen medium value of W

HIGH_W = #HIGHW;
! replace #HIGHW in the code with your chosen high value of W
! Now calc indirect and total effects for each value of W

a1b1 = a1*b1;
LWa2b2 = a2*b2 + a2*b3*LOW_W;
MWa2b2 = a2*b2 + a2*b3*MED_W;
HWa2b2 = a2*b2 + a2*b3*HIGH_W;
! Conditional indirect effects of X on Y via M2 only given values of W
LWa1d1b2 = a1*d1*b2 + a1*d1*b3*LOW_W;
MWa1d1b2 = a1*d1*b2 + a1*d1*b3*MED_W;
HWa1d1b2 = a1*d1*b2 + a1*d1*b3*HIGH_W;
! Conditional indirect effects of X on Y via M1 and M2 given values of W

这部分计算在 W 的低、中、高值下，X 通过 M1 和 M2 对 Y 的条件间接效应。

代码解读14

! Predictor variable - X
! Mediator variable(s) - M1, M2
! Moderator variable(s) - W
! Outcome variable - Y
USEVARIABLES = X M1 M2 W Y XW M2W;
! Note that it has to be placed at end of USEVARIABLES subcommand above
DEFINE:

XW = X*W;
M2W = M2*W;
ANALYSIS:

TYPE = GENERAL;
ESTIMATOR = ML;
BOOTSTRAP = 10000;
! In model statement name each path using parentheses
MODEL:

Y ON M1 (b1);
Y ON M2 (b2);
Y ON M2W (b3);
Y ON X (cdash1);
Y ON W (cdash2);
Y ON XW (cdash3);
M1 ON X (a1);
M2 ON X (a2);
M2 ON M1 (d1);
! Use model constraint subcommand to test simple slopes
! You need to pick low, medium and high moderator values,
! for example, of 1 SD below mean, mean, 1 SD above mean
! Also calc total effects at lo, med, hi values of moderator
MODEL CONSTRAINT:

NEW(LOW_W MED_W HIGH_W a1b1
LWa2b2 MWa2b2 HWa2b2
LWa1d1b2 MWa1d1b2 HWa1d1b2
IMM_A IMM_B
DIR_LW DIR_MW DIR_HW
TOT_LOWW TOT_MEDW TOT_HIW);
LOW_W = #LOWW;
! replace #LOWW in the code with your chosen low value of W

MED_W = #MEDW;
! replace #MEDW in the code with your chosen medium value of W

HIGH_W = #HIGHW;
! replace #HIGHW in the code with your chosen high value of W
! Now calc indirect and total effects for each value of W

a1b1 = a1*b1;
LWa2b2 = a2*b2 + a2*b3*LOW_W;
MWa2b2 = a2*b2 + a2*b3*MED_W;
HWa2b2 = a2*b2 + a2*b3*HIGH_W;
! Conditional indirect effects of X on Y via M2 only given values of W
LWa1d1b2 = a1*d1*b2 + a1*d1*b3*LOW_W;
MWa1d1b2 = a1*d1*b2 + a1*d1*b3*MED_W;
HWa1d1b2 = a1*d1*b2 + a1*d1*b3*HIGH_W;
! Conditional indirect effects of X on Y via M1 and M2 given values of W
! Indices of Moderated Mediation
IMM_A = a2*b3;
IMM_B = a1*d1*b3;

这里计算有调节的中介效应指标。IMM_A = a2*b3 和 IMM_B = a1*d1*b3 分别表示不同的有调节中介效应指标。

代码解读15

! Predictor variable - X
! Mediator variable(s) - M1, M2
! Moderator variable(s) - W
! Outcome variable - Y
USEVARIABLES = X M1 M2 W Y XW M2W;
! Note that it has to be placed at end of USEVARIABLES subcommand above
DEFINE:

XW = X*W;
M2W = M2*W;
ANALYSIS:

TYPE = GENERAL;
ESTIMATOR = ML;
BOOTSTRAP = 10000;
! In model statement name each path using parentheses
MODEL:

Y ON M1 (b1);
Y ON M2 (b2);
Y ON M2W (b3);
Y ON X (cdash1);
Y ON W (cdash2);
Y ON XW (cdash3);
M1 ON X (a1);
M2 ON X (a2);
M2 ON M1 (d1);
! Use model constraint subcommand to test simple slopes
! You need to pick low, medium and high moderator values,
! for example, of 1 SD below mean, mean, 1 SD above mean
! Also calc total effects at lo, med, hi values of moderator
MODEL CONSTRAINT:

NEW(LOW_W MED_W HIGH_W a1b1
LWa2b2 MWa2b2 HWa2b2
LWa1d1b2 MWa1d1b2 HWa1d1b2
IMM_A IMM_B
DIR_LW DIR_MW DIR_HW
TOT_LOWW TOT_MEDW TOT_HIW);
LOW_W = #LOWW;
! replace #LOWW in the code with your chosen low value of W

MED_W = #MEDW;
! replace #MEDW in the code with your chosen medium value of W

HIGH_W = #HIGHW;
! replace #HIGHW in the code with your chosen high value of W
! Now calc indirect and total effects for each value of W

a1b1 = a1*b1;
LWa2b2 = a2*b2 + a2*b3*LOW_W;
MWa2b2 = a2*b2 + a2*b3*MED_W;
HWa2b2 = a2*b2 + a2*b3*HIGH_W;
! Conditional indirect effects of X on Y via M2 only given values of W
LWa1d1b2 = a1*d1*b2 + a1*d1*b3*LOW_W;
MWa1d1b2 = a1*d1*b2 + a1*d1*b3*MED_W;
HWa1d1b2 = a1*d1*b2 + a1*d1*b3*HIGH_W;
! Conditional indirect effects of X on Y via M1 and M2 given values of W
! Indices of Moderated Mediation
IMM_A = a2*b3;
IMM_B = a1*d1*b3;
! Conditional direct effects of X on Y given values of W
DIR_LW = cdash1 + cdash3*LOW_W;
DIR_MW = cdash1 + cdash3*MED_W;;
DIR_HW = cdash1 + cdash3*HIGH_W;;

这部分计算在 W 的低、中、高值下，X 对 Y 的条件直接效应。

代码解读16

! Predictor variable - X
! Mediator variable(s) - M1, M2
! Moderator variable(s) - W
! Outcome variable - Y
USEVARIABLES = X M1 M2 W Y XW M2W;
! Note that it has to be placed at end of USEVARIABLES subcommand above
DEFINE:

XW = X*W;
M2W = M2*W;
ANALYSIS:

TYPE = GENERAL;
ESTIMATOR = ML;
BOOTSTRAP = 10000;
! In model statement name each path using parentheses
MODEL:

Y ON M1 (b1);
Y ON M2 (b2);
Y ON M2W (b3);
Y ON X (cdash1);
Y ON W (cdash2);
Y ON XW (cdash3);
M1 ON X (a1);
M2 ON X (a2);
M2 ON M1 (d1);
! Use model constraint subcommand to test simple slopes
! You need to pick low, medium and high moderator values,
! for example, of 1 SD below mean, mean, 1 SD above mean
! Also calc total effects at lo, med, hi values of moderator
MODEL CONSTRAINT:

NEW(LOW_W MED_W HIGH_W a1b1
LWa2b2 MWa2b2 HWa2b2
LWa1d1b2 MWa1d1b2 HWa1d1b2
IMM_A IMM_B
DIR_LW DIR_MW DIR_HW
TOT_LOWW TOT_MEDW TOT_HIW);
LOW_W = #LOWW;
! replace #LOWW in the code with your chosen low value of W

MED_W = #MEDW;
! replace #MEDW in the code with your chosen medium value of W

HIGH_W = #HIGHW;
! replace #HIGHW in the code with your chosen high value of W
! Now calc indirect and total effects for each value of W

a1b1 = a1*b1;
LWa2b2 = a2*b2 + a2*b3*LOW_W;
MWa2b2 = a2*b2 + a2*b3*MED_W;
HWa2b2 = a2*b2 + a2*b3*HIGH_W;
! Conditional indirect effects of X on Y via M2 only given values of W
LWa1d1b2 = a1*d1*b2 + a1*d1*b3*LOW_W;
MWa1d1b2 = a1*d1*b2 + a1*d1*b3*MED_W;
HWa1d1b2 = a1*d1*b2 + a1*d1*b3*HIGH_W;
! Conditional indirect effects of X on Y via M1 and M2 given values of W
! Indices of Moderated Mediation
IMM_A = a2*b3;
IMM_B = a1*d1*b3;
! Conditional direct effects of X on Y given values of W
DIR_LW = cdash1 + cdash3*LOW_W;
DIR_MW = cdash1 + cdash3*MED_W;;
DIR_HW = cdash1 + cdash3*HIGH_W;;
! Conditional total effects of X on Y given values of W
TOT_LOWW = LWa1d1b2 + LWa2b2 + a1b1 + DIR_LW;
TOT_MEDW = MWa1d1b2 + MWa2b2 + a1b1 + DIR_MW;
TOT_HIW = HWa1d1b2 + HWa2b2 + a1b1 + DIR_HW;

这里计算在 W 的低、中、高值下，X 对 Y 的总效应，即所有直接和间接效应的总和。

代码解读17

! Predictor variable - X
! Mediator variable(s) - M1, M2
! Moderator variable(s) - W
! Outcome variable - Y
USEVARIABLES = X M1 M2 W Y XW M2W;
! Note that it has to be placed at end of USEVARIABLES subcommand above
DEFINE:

XW = X*W;
M2W = M2*W;
ANALYSIS:

TYPE = GENERAL;
ESTIMATOR = ML;
BOOTSTRAP = 10000;
! In model statement name each path using parentheses
MODEL:

Y ON M1 (b1);
Y ON M2 (b2);
Y ON M2W (b3);
Y ON X (cdash1);
Y ON W (cdash2);
Y ON XW (cdash3);
M1 ON X (a1);
M2 ON X (a2);
M2 ON M1 (d1);
! Use model constraint subcommand to test simple slopes
! You need to pick low, medium and high moderator values,
! for example, of 1 SD below mean, mean, 1 SD above mean
! Also calc total effects at lo, med, hi values of moderator
MODEL CONSTRAINT:

NEW(LOW_W MED_W HIGH_W a1b1
LWa2b2 MWa2b2 HWa2b2
LWa1d1b2 MWa1d1b2 HWa1d1b2
IMM_A IMM_B
DIR_LW DIR_MW DIR_HW
TOT_LOWW TOT_MEDW TOT_HIW);
LOW_W = #LOWW;
! replace #LOWW in the code with your chosen low value of W

MED_W = #MEDW;
! replace #MEDW in the code with your chosen medium value of W

HIGH_W = #HIGHW;
! replace #HIGHW in the code with your chosen high value of W
! Now calc indirect and total effects for each value of W

a1b1 = a1*b1;
LWa2b2 = a2*b2 + a2*b3*LOW_W;
MWa2b2 = a2*b2 + a2*b3*MED_W;
HWa2b2 = a2*b2 + a2*b3*HIGH_W;
! Conditional indirect effects of X on Y via M2 only given values of W
LWa1d1b2 = a1*d1*b2 + a1*d1*b3*LOW_W;
MWa1d1b2 = a1*d1*b2 + a1*d1*b3*MED_W;
HWa1d1b2 = a1*d1*b2 + a1*d1*b3*HIGH_W;
! Conditional indirect effects of X on Y via M1 and M2 given values of W
! Indices of Moderated Mediation
IMM_A = a2*b3;
IMM_B = a1*d1*b3;
! Conditional direct effects of X on Y given values of W
DIR_LW = cdash1 + cdash3*LOW_W;
DIR_MW = cdash1 + cdash3*MED_W;;
DIR_HW = cdash1 + cdash3*HIGH_W;;
! Conditional total effects of X on Y given values of W
TOT_LOWW = LWa1d1b2 + LWa2b2 + a1b1 + DIR_LW;
TOT_MEDW = MWa1d1b2 + MWa2b2 + a1b1 + DIR_MW;
TOT_HIW = HWa1d1b2 + HWa2b2 + a1b1 + DIR_HW;
! Use loop plot to plot total effect of X on Y for low, med, high values of W
! NOTE - values of 1,5 in LOOP() statement need to be replaced by
! logical min and max limits of predictor X used in analysis

这里是注释，解释了接下来的代码是用于绘制 X 对 Y 的总效应在不同 W 值下的变化，并提醒 LOOP 语句中的 1 和 5 需要替换成分析中 X 变量的逻辑最小值和最大值。

代码解读18

! Predictor variable - X
! Mediator variable(s) - M1, M2
! Moderator variable(s) - W
! Outcome variable - Y
USEVARIABLES = X M1 M2 W Y XW M2W;
! Note that it has to be placed at end of USEVARIABLES subcommand above
DEFINE:

XW = X*W;
M2W = M2*W;
ANALYSIS:

TYPE = GENERAL;
ESTIMATOR = ML;
BOOTSTRAP = 10000;
! In model statement name each path using parentheses
MODEL:

Y ON M1 (b1);
Y ON M2 (b2);
Y ON M2W (b3);
Y ON X (cdash1);
Y ON W (cdash2);
Y ON XW (cdash3);
M1 ON X (a1);
M2 ON X (a2);
M2 ON M1 (d1);
! Use model constraint subcommand to test simple slopes
! You need to pick low, medium and high moderator values,
! for example, of 1 SD below mean, mean, 1 SD above mean
! Also calc total effects at lo, med, hi values of moderator
MODEL CONSTRAINT:

NEW(LOW_W MED_W HIGH_W a1b1
LWa2b2 MWa2b2 HWa2b2
LWa1d1b2 MWa1d1b2 HWa1d1b2
IMM_A IMM_B
DIR_LW DIR_MW DIR_HW
TOT_LOWW TOT_MEDW TOT_HIW);
LOW_W = #LOWW;
! replace #LOWW in the code with your chosen low value of W

MED_W = #MEDW;
! replace #MEDW in the code with your chosen medium value of W

HIGH_W = #HIGHW;
! replace #HIGHW in the code with your chosen high value of W
! Now calc indirect and total effects for each value of W

a1b1 = a1*b1;
LWa2b2 = a2*b2 + a2*b3*LOW_W;
MWa2b2 = a2*b2 + a2*b3*MED_W;
HWa2b2 = a2*b2 + a2*b3*HIGH_W;
! Conditional indirect effects of X on Y via M2 only given values of W
LWa1d1b2 = a1*d1*b2 + a1*d1*b3*LOW_W;
MWa1d1b2 = a1*d1*b2 + a1*d1*b3*MED_W;
HWa1d1b2 = a1*d1*b2 + a1*d1*b3*HIGH_W;
! Conditional indirect effects of X on Y via M1 and M2 given values of W
! Indices of Moderated Mediation
IMM_A = a2*b3;
IMM_B = a1*d1*b3;
! Conditional direct effects of X on Y given values of W
DIR_LW = cdash1 + cdash3*LOW_W;
DIR_MW = cdash1 + cdash3*MED_W;;
DIR_HW = cdash1 + cdash3*HIGH_W;;
! Conditional total effects of X on Y given values of W
TOT_LOWW = LWa1d1b2 + LWa2b2 + a1b1 + DIR_LW;
TOT_MEDW = MWa1d1b2 + MWa2b2 + a1b1 + DIR_MW;
TOT_HIW = HWa1d1b2 + HWa2b2 + a1b1 + DIR_HW;
! Use loop plot to plot total effect of X on Y for low, med, high values of W
! NOTE - values of 1,5 in LOOP() statement need to be replaced by
! logical min and max limits of predictor X used in analysis
PLOT(LOMOD MEDMOD HIMOD);
LOOP(XVAL,1,5,0.1);
LOMOD = TOT_LOWW*XVAL;
MEDMOD = TOT_MEDW*XVAL;
HIMOD = TOT_HIW*XVAL;

PLOT 命令声明要绘制的变量。LOOP 命令创建一个循环，使得 X 的值 XVAL 从 1 到 5，步长为 0.1。在循环中，计算 W 的低、中、高值下，不同的 X 值对应的 Y 的预测值。

代码解读19

! Predictor variable - X
! Mediator variable(s) - M1, M2
! Moderator variable(s) - W
! Outcome variable - Y
USEVARIABLES = X M1 M2 W Y XW M2W;
! Note that it has to be placed at end of USEVARIABLES subcommand above
DEFINE:

XW = X*W;
M2W = M2*W;
ANALYSIS:

TYPE = GENERAL;
ESTIMATOR = ML;
BOOTSTRAP = 10000;
! In model statement name each path using parentheses
MODEL:

Y ON M1 (b1);
Y ON M2 (b2);
Y ON M2W (b3);
Y ON X (cdash1);
Y ON W (cdash2);
Y ON XW (cdash3);
M1 ON X (a1);
M2 ON X (a2);
M2 ON M1 (d1);
! Use model constraint subcommand to test simple slopes
! You need to pick low, medium and high moderator values,
! for example, of 1 SD below mean, mean, 1 SD above mean
! Also calc total effects at lo, med, hi values of moderator
MODEL CONSTRAINT:

NEW(LOW_W MED_W HIGH_W a1b1
LWa2b2 MWa2b2 HWa2b2
LWa1d1b2 MWa1d1b2 HWa1d1b2
IMM_A IMM_B
DIR_LW DIR_MW DIR_HW
TOT_LOWW TOT_MEDW TOT_HIW);
LOW_W = #LOWW;
! replace #LOWW in the code with your chosen low value of W

MED_W = #MEDW;
! replace #MEDW in the code with your chosen medium value of W

HIGH_W = #HIGHW;
! replace #HIGHW in the code with your chosen high value of W
! Now calc indirect and total effects for each value of W

a1b1 = a1*b1;
LWa2b2 = a2*b2 + a2*b3*LOW_W;
MWa2b2 = a2*b2 + a2*b3*MED_W;
HWa2b2 = a2*b2 + a2*b3*HIGH_W;
! Conditional indirect effects of X on Y via M2 only given values of W
LWa1d1b2 = a1*d1*b2 + a1*d1*b3*LOW_W;
MWa1d1b2 = a1*d1*b2 + a1*d1*b3*MED_W;
HWa1d1b2 = a1*d1*b2 + a1*d1*b3*HIGH_W;
! Conditional indirect effects of X on Y via M1 and M2 given values of W
! Indices of Moderated Mediation
IMM_A = a2*b3;
IMM_B = a1*d1*b3;
! Conditional direct effects of X on Y given values of W
DIR_LW = cdash1 + cdash3*LOW_W;
DIR_MW = cdash1 + cdash3*MED_W;;
DIR_HW = cdash1 + cdash3*HIGH_W;;
! Conditional total effects of X on Y given values of W
TOT_LOWW = LWa1d1b2 + LWa2b2 + a1b1 + DIR_LW;
TOT_MEDW = MWa1d1b2 + MWa2b2 + a1b1 + DIR_MW;
TOT_HIW = HWa1d1b2 + HWa2b2 + a1b1 + DIR_HW;
! Use loop plot to plot total effect of X on Y for low, med, high values of W
! NOTE - values of 1,5 in LOOP() statement need to be replaced by
! logical min and max limits of predictor X used in analysis
PLOT(LOMOD MEDMOD HIMOD);
LOOP(XVAL,1,5,0.1);
LOMOD = TOT_LOWW*XVAL;
MEDMOD = TOT_MEDW*XVAL;
HIMOD = TOT_HIW*XVAL;
PLOT:

TYPE = plot2;

PLOT 命令块设置绘图参数，TYPE = plot2 指定绘图类型为二维线图。

代码解读20

! Predictor variable - X
! Mediator variable(s) - M1, M2
! Moderator variable(s) - W
! Outcome variable - Y
USEVARIABLES = X M1 M2 W Y XW M2W;
! Note that it has to be placed at end of USEVARIABLES subcommand above
DEFINE:

XW = X*W;
M2W = M2*W;
ANALYSIS:

TYPE = GENERAL;
ESTIMATOR = ML;
BOOTSTRAP = 10000;
! In model statement name each path using parentheses
MODEL:

Y ON M1 (b1);
Y ON M2 (b2);
Y ON M2W (b3);
Y ON X (cdash1);
Y ON W (cdash2);
Y ON XW (cdash3);
M1 ON X (a1);
M2 ON X (a2);
M2 ON M1 (d1);
! Use model constraint subcommand to test simple slopes
! You need to pick low, medium and high moderator values,
! for example, of 1 SD below mean, mean, 1 SD above mean
! Also calc total effects at lo, med, hi values of moderator
MODEL CONSTRAINT:

NEW(LOW_W MED_W HIGH_W a1b1
LWa2b2 MWa2b2 HWa2b2
LWa1d1b2 MWa1d1b2 HWa1d1b2
IMM_A IMM_B
DIR_LW DIR_MW DIR_HW
TOT_LOWW TOT_MEDW TOT_HIW);
LOW_W = #LOWW;
! replace #LOWW in the code with your chosen low value of W

MED_W = #MEDW;
! replace #MEDW in the code with your chosen medium value of W

HIGH_W = #HIGHW;
! replace #HIGHW in the code with your chosen high value of W
! Now calc indirect and total effects for each value of W

a1b1 = a1*b1;
LWa2b2 = a2*b2 + a2*b3*LOW_W;
MWa2b2 = a2*b2 + a2*b3*MED_W;
HWa2b2 = a2*b2 + a2*b3*HIGH_W;
! Conditional indirect effects of X on Y via M2 only given values of W
LWa1d1b2 = a1*d1*b2 + a1*d1*b3*LOW_W;
MWa1d1b2 = a1*d1*b2 + a1*d1*b3*MED_W;
HWa1d1b2 = a1*d1*b2 + a1*d1*b3*HIGH_W;
! Conditional indirect effects of X on Y via M1 and M2 given values of W
! Indices of Moderated Mediation
IMM_A = a2*b3;
IMM_B = a1*d1*b3;
! Conditional direct effects of X on Y given values of W
DIR_LW = cdash1 + cdash3*LOW_W;
DIR_MW = cdash1 + cdash3*MED_W;;
DIR_HW = cdash1 + cdash3*HIGH_W;;
! Conditional total effects of X on Y given values of W
TOT_LOWW = LWa1d1b2 + LWa2b2 + a1b1 + DIR_LW;
TOT_MEDW = MWa1d1b2 + MWa2b2 + a1b1 + DIR_MW;
TOT_HIW = HWa1d1b2 + HWa2b2 + a1b1 + DIR_HW;
! Use loop plot to plot total effect of X on Y for low, med, high values of W
! NOTE - values of 1,5 in LOOP() statement need to be replaced by
! logical min and max limits of predictor X used in analysis
PLOT(LOMOD MEDMOD HIMOD);
LOOP(XVAL,1,5,0.1);
LOMOD = TOT_LOWW*XVAL;
MEDMOD = TOT_MEDW*XVAL;
HIMOD = TOT_HIW*XVAL;
PLOT:

TYPE = plot2;
OUTPUT:

STAND CINT(bcbootstrap);

OUTPUT 命令块用于设置输出选项。STAND 表示输出标准化系数，CINT(bcbootstrap) 表示使用 Bootstrap 方法计算置信区间。

资源汇总

• 本视频讲义地址: https://mlln.cn/mplus-model-templates/model90.html
• 图书《MPlus中介调节模型》打包下载: 点击下载
• 图书《MPlus中介调节模型》在线看: 点击查看
• 视频教程: 点击这里打开视频
• Mplus 模型模板教程列表: https://mlln.cn/mplus-model-templates
• 统计咨询: https://wx.zsxq.com/group/88888188828842

感谢大家的收看！ 本节课的资源已全部提供，请大家点击页面上的链接下载或查看。 希望这份资料对大家的学习有所帮助，也欢迎大家点赞收藏，方便日后查阅！