【问题标题】:Student requesting help interpreting Lavaan output Parallel Multiple Mediation学生请求帮助解释 Lavaan 输出并行多重调解
【发布时间】:2022-06-16 01:39:36
【问题描述】:

我正在使用清单变量的并行多重中介模型路径分析来分析我的硕士论文的数据。这是我第一次进行路径分析,我的 PI 对 R 不熟悉,所以我一个人。我希望我能得到一些帮助来解释我的意见并回答在此过程中出现的一些问题

Path diagram of my model

这是我的模型规格代码:

#### PARALLEL MULTIPLE MEDIATION MODEL ####

#Recoding Variables for Parallel Multiple Mediation Model specification
#SDO and Power are independent variables recoded as X1 and X2
impdata$X1 <- impdata$sdo.scores
impdata$X2 <- impdata$pow.scores
#Empathy, Reciprocity, Paternalism, Colourblindess, the mediators, are M1, M2, M3, and M4 respectively
impdata$M1 <- impdata$empath.scores
impdata$M2 <- impdata$recip.scores
impdata$M3 <- impdata$pat.scores
impdata$M4 <- impdata$colour.scores
#Autonomy Support and Talk with Family are dependent variables recoded as Y1 and Y2
impdata$Y1 <- impdata$autonomysupport
impdata$Y2 <- impdata$talkfamily

myModel <- '
#direct effects
Y1 ~ b1 * M1 + b2 * M2 + b3 * M3 + b4 * M4 + c1 * X1 + c3 * X2
Y2 ~ b5 * M1 + b6 * M2 + b7 * M3 + b8 * M4 + c2 * X1 + c4 * X2

#mediators
M1 ~ a1 * X1 + a5 * X2
M2 ~ a2 * X1 + a6 * X2
M3 ~ a3 * X1 + a7 * X2
M4 ~ a4 * X1 + a8 * X2

#indirect effects
indirect1 := a1 * b5
indirect2 := a2 * b6
indirect3 := a3 * b7
indirect4 := a4 * b8
indirect5 := a5 * b1
indirect6 := a6 * b2
indirect7 := a7 * b3
indirect8 := a8 * b4
indirect9 := a1 * b1
indirect10 := a2 * b2
indirect11 := a3 * b3
indirect12 := a4 * b4
indirect13 := a5 * b5
indirect14 := a6 * b6
indirect15 := a7 * b7
indirect16 := a8 * b8

# total effect (C)
#autonomy support
total1 := c1 + (a1 * b1) + (a2 * b2) + (a3 * b3) + (a4 * b4)
total2 := c3 + (a5 * b1) + (a6 * b2) + (a7 * b3) + (a8 * b4)
#communication
total3 := c2 + (a1 * b5) + (a2 * b6) + (a3 * b7) + (a4 * b8)
total4 := c4 + (a5 * b5) + (a6 * b6) + (a7 * b7) + (a8 * b8)

# covariates
M1 ~~ M2
M2 ~~ M3
M2 ~~ M4
M1 ~~ M4
M1 ~~ M3
M3 ~~ M4
'

#bootstap CI's
require("lavaan")
fit <- sem(myModel,
           data= impdata,
           se = "bootstrap",
           bootstrap = 5000)

#obtaining more fit measures
summary(fit, fit.measures=TRUE,
        standardize=TRUE,
        rsquare=TRUE,
        estimates = TRUE,
        ci = TRUE)

拉凡输出 我已将我的输出包含为图像以增加可读性 Model fit indicesRegressions, Covariances, Variances Direct and Indirect Effects

我的问题:

  • 我认为我指定的模型是我的测试模型,但输出将其作为基本模型。我必须先指定基本模型吗?
  • 由于上述原因,我如何报告拟合指数?
  • 据我了解,我的间接(中介)影响都不显着,所以现在我只能报告回归?

任何帮助将不胜感激!谢谢!艾米莉 :)

【问题讨论】:

  • 嗨,欢迎来到 SO。你能解释一下在指定模型后你还尝试了什么吗?你跑sem(model,data)了吗?显示哪种错误消息?
  • @hamagust 感谢您的热烈欢迎 :) 我现在可以通过一些调整来运行我的模型,但现在对解释输出感到困惑。我将更新我的问题以反映当前的代码和输出。据我了解,因为我的间接影响都不显着,所以我只能报告回归。但是,我对拟合指数感到困惑,因为输出将我的测试模型显示为输出模型。任何帮助将不胜感激!

标签: r output r-lavaan structural-equation-model


猜你喜欢
  • 2013-10-09
  • 1970-01-01
  • 1970-01-01
  • 1970-01-01
  • 1970-01-01
  • 2011-03-31
  • 1970-01-01
  • 2021-09-29
  • 1970-01-01
相关资源
最近更新 更多