O.2 Report-medmod

Mediation and Moderation

1 Read and Merge

#~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
# Read and Merge
#~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
wd <- "E:/Cinetic idei noi/EXPERIMENTE OGL Frontiers (O.2 & O.0.3 & O.0.2)"
setwd(wd)
DataTrust <- rio::import(file.path(wd, "O.2 REZULTATE\\O.2 Date PrelucrareSPSS cu NEO si STAI", "O.2 Date pt Trust BUN cu NEO si STAIY.xlsx"))
DataDG <- rio::import(file.path(wd, "O.2 REZULTATE\\O.2 Date PrelucrareSPSS cu NEO si STAI", "O.2 Date pt DG BUN cu NEO si STAIY.xlsx"))
DataVAS <- rio::import(file.path(wd, "O.2 REZULTATE\\O.2 VAS,IOS", "O.2 Date PrelucrareSPSS.xlsx"))
DataBIO <- rio::import(file.path(wd, "O.2 BIO", "O.2 Ox si Cortizol.xlsx"))
  
Data_merge1 <- merge(DataVAS, DataTrust)  
Data_merge2 <- merge(Data_merge1, DataDG) 
Data_merge3 <- merge(Data_merge2, DataBIO)
Data <- Data_merge3
test_names <- unique(unlist(lapply(list(DataTrust, DataDG, DataVAS, DataBIO), names)))
merge_names <- names(Data)
if(identical(merge_names[order(merge_names)], test_names[order(test_names)])){    # the order matters in identical()
  cat("**Merge was succesful**")
  rm("Data_merge1", "Data_merge2", "Data_merge3", "DataBIO", "DataDG", "DataTrust", "DataVAS", "test_names", "merge_names")
}else cat("**Merge unsuccesful**") 

Merge was succesful

2 Derive new variables (not used here)

#~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
# Derive new variables
#~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Data$D_VasS_Poz <- Data[, "VasS_postPoz"] - Data[, "VasS_prePoz"] 
Data$D_VasS_Neg <- Data[, "VasS_postNeg"] - Data[, "VasS_preNeg"]
Data$D_VasB_Poz <- Data[, "VasB_postPoz"] - Data[, "VasB_prePoz"] 
Data$D_VasB_Neg <- Data[, "VasB_postNeg"] - Data[, "VasB_preNeg"]
Data$D_IOS_Poz <- Data[, "IOS_postPoz"] - Data[, "IOS_prePoz"] 
Data$D_IOS_Neg <- Data[, "IOS_postNeg"] - Data[, "IOS_preNeg"]

Data$D_Sam1_Poz <- Data[, "Sam1_postPoz"] - Data[, "Sam1_prePoz"] 
Data$D_Sam1_Neg <- Data[, "Sam1_postNeg"] - Data[, "Sam1_preNeg"]
Data$D_Sam2_Poz <- Data[, "Sam2_postPoz"] - Data[, "Sam2_prePoz"] 
Data$D_Sam2_Neg <- Data[, "Sam2_postNeg"] - Data[, "Sam2_preNeg"]
Data$D_Sam3_Poz <- Data[, "Sam3_postPoz"] - Data[, "Sam3_prePoz"] 
Data$D_Sam3_Neg <- Data[, "Sam3_postNeg"] - Data[, "Sam3_preNeg"]

Data$D_DG_Poz <- Data[, "DG_postPozTot"] - Data[, "DG_prePozTot"] 
Data$D_DG_Neg <- Data[, "DG_postNegTot"] - Data[, "DG_preNegTot"]

Data$D_TrustMin_Poz <- Data[, "TrustMinPozPost"] - Data[, "TrustMinPozPre"] 
Data$D_TrustMin_Neg <- Data[, "TrustMinNegPost"] - Data[, "TrustMinNegPre"]
Data$D_TrustTot_Poz <- Data[, "TrustTotPozPost"] - Data[, "TrustTotPozPre"] 
Data$D_TrustTot_Neg <- Data[, "TrustTotNegPost"] - Data[, "TrustTotNegPre"]

Data$D_Cort_Poz <- Data[, "Cort_post_Poz"] - Data[, "Cort_pre_Poz"] 
Data$D_Cort_Neg <- Data[, "Cort_post_Neg"] - Data[, "Cort_pre_Neg"]
Data$D_Ox_Poz <- Data[, "Ox_post_Poz"] - Data[, "Ox_pre_Poz"] 
Data$D_Ox_Neg <- Data[, "Ox_post_Neg"] - Data[, "Ox_pre_Neg"]

2.1 Define Functions

find_med <- function(df, dfp = NULL, num_only = TRUE, verbose = TRUE) {
  count = 0
  mediation_model_list <<- list()
  
  if(num_only == TRUE){
  numeric_cols <- unlist(lapply(df, is.numeric))                                      # get only numeric columns
  df <- df[, numeric_cols]
  }
  
  # permutations for Mediation - Check: factorial(len_names)/factorial(len_names-3)
  names <- colnames(df)
  len_names = length(names)
  
  if(is.null(dfp)){
    dfp <- lapply(1:len_names, function(i){
      tmp <- lapply(1:len_names, function(j){
        tmp <- lapply(1:len_names, function(k){
          if(j != i & k != i & k != j) c(names[i], names[j], names[k])
        })
        do.call(rbind, tmp)
      })
      do.call(rbind, tmp)
    })
    dfp <- do.call(rbind.data.frame, dfp)
    names(dfp) <- paste("var", 1:3, sep = "_")
    dfp[, ] <- lapply(dfp[, ], as.character)    
  } else {
    dfp <- dfp
  }
  
  
  for (row in 1:nrow(dfp)) {                
    
    results <- medmod::med(data = df,                                                  
                          dep = dfp[row, 1], med = dfp[row, 2], pred = dfp[row, 3], 
                          ci = TRUE, label = TRUE, 
                          paths = TRUE, pm = TRUE, 
                          estPlot = FALSE)             
    
    pmed <- as.data.frame(results$med)[1, 8]           # p-value of Indirect Effect
    if(pmed < 0.05 && !is.na(pmed)) {
      count <- count + 1
      if(verbose == TRUE) {
        cat("Mediator Variable:", dfp[row, 2], "| Predictor:", dfp[row, 3], "| Dependent:", dfp[row, 1])
        print(results$med)
        print(results$paths)
      }
      mediation_model_list[["MedEs"]][[paste("model", count, sep = "_")]] <<- as.data.frame(results$med)   # return as list of dataframes
      mediation_model_list[["PathEs"]][[paste("model", count, sep = "_")]] <<- as.data.frame(results$paths)
      mediation_model_list[["Syntax"]][[paste("model", count, sep = "_")]] <<- results$modelSyntax
    }
  }
  cat("\n","Report: ", count, "significant mediations out of", row, "total tries.")
}
find_mod <- function(df, dfp = NULL, num_only = TRUE, verbose = TRUE) {
  count = 0
  moderation_model_list <<- list()
  if(num_only == TRUE){
  numeric_cols <- unlist(lapply(df, is.numeric))                                      # get only numeric columns
  df <- df[, numeric_cols]
  }
  
  # restricted permutations for Moderation - Check: choose(len_names, 3)*3
  names <- colnames(df)
  len_names = length(names)
  
  if(is.null(dfp)){
    dfp <- lapply(1:len_names, function(i){
      tmp <- lapply(1:(len_names-1), function(j){
        tmp <- lapply((j+1):len_names, function(k){
          if(j != i & k != i) c(names[i], names[j], names[k])
        })
        do.call(rbind, tmp)
      })
      do.call(rbind, tmp)
    })
    dfp <- do.call(rbind.data.frame, dfp)
    names(dfp) <- paste("var", 1:3, sep = "_")
    dfp[, ] <- lapply(dfp[, ], as.character)    
  } else {
    dfp <- dfp
  }
  
  for (row in 1:nrow(dfp)) {                
    
    results <- medmod::mod(data = df,                                                   # mod does centering automatically
                          dep = dfp[row, 1], mod = dfp[row, 2], pred = dfp[row, 3], 
                          estMethod = "standard", test = TRUE, 
                          simpleSlopeEst = FALSE, simpleSlopePlot = FALSE)             # when testing use estMethod = 'bootstrap', bootstrap = 500 
    
    pmod <- as.data.frame(results$mod)[3,5]
    if(pmod < 0.05 && !is.na(pmod)) {
      count <- count + 1
        if(verbose == TRUE) {
        cat("Dependent Variable:", dfp[row, 1])
        print(results$mod) 
        }
      moderation_model_list[["Model"]][[paste("model", count, sep = "_")]] <<- as.data.frame(results$mod)   # return as list of dataframes
      moderation_model_list[["Syntax"]][[paste("model", count, sep = "_")]] <<- results$modelSyntax
      
    }
  }
  cat("\n","Report: ", count, "significant moderations out of", row, "total tries.")
}

#######################################################################################################################################
## LCS ANCOVA mediation function (adapted for lavaan from doi: 10.1080/10705511.2016.1274657)
#######################################################################################################################################
library(lavaan)
library(semPlot)
lcs_ancova_med <- function(df, x, y1, y2, m1, m2){
  
  arguments <- as.list(match.call())
  x = eval(arguments$x, df)
  y1 = eval(arguments$y1, df)
  y2 = eval(arguments$y2, df)
  m1 = eval(arguments$m1, df)
  m2 = eval(arguments$m2, df)  
  
  df_mod <- data.frame(x, y1, y2, m1, m2)     # use this df creating part like this: lcs_ancova_med(mtcars, mpg, cyl, disp, hp, drat)
  mod_syntax <- 
  '
  # Defining change in M as a function of M1 and M2
  deltam =~ 1*m2
  deltam ~~ deltam
  deltam ~ 1
  m2 ~ 1*m1
  m2 ~~ 0*m1
  m2 ~~ 0*m2
  m2 ~ 0*1
  m1 ~ 1
  # Defining the change in Y as a function of Y1 and Y2
  deltay =~ 1*y2
  deltay ~~ deltay
  deltay ~ 1
  y2 ~ 1*y1
  y2 ~~ 0*y1
  y2 ~~ 0*y2
  y2 ~ 0*1
  y1 ~ 1
  # Estimating the Pretest correlation between M1 and Y1 and Variance of X
  m1 ~~ y1
  # Estimated covariance between M1 and X and Y1 and X because these covariances may not be equal to zero especially 
  # if X is not a randomized experiment without these the model has 2 degrees of freedom (covariances are only constrained to zero) 
  # but ANCOVA model should start out as saturated and have 0 degrees of freedom 
  m1 ~~ x # these covariances may not be equal to zero especially if X is not a randomized experiment
  x ~~ y1 # these covariances may not be equal to zero especially if X is not a randomized experiment
  # Regression of change in M on X and pretest measures
  deltam ~ am2x*x + sm1*m1 + y1
  # Regression of change in Y on X, change in M, and pretest measures
  deltay ~ x + by2m2*deltam + b*m1 + sy1*y1
  
  # Making constraints to match estimates to ANCOVA
  # Estimate of effect of M1 on M2 in ANCOVA
  sm := sm1+1 
  # Estimate of effect of Y1 on Y2 in ANCOVA
  sy := sy1+1 
  # Estimate of effect of M1 on Y2 in ANCOVA
  by2m1 := b-by2m2 
  # Estimate of mediated effect
  med := am2x*by2m2 
  '  
  
  mod <- lavaan::sem(model = mod_syntax, data = df_mod, test = "bootstrap")  
}
#######################################################################################################################################
# Use - dont run
#######################################################################################################################################
# mod <- lcs_ancova_med(df = df2, x = Cond, y1 = Out1, y2 = Out2, m1 = Med1, m2 = Med2)
# 
# summary(mod, standardized = TRUE)
# semPlot::semPaths(mod, layout = "spring",  nCharNodes = 0, nCharEdges = 0, what = "path", whatLabels = "path", edge.label.cex = 0.8) 

3 Downloadable Table

Data %>% 
  select(-"Nume Prenume") %>%
    DT::datatable(                                  # excel downloadable  DT table
      extensions = 'Buttons',
      options = list(pageLength = 10,
                     scrollX='500px', 
                     dom = 'Bfrtip', 
                     buttons = c('excel', "csv")))

4 Analyses

4.1 Melt Data

Data_med <- Data %>%
  select(-c(1:4,6,7), 
         -c("DG_prePoz1", "DG_prePoz2", "DG_prePoz3", "DG_postPoz1", "DG_postPoz2", "DG_postPoz3",  
           "DG_preNeg1", "DG_preNeg2", "DG_preNeg3" , "DG_postNeg1", "DG_postNeg2", "DG_postNeg3"))
## Data not melted to long
# Data_med %>%
#   find_med()       # not run here to keep report clean
## Data melted to long (Cond = Poz/Neg)
# Data_med_melt <- 
#   Data_med %>%
#     dplyr::rename_all(list(~stringr::str_replace_all(., "Pre|_pre|_Pre", "_pre"))) %>%         # consistent Condition and Time
#     dplyr::rename_all(list(~stringr::str_replace_all(., "Post|_post|_Post", "_post"))) %>%
#     dplyr::rename_all(list(~stringr::str_replace_all(., "Poz|poz|_Poz|_poz", "_poz"))) %>%
#     dplyr::rename_all(list(~stringr::str_replace_all(., "Neg|neg|_Neg|_neg", "_neg"))) %>%
#     dplyr::rename_all(list(~stringr::str_replace_all(., "_neg_pre", "_pre_neg"))) %>%          # consistent ordering in names
#     dplyr::rename_all(list(~stringr::str_replace_all(., "_neg_post", "_post_neg"))) %>%
#     dplyr::rename_all(list(~stringr::str_replace_all(., "_poz_pre", "_pre_poz"))) %>%
#     dplyr::rename_all(list(~stringr::str_replace_all(., "_poz_post", "_post_poz"))) %>%
#     dplyr::rename_all(list(~stringr::str_replace_all(., "Tot", ""))) %>%                       # detele Tot "DG_pre_negTot", "TrustTot_post_poz"
#     gather(Var, Val, -c(1:38)) %>%
#     separate(Var, into = c("Var", "Time", "Condition")) %>%
#     mutate(Time = factor(Time, levels = c("pre", "post"))) %>%
#     mutate(Condition = factor(Condition, levels = c("poz", "neg"))) %>%
#     spread(Var, Val) 
Data_med_melt <- 
  Data_med %>%
    dplyr::rename_all(list(~stringr::str_replace_all(., "Pre|_pre|_Pre", "_pre"))) %>%         # consistent Condition and Time
    dplyr::rename_all(list(~stringr::str_replace_all(., "Post|_post|_Post", "_post"))) %>%
    dplyr::rename_all(list(~stringr::str_replace_all(., "Poz|poz|_Poz|_poz", "_poz"))) %>%
    dplyr::rename_all(list(~stringr::str_replace_all(., "Neg|neg|_Neg|_neg", "_neg"))) %>%
    dplyr::rename_all(list(~stringr::str_replace_all(., "_neg_pre", "_pre_neg"))) %>%          # consistent ordering in names
    dplyr::rename_all(list(~stringr::str_replace_all(., "_neg_post", "_post_neg"))) %>%
    dplyr::rename_all(list(~stringr::str_replace_all(., "_poz_pre", "_pre_poz"))) %>%
    dplyr::rename_all(list(~stringr::str_replace_all(., "_poz_post", "_post_poz"))) %>%
    dplyr::rename_all(list(~stringr::str_replace_all(., "Tot", ""))) %>%                       # detele Tot "DG_pre_negTot", "TrustTot_post_poz"
    gather(Var, Val, -c(1:38)) %>%
    tidyr::separate(Var, into = c("Var", "Time", "Condition")) %>%
    tidyr::unite("Var", c("Var", "Time")) %>%
    mutate(Condition = factor(Condition, levels = c("poz", "neg"))) %>%
    mutate(Condition = dplyr::recode(Condition, "poz" = 1, "neg" = 2)) %>% 
    spread(Var, Val) 

4.2 Search Mediations and Moderations

dfp <- data.frame(
  var1 = colnames(Data_med_melt)[grep("_post", colnames(Data_med_melt))],
  var2 = rep("Condition", 13),
  var3 = colnames(Data_med_melt)[grep("_pre", colnames(Data_med_melt))],
  stringsAsFactors = FALSE
)
find_med(df = Data_med_melt, dfp = dfp, num_only = FALSE)                 # nothing
find_mod(df = Data_med_melt, dfp = dfp, num_only = FALSE)                 # DG_pre:Condition
# medmod::med(data = Data_med_melt,                                                  
#            dep = "Ox_post", med = "Condition", pred = "Ox_pre", 
#            ci = TRUE, label = TRUE, 
#            paths = TRUE, pm = TRUE, 
#            estPlot = FALSE)
# 
# medmod::mod(data = Data_med_melt,                                                   # mod does centering automatically
#             dep = "Ox_post", mod = "Condition", pred = "Ox_pre",  
#             estMethod = "standard", test = TRUE, 
#             simpleSlopeEst = FALSE, simpleSlopePlot = FALSE)

# mediation_model_list$Model   # no models to show
moderation_model_list$Model %>% 
  knitr::kable(digits = 2)

term	est	se	z	p
DG_pre	0.64	0.06	10.50	0.00
Condition	40.39	29.01	1.39	0.16
DG_pre:Condition	-0.36	0.12	-2.92	0.00

4.3 LCS ANCOVA longitudinal mediation

Data_med_melt_transform <- Data_med_melt    # some observed variances are (at least) a factor 1000 times larger than others
Data_med_melt_transform$DG_pre <- Data_med_melt_transform$DG_pre / 100  
Data_med_melt_transform$DG_post <- Data_med_melt_transform$DG_post / 100
mod <- lcs_ancova_med(df = Data_med_melt_transform, x = Condition, y1 = DG_pre, y2 = DG_post, m1 = Ox_pre, m2 = Ox_post)
summary(mod, standardized = TRUE)    # nope

lavaan 0.6-4 ended normally after 47 iterations

  Optimization method                           NLMINB
  Number of free parameters                         20

                                                  Used       Total
  Number of observations                            59          60

  Estimator                                         ML
  Model Fit Test Statistic                       0.000
  Degrees of freedom                                 0
  P-value (Bollen-Stine Bootstrap)               0.644

Parameter Estimates:

  Information                                 Expected
  Information saturated (h1) model          Structured
  Standard Errors                             Standard

Latent Variables:
                   Estimate  Std.Err  z-value  P(>|z|)   Std.lv  Std.all
  deltam =~                                                             
    m2                1.000                               0.629    0.897
  deltay =~                                                             
    y2                1.000                               1.509    0.775

Regressions:
                   Estimate  Std.Err  z-value  P(>|z|)   Std.lv  Std.all
  m2 ~                                                                  
    m1                1.000                               1.000    0.770
  y2 ~                                                                  
    y1                1.000                               1.000    1.227
  deltam ~                                                              
    x       (am2x)    0.023    0.158    0.144    0.885    0.036    0.018
    m1       (sm1)   -0.320    0.146   -2.188    0.029   -0.508   -0.274
    y1               -0.036    0.033   -1.104    0.269   -0.057   -0.137
  deltay ~                                                              
    x                 0.397    0.310    1.280    0.200    0.263    0.131
    deltam  (by22)    0.485    0.256    1.894    0.058    0.202    0.202
    m1         (b)    0.090    0.299    0.303    0.762    0.060    0.032
    y1       (sy1)   -0.352    0.065   -5.427    0.000   -0.233   -0.558

Covariances:
                   Estimate  Std.Err  z-value  P(>|z|)   Std.lv  Std.all
 .m2 ~~                                                                 
    m1                0.000                               0.000      NaN
 .y2 ~~                                                                 
    y1                0.000                               0.000      NaN
  m1 ~~                                                                 
    y1                0.100    0.169    0.596    0.551    0.100    0.078
    x                -0.042    0.036   -1.175    0.240   -0.042   -0.155
  y1 ~~                                                                 
    x                 0.053    0.156    0.341    0.733    0.053    0.044

Intercepts:
                   Estimate  Std.Err  z-value  P(>|z|)   Std.lv  Std.all
   .deltam            0.790    0.359    2.204    0.028    1.256    1.256
   .m2                0.000                               0.000    0.000
    m1                0.776    0.070   11.027    0.000    0.776    1.436
   .deltay            2.190    0.734    2.983    0.003    1.451    1.451
   .y2                0.000                               0.000    0.000
    y1                7.229    0.311   23.227    0.000    7.229    3.024
    x                 1.508    0.065   23.177    0.000    1.508    3.017

Variances:
                   Estimate  Std.Err  z-value  P(>|z|)   Std.lv  Std.all
   .deltam            0.356    0.066    5.431    0.000    0.898    0.898
   .m2                0.000                               0.000    0.000
   .deltay            1.379    0.254    5.431    0.000    0.605    0.605
   .y2                0.000                               0.000    0.000
    m1                0.292    0.054    5.431    0.000    0.292    1.000
    y1                5.715    1.052    5.431    0.000    5.715    1.000
    x                 0.250    0.046    5.431    0.000    0.250    1.000

Defined Parameters:
                   Estimate  Std.Err  z-value  P(>|z|)   Std.lv  Std.all
    sm                0.680    0.146    4.661    0.000    0.492    0.726
    sy                0.648    0.065    9.976    0.000    0.767    0.442
    by2m1            -0.395    0.336   -1.175    0.240   -0.142   -0.170
    med               0.011    0.077    0.144    0.886    0.007    0.004

semPlot::semPaths(mod, layout = "spring",  nCharNodes = 0, nCharEdges = 0, what = "path", whatLabels = "path", edge.label.cex = 0.8) 

# mod <- lcs_ancova_med(df = Data_med_melt, x = Condition, y1 = VasS_pre, y2 = VasS_post, m1 = Ox_pre, m2 = Ox_post)
# summary(mod, standardized = TRUE)    # nope
# 
# mod <- lcs_ancova_med(df = Data_med_melt, x = Condition, y1 = VasB_pre, y2 = VasB_post, m1 = Ox_pre, m2 = Ox_post)
# summary(mod, standardized = TRUE)    # nope
# 
# mod <- lcs_ancova_med(df = Data_med_melt, x = Condition, y1 = IOS_pre, y2 = IOS_post, m1 = Ox_pre, m2 = Ox_post)
# summary(mod, standardized = TRUE)    # nope
# 
# mod <- lcs_ancova_med(df = Data_med_melt, x = Condition, y1 = Trust_pre, y2 = Trust_post, m1 = Ox_pre, m2 = Ox_post)
# summary(mod, standardized = TRUE)    # nope
# 
# 
# 
# mod <- lcs_ancova_med(df = Data_med_melt, x = Condition, y1 = Ox_pre, y2 = Ox_post, m1 = IOS_pre, m2 = IOS_post)
# summary(mod, standardized = TRUE)    # nope
# 
# mod <- lcs_ancova_med(df = Data_med_melt, x = Condition, y1 = Ox_pre, y2 = Ox_post, m1 = VasS_pre, m2 = VasS_post)
# summary(mod, standardized = TRUE)    # nope
# 
# mod <- lcs_ancova_med(df = Data_med_melt, x = Condition, y1 = Ox_pre, y2 = Ox_post, m1 = VasB_pre, m2 = VasB_post)
# summary(mod, standardized = TRUE)    # nope
# 
# 
# 
# mod <- lcs_ancova_med(df = Data_med_melt, x = Condition, y1 = Cort_pre, y2 = Cort_post, m1 = IOS_pre, m2 = IOS_post)
# summary(mod, standardized = TRUE)    # nope
# 
# mod <- lcs_ancova_med(df = Data_med_melt, x = Condition, y1 = Cort_pre, y2 = Cort_post, m1 = VasS_pre, m2 = VasS_post)
# summary(mod, standardized = TRUE)    # nope
# 
# mod <- lcs_ancova_med(df = Data_med_melt, x = Condition, y1 = Cort_pre, y2 = Cort_post, m1 = VasB_pre, m2 = VasB_post)
# summary(mod, standardized = TRUE)    # nope

5 Session Info

R version 3.5.2 (2018-12-20)
Platform: x86_64-w64-mingw32/x64 (64-bit)
Running under: Windows >= 8 x64 (build 9200)

Matrix products: default

locale:
[1] LC_COLLATE=Romanian_Romania.1250  LC_CTYPE=Romanian_Romania.1250    LC_MONETARY=Romanian_Romania.1250 LC_NUMERIC=C                     
[5] LC_TIME=Romanian_Romania.1250    

attached base packages:
[1] stats     graphics  grDevices utils     datasets  methods   base     

other attached packages:
 [1] semPlot_1.1        lavaan_0.6-4       rio_0.5.16         plyr_1.8.4         summarytools_0.9.3 DT_0.5             ggpubr_0.2        
 [8] magrittr_1.5       broom_0.5.1        papaja_0.1.0.9842  psych_1.8.10       forcats_0.3.0      stringr_1.3.1      dplyr_0.7.8       
[15] purrr_0.2.5        readr_1.3.0        tidyr_0.8.2        tibble_1.4.2       ggplot2_3.2.0      tidyverse_1.2.1    pacman_0.5.1      

loaded via a namespace (and not attached):
  [1] readxl_1.1.0         backports_1.1.3      Hmisc_4.1-1          BDgraph_2.53         igraph_1.2.4.1       lazyeval_0.2.1      
  [7] splines_3.5.2        crosstalk_1.0.0      pryr_0.1.4           digest_0.6.18        htmltools_0.3.6      matrixcalc_1.0-3    
 [13] magick_2.0           checkmate_1.8.5      lisrelToR_0.1.4      cluster_2.0.7-1      openxlsx_4.1.0       sna_2.4             
 [19] modelr_0.1.2         matrixStats_0.54.0   jpeg_0.1-8           sem_3.1-9            colorspace_1.3-2     rvest_0.3.2         
 [25] haven_2.1.1          xfun_0.8             tcltk_3.5.2          crayon_1.3.4         RCurl_1.95-4.11      jsonlite_1.6        
 [31] lme4_1.1-19          bindr_0.1.1          survival_2.43-3      glue_1.3.1           gtable_0.2.0         mi_1.0              
 [37] medmod_1.0.0         ggm_2.3              abind_1.4-5          rapportools_1.0      scales_1.0.0         Rcpp_1.0.2          
 [43] xtable_1.8-3         htmlTable_1.12       foreign_0.8-71       Formula_1.2-3        stats4_3.5.2         htmlwidgets_1.3     
 [49] httr_1.4.0           RColorBrewer_1.1-2   acepack_1.4.1        pkgconfig_2.0.2      XML_3.98-1.16        nnet_7.3-12         
 [55] kutils_1.69          later_0.7.5          tidyselect_0.2.5     rlang_0.3.0.1        reshape2_1.4.3       munsell_0.5.0       
 [61] cellranger_1.1.0     tools_3.5.2          jmvcore_0.9.5.2      cli_1.0.1            generics_0.0.2       statnet.common_4.1.4
 [67] evaluate_0.14        fdrtool_1.2.15       arm_1.10-1           yaml_2.2.0           knitr_1.24           zip_1.0.0           
 [73] pander_0.6.3         bindrcpp_0.2.2       glasso_1.10          pbapply_1.3-4        nlme_3.1-137         mime_0.6            
 [79] whisker_0.3-2        xml2_1.2.0           compiler_3.5.2       rstudioapi_0.8       curl_4.0             png_0.1-7           
 [85] huge_1.2.7           pbivnorm_0.6.0       stringi_1.2.4        highr_0.7            qgraph_1.5           rockchalk_1.8.129   
 [91] lattice_0.20-38      Matrix_1.2-15        nloptr_1.2.1         pillar_1.3.1         OpenMx_2.11.5        data.table_1.12.2   
 [97] bitops_1.0-6         corpcor_1.6.9        httpuv_1.4.5         R6_2.4.0             latticeExtra_0.6-28  promises_1.0.1      
[103] network_1.13.0.1     gridExtra_2.3        codetools_0.2-15     boot_1.3-20          MASS_7.3-51.1        gtools_3.8.1        
[109] assertthat_0.2.1     rjson_0.2.20         withr_2.1.2          mnormt_1.5-5         parallel_3.5.2       hms_0.4.2           
[115] grid_3.5.2           rpart_4.1-13         coda_0.19-2          minqa_1.2.4          rmarkdown_1.14       carData_3.0-2       
[121] d3Network_0.5.2.1    semTools_0.5-1       shiny_1.2.0          lubridate_1.7.4      base64enc_0.1-3      ellipse_0.4.1       

 

---

A work by Claudiu Papasteri

claudiu.papasteri@gmail.com