Consistency and Change in Idiographic Personality: A Longitudinal ESM Network Study Supplementary Code

Left off formatting at the beginning of the consistency section maybe make some sort of reference for the footnotes

Workspace

Packages

library(lavaan)
library(qgraph)
library(igraph)
library(glasso)
library(mlVAR)
library(graphicalVAR)
library(gimme)
library(knitr)
library(kableExtra)
library(gridExtra)
library(Rmisc)
library(psych)
library(stargazer)
library(Matrix)
library(papaja)
library(pander)
library(RColorBrewer)
library(stringr)
library(magrittr)
library(data.table)
library(plyr)
library(tidyverse)

res_path <- "https://github.com/emoriebeck/Idiographic-Network-Consistency"
esm_codebook <- tbl_df(readr::read_csv(sprintf("%s/raw/master/data/Codebook.csv", res_path)))

meanSD_r2z2r <- function(x) {
  z <- fisherz(x)
  z[is.infinite(z)] <- NA
  x_bar <- mean(z, na.rm = T)
  x_sd <- sd(z, na.rm = T)
  r_bar <- fisherz2r(x_bar)
  r_sd <- fisherz2r(x_sd)
  return(c(r_bar, r_sd))
}

Prepare Data

The data include three waves of experience sampling method data from the Personality and Intimate Relationship Study. Data were previously cleaned to remove data points that did not meet inclusion criteria.

Load Data

wave1_all <- readr::read_csv(sprintf("%s/raw/master/data/esm_w1_redacted.csv", res_path)) %>% tbl_df
wave4_all <- readr::read_csv(sprintf("%s/raw/master/data/esm_w4_redacted.csv", res_path)) %>% tbl_df
wave7_all <- readr::read_csv(sprintf("%s/raw/master/data/esm_w7_redacted.csv", res_path)) %>% tbl_df

Clean Data

Because the data sets include data that are not being used in this study, we extract the relevant columns (Subject ID, frequency, hour block, day of study, measurement point, and personality items) from the original data frames. Next, we rename the columns for later ease of use and visualization. Finally, because of the small sample size for waves 4 and 7, we merge those data sets.

old.names <- esm_codebook$old_name
new.names <- esm_codebook$new_name

#Getting necessary columns
#Keeping subject ID and all esm.BFI items
w1 <- wave1_all %>%
  select(one_of(paste(old.names, "w1", sep = "."))) %>%
  setNames(new.names) %>% # change column names
  mutate(wave = "1") 
w4 <- wave4_all %>%
  select(one_of(paste(old.names, "w4", sep = "."))) %>%
  setNames(new.names) %>% # change column names
  mutate(wave = "4")
w7 <- wave7_all %>%
  select(one_of(paste(old.names, "w7", sep = "."))) %>%
  setNames(new.names) %>% # change column names
  mutate(wave = "7")

# merge wave 4 and 7 data sets
w2 <- w4 %>% full_join(w7) 

head(w1)

head(w2)

Variable	New Name	Description
esm.IDnum.w1	SID	numeric variable; identification number
esm.BFI37.w1	A_rude	agreeablness, negative; “During the last hour, how rude were you?” Likert scale from 1 to 5; 1 = Not a lot, 3 = Somewhat, 5 = Very
esm.BFI21.w1	E_quiet	extraversion, negative; “During the last hour, how quiet were you?” Likert scale from 1 to 5; 1 = Not a lot, 3 = Somewhat, 5 = Very
esm.BFI23.w1	C_lazy	conscientiousness, negative; “During the last hour, how lazy were you?” Likert scale from 1 to 5; 1 = Not a lot, 3 = Somewhat, 5 = Very
esm.BFI09.w1	N_relaxed	neuroticism, positive; “During the last hour, how relaxed were you?” Likert scale from 1 to 5; 1 = Not a lot, 3 = Somewhat, 5 = Very
esm.BFI04.w1	N_depressed	neuroticism, positive; “During the last hour, did you feel ‘depressed, blue’?” Likert scale from 1 to 5; 1 = Not a lot, 3 = Somewhat, 5 = Very
esm.BFI36.w1	E_outgoing	extraversion, positive; “During the last hour, how ‘outgoing, sociable’ were you?” Likert scale from 1 to 5; 1 = Not a lot, 3 = Somewhat, 5 = Very
esm.BFI32.w1	A_kind	agreeablness, positive; “During the last hour, how ‘considerate, kind’ were you?” Likert scale from 1 to 5; 1 = Not a lot, 3 = Somewhat, 5 = Very
esm.BFI13.w1	C_reliable	conscientiousness, positive; “During the last hour, how reliable were you?” Likert scale from 1 to 5; 1 = Not a lot, 3 = Somewhat, 5 = Very
esm.BFI19.w1	N_worried	neuroticism, positive; “During the last hour, how worried were you?” Likert scale from 1 to 5; 1 = Not a lot, 3 = Somewhat, 5 = Very

Missing Data Handling

Participants in the study only answered Agreeableness items if they indicated they were interacting with another person during the hour block previous to responding. To retain those measurement points for use in models later, we fill in gaps using within-person means of Agreeabless items.

for (i in unique(w1$SID)){
  mean_A_rude <- mean(w1$A_rude[w1$SID == i], na.rm = T)
  w1$A_rude[is.na(w1$A_rude) & w1$SID == i] <- mean_A_rude
  mean_A_kind <- mean(w1$A_kind[w1$SID == i], na.rm = T)
  w1$A_kind[is.na(w1$A_kind) & w1$SID == i] <- mean_A_kind
}

for (i in unique(w2$SID)){
  mean_A_rude <- mean(w2$A_rude[w2$SID == i], na.rm = T)
  w2$A_rude[is.na(w2$A_rude) & w2$SID == i] <- mean_A_rude
  mean_A_kind <- mean(w2$A_kind[w2$SID == i], na.rm = T)
  w2$A_kind[is.na(w2$A_kind) & w2$SID == i] <- mean_A_kind
}

Response Order & Screening Participants

Because we want to know the precise order of responses of actually collected responses, we make a column with sequential numbering. Then we calculate composites for at item for later use.
To be able to construct individual networks for participants, we ideally need approximately 50 measurement points. However, for current purposes, we will keep all participants who have at least 10 responses, lest we eliminate a large portion of our subjects.

w1 <- w1 %>%
  filter(complete.cases(.)) %>%
  group_by(SID) %>%
  arrange(day, hourBlock) %>%
  mutate(beep_seq = seq(1, n(), 1)) %>%
  group_by(SID) %>%
  mutate_at(vars(A_rude:N_worried), funs(comp = mean)) %>%
  ungroup() 

w2 <- w2 %>%
  filter(complete.cases(.)) %>%
  group_by(SID) %>%
  arrange(wave, day, hourBlock) %>%
  mutate(beep_seq = seq(1, n(), 1)) %>%
  group_by(SID) %>%
  mutate_at(vars(A_rude:N_worried), funs(comp = mean)) %>%
  ungroup() 

jitter_fun <- function(df){
  sd_fun <- function(x){if(sd(x, na.rm = T) == 0 | (length(unique(x)) == 2 & any(table(x) < 2))) jitter(x, amount = runif(1,0,.05)) else x}
  df2 <- data.frame(apply(df, 2, sd_fun))
  colnames(df2) <- colnames(df2)
  return(df2)
}

# filter out people who had < 10 responses and 
#calculate SD's to find people with no variance in responses
w1 <- w1 %>%
  select(SID, wave, beep_seq, A_rude:N_worried) %>%
  group_by(SID) %>%
  mutate_if(is.integer, as.numeric) %>%
  mutate(count = n()) %>%
  filter(count > 10) %>%
  group_by(SID, count, wave) %>% 
  nest() %>%
  mutate(data2 = map(data, jitter_fun)) %>%
  unnest(data2, .drop = T)

w2 <- w2 %>%
  select(SID, beep_seq, A_rude:N_worried) %>%
  group_by(SID) %>%
  mutate_if(is.integer, as.numeric) %>%
  mutate(count = n(), wave = "2") %>%
  filter(count > 10) %>%
  group_by(SID, count, wave) %>% 
  nest() %>%
  mutate(data2 = map(data, jitter_fun)) %>%
  unnest(data2, .drop = T)

head(w1)

head(w2)

Descriptive Statistics of ESM Items (Table 1)

### Table 1 ###
w1 %>% full_join(w2) %>% 
  gather(key = new_name, value = value, A_rude:N_worried, na.rm = T) %>%
  left_join(esm_codebook %>% select(new_name, item_text)) %>%
  group_by(SID, wave, new_name, item_text) %>%
  summarize(mean = mean(value, na.rm = T)) %>%
  group_by(wave, new_name, item_text) %>%
  summarize(M = mean(mean, na.rm = T), 
            SD = sd(mean, na.rm = T)) %>%
  ungroup() %>%
  mutate(item_text = Hmisc::capitalize(str_remove_all(item_text, 
            "During the last hour, "))) %>%
  gather(est, value, M, SD) %>%
  unite(tmp, wave, est, sep = ".") %>%
  spread(tmp, value) %>%
  kable(., "html", escape = F, booktabs = T, digits = 2,
        col.names = c("Variable Name", "Item Text", rep(c("M", "SD"), times = 2)),
        caption = "Item Text and Descriptive Statistics of ESM Items for Waves 1 and 2",
        align = c("l", "r", rep("c", 4))) %>%
  kable_styling(full_width = F) %>%
  add_header_above(c(" " = 2, "Wave 1" = 2, "Wave 2" = 2))

Item Text and Descriptive Statistics of ESM Items for Waves 1 and 2

		Wave 1		Wave 2
Variable Name	Item Text	M	SD	M	SD
A_kind	How ‘considerate, kind’ were you?	3.42	0.60	3.37	0.63
A_rude	How rude were you?	1.46	0.37	1.51	0.40
C_lazy	How lazy were you?	2.48	0.53	2.40	0.56
C_reliable	How reliable were you?	3.69	0.56	3.69	0.55
E_outgoing	How ‘outgoing, sociable’ were you?	2.79	0.50	2.66	0.52
E_quiet	How quiet were you?	3.23	0.44	3.34	0.45
N_depressed	Did you feel ‘depressed, blue’?	1.64	0.53	1.60	0.48
N_relaxed	How relaxed were you?	3.44	0.49	3.33	0.49
N_worried	How worried were you?	2.27	0.60	2.31	0.59

Question 1: Idiographic Network Structure

Although mlVAR includes both population and subject level effects, it represents subject level effects as deviations from population effects rather than exmaning unique subject-level patterns. To assess such unique effects, below we construct individual networks for all subjects at each wave.

Run Models

For idiographic networks, we estimate a Gaussian graphical model (GGM) variation of the vector autoregressive model (VAR), which estimates a partial correlation network in which correlations represent the correlation between variables after conditioning on all other variables. These models are regularized using a variant of the least absolute shrinkage and selection operator (LASSO), graphical LASSO (glasso). In addition, glasso includes a tuning parameter that can be set to control the sparsity of the network. Different values of the parameter can be chosen to optimize prediction accuracy to minimize an information criterion, such as the Bayesian information criterion (BIC) or the extended BIC (EBIC; Chen & Chen, 2008).

Note that a few subjects are manually removed below because the graphicalVAR function is still developmental and crashes R when run on their data.

load(url(sprintf("%s/blob/master/results/graphicalVAR_allvariations.RData?raw=true", res_path)))
# save those subjects to a vector
gVAR_fun <- function(sid, Wave, lambda_scale){
  gamma <- 0
  x <- all_dat %>% filter(SID == sid & wave == Wave) %>% select(A_rude:N_worried)
  n <- nrow(x)
  print(sprintf("%s Wave %s %s", sid, Wave, lambda_scale))
  if(!lambda_scale %in% c("association", "mlVAR_EB")){
    if(lambda_scale == ".25"){lambda <- seq(.025, .25, .025)}
    else if(lambda_scale == ".5"){lambda <- seq(.025, .5, .025)}
    else if(lambda_scale == "unregularized"){lambda = 0}
    else{lambda <- n * .01}
    fit <-
      graphicalVAR(x, gamma = gamma, maxit.in = 1000, maxit.out = 1000,
                        lambda_beta = lambda, lambda_kappa = lambda,
                        verbose = T, scale = F, centerWithin = F)
  } else if (lambda_scale == "association"){
    x_lag <- x %>% mutate_all(funs(lag = lag))
    CC <- cor(x, use = "pairwise")
    DC <- cor(x_lag[,10:18], x_lag[,1:9], use = "pairwise")
    rownames(DC) <- colnames(DC)
    fit <- list(PDC = DC, PCC = CC)
  } else {
    model <- switch(as.numeric(Wave), fit1_w1, fit1_w2)
    sub <- which(model$IDs == sid)
    PDC <- model$results$Beta$subject[[sub]][,,1]
    PCC <- model$results$Theta$pcor$subject[[sub]]
    fit <- list(PDC = PDC, PCC = PCC)
  }
  return(fit)
}

all_dat <- w1 %>% full_join(w2) %>%
  arrange(wave, SID)

gVAR_fit1 <- crossing(
    SID = unique(w1$SID),
    wave = 1,
    lambda_scale = c(".5", ".25", "unregularized", "scaled", "association", "mlVAR_EB")
  ) %>%
full_join(
crossing(
    SID = unique(w2$SID),
    wave = 2,
    lambda_scale = c(".5", ".25", "unregularized", "scaled", "association", "mlVAR_EB")
  )
) %>%
  arrange(SID, wave, lambda_scale) %>%
  filter(!(wave == "2" & SID %in% c("89078", "65365", "20054")) &
         !(wave == "1" & SID %in% c("34960", "27127", "47525"))) %>%
  filter(lambda_scale == "association") %>%
  mutate(gVAR_fit = pmap(list(SID, wave, lambda_scale), possibly(gVAR_fun, NA_real_)))
  
# save(gVAR_fit, file = sprintf("%s/results/graphicalVAR_allVariations.RData", res_path))

GIMME

First, we have to create individual files for each subject, per the procedure to run GIMME.

dir.create(sprintf("%s/results/GIMME", res_path))
dir.create(sprintf("%s/results/GIMME/wave_1_data", res_path))
dir.create(sprintf("%s/results/GIMME/wave_2_data", res_path))

jitter_fun <- function(df){
  u_fun <- function(x){if(length(unique(x)) == 1 | (length(unique(x)) == 2 & any(table(x) < 2))) jitter(x, amount = runif(1,0,.05)) else x}
  df2 <- data.frame(apply(df, 2, u_fun))
  colnames(df2) <- colnames(df2)
  return(df2)
}

# Create separate files for each individual 
write_file_fun <- function(x, sid, Wave){
  x <- x %>% select(A_rude:N_worried)
  write.csv(x, file = sprintf("%s/results/GIMME/wave_%s_data/%s.csv", res_path, Wave, sid), row.names = F)
}

w1 %>% full_join(w2) %>% 
  group_by(SID, wave) %>% 
  nest() %>% 
  filter(!(SID %in% c(1996, 1119, 2650, 28690, 42243, 43460,
                     52374, 56105, 60148, 60311, 8608, 10981,
                     94389, 77407, 76636, 59057, 38258, 42816,
                     81362, 85239, 95555) & wave == 1) | 
           !(SID %in% c(17656, 29431, 40883, 4481, 50556,
                        60759, 6777, 76636, 77453, 79375,
                        85741) & wave == 2)) %>%
  mutate(data = map(data, jitter_fun), 
         write = pmap(list(data, SID, wave), write_file_fun))

Now, we can run the procedure. The results have to be saved to files in a separate directory.

# create results folders
dir.create(sprintf("%s/results/GIMME/wave_1_results", res_path))
dir.create(sprintf("%s/results/GIMME/wave_2_results", res_path))

gimme_w1 <- gimme(         # can use "gimme" or "gimmeSEM"
  data = sprintf("%s/results/GIMME/wave_1_data", res_path),  # source directory where your data are 
  out = sprintf("%s/results/GIMME/wave_1_results", res_path),# output directory where you'd like your output to go
  sep = ",",           # how data are separated. "" for space; "," for comma, "/t" for tab-delimited
  header = TRUE,          # TRUE or FALSE, is there a header
  ar = TRUE,          # TRUE (default) or FALSE, start with autoregressive paths open
  plot = TRUE,        # TRUE (default) or FALSE, generate plots
  subgroup = FALSE,   # TRUE or FALSE (default), cluster individuals based on similarities in effects
  paths = NULL,       # option to list paths that will be group-level (semi-confirmatory)
  groupcutoff = .75,  # the proportion that is considered the majority at the group level
  subcutoff = .5      # the proportion that is considered the majority at the subgroup level
)     

gimme_w2 <- gimme(         # can use "gimme" or "gimmeSEM"
  data = sprintf("%s/results/GIMME/wave_2_data", res_path),  # source directory where your data are 
  out = sprintf("%s/results/GIMME/wave_2_results", res_path),# output directory where you'd like your output to go
  sep = ",",           # how data are separated. "" for space; "," for comma, "/t" for tab-delimited
  header = TRUE,          # TRUE or FALSE, is there a header
  ar = TRUE,          # TRUE (default) or FALSE, start with autoregressive paths open
  plot = TRUE,        # TRUE (default) or FALSE, generate plots
  subgroup = FALSE,   # TRUE or FALSE (default), cluster individuals based on similarities in effects
  paths = NULL,       # option to list paths that will be group-level (semi-confirmatory)
  groupcutoff = .75,  # the proportion that is considered the majority at the group level
  subcutoff = .5      # the proportion that is considered the majority at the subgroup level
)     

load(url(sprintf("%s/blob/master/results/gimme.RData?raw=true", res_path)))

gimme_fun <- function(fit){
  PCC <- fit[,!grepl("lag", colnames(fit))]
  PCC <- apply(simplify2array(list(PCC, t(PCC))), 1:2, mean)
  diag(PCC) <- 1
  fit <- list(PDC = fit[,grepl("lag", colnames(fit))],
              PCC = PCC)
  colnames(fit$PDC) <- colnames(fit$PCC)
  fit
}

gVAR_fit <- gVAR_fit %>%
  bind_rows(
  tibble(SID = as.numeric(names(gimme_w1$path_est_mats)), 
       gVAR_fit = gimme_w1$path_est_mats, 
       wave = 1) %>%
  bind_rows(tibble(
       SID = as.numeric(names(gimme_w2$path_est_mats)), 
       gVAR_fit = gimme_w2$path_est_mats,
       wave = 2)) %>%
  mutate(lambda_scale = "gimme",
         gVAR_fit = map(gVAR_fit, gimme_fun))
  )

Edge Weights: What Is the Structure of Idiographic Personality Networks?

# short function for extracting lagged results and saving to a long format df
temp_fun <- function(fit, SID){
  PDC <- fit$PDC
  from <- row.names(PDC)
  PDC.long <- tbl_df(PDC) %>%
    mutate(from = from, type = "Lagged") %>%
    gather(key = to, value = weight, A_rude:N_worried)
}

# short fun for extracting contemporaneous matrix
contemp_mat_fun <- function(fit){fit$PCC}

# short fun for extracting contemporaneous results and saving to long format df
contemp_long_fun <- function(fit){
  PCC <- fit$PCC
  PCC <- PCC[,order(colnames(PCC))]
  PCC <- PCC[order(rownames(PCC)),]
  PCC[lower.tri(PCC, diag = T)] <- NA
  vars <- rownames(PCC)
  PCC.long <- tbl_df(PCC) %>%
    mutate(Var1 = vars,
           type = "Contemporaneous") %>%
    gather(key = Var2, value = weight, A_kind:N_worried) %>%
    filter(!is.na(weight)) %>%
    unite(var, Var1, Var2, sep = ".", remove = F)
}

Cross-Sample Distributions and Descriptives

Lagged

# load idiographic networks #
load(url(sprintf("%s/raw/master/results/graphicalVAR_allvariations.RData?raw=true", res_path)))

# run functions for extraction
gVAR_fit <- gVAR_fit %>%
  filter(!is.na(gVAR_fit)) %>%
  mutate(temp = map2(gVAR_fit, SID, possibly(temp_fun, NA_real_)),
         contemp_mat = map(gVAR_fit, possibly(contemp_mat_fun, NA_real_)),
         contemp = map(gVAR_fit, possibly(contemp_long_fun, NA_real_)))

# Lagged: Partial Directed Correlations
temp_long <- gVAR_fit %>%
  unnest(temp, .drop = T) %>%
  mutate(type = "Lagged")

# grab and print averages
temp_long %>%
  group_by(wave, lambda_scale, from, to) %>%
  summarise_at(vars(weight), funs(M = fisherz2r(mean(fisherz(.), na.rm = T)), 
                             sd(., na.rm = T), kurtosi, skew)) %>%
  gather(est, value, M:skew) %>%
  unite(tmp, wave, est, sep = ".") %>%
  # arrange(desc(r)) %>%
  spread(key = tmp, value = value) %>%
  kable(., "html", booktabs = T, longtable = T, digits = 2,
        caption = "Descriptives of Lagged Edge Weights",
        col.names = c("Network Type", "From", "To", rep(c("Kurtosi", "M", "SD", "Skew"),2))) %>%
  kable_styling(bootstrap_options = c("striped"),full_width = T) %>%
  add_header_above(c(" " = 3, "Wave 1" = 4, "Wave 2" = 4)) %>%
  scroll_box(width = "750px", height = "400px")

Descriptives of Lagged Edge Weights

			Wave 1				Wave 2
Network Type	From	To	Kurtosi	M	SD	Skew	Kurtosi	M	SD	Skew
.25	A_kind	A_kind	10.97	-0.01	0.16	0.05	7.06	-0.02	0.16	-1.90
.25	A_kind	A_rude	9.43	0.06	0.21	0.88	7.28	0.07	0.23	-0.02
.25	A_kind	C_lazy	17.73	0.00	0.16	0.00	10.65	0.10	0.20	2.77
.25	A_kind	C_reliable	15.79	0.02	0.15	1.90	8.49	-0.02	0.22	0.08
.25	A_kind	E_outgoing	19.21	-0.07	0.17	-2.04	11.56	-0.04	0.19	-2.23
.25	A_kind	E_quiet	18.10	0.01	0.16	0.72	14.50	0.07	0.16	2.89
.25	A_kind	N_depressed	10.28	-0.04	0.20	-1.08	6.18	0.05	0.27	0.21
.25	A_kind	N_relaxed	21.55	-0.01	0.15	-0.47	12.57	-0.06	0.20	0.08
.25	A_kind	N_worried	16.96	0.02	0.17	0.87	11.67	0.03	0.19	0.50
.25	A_rude	A_kind	14.13	-0.01	0.14	0.00	9.41	0.02	0.20	0.26
.25	A_rude	A_rude	8.93	-0.01	0.16	-0.44	9.07	0.00	0.20	0.02
.25	A_rude	C_lazy	24.70	0.00	0.13	-0.27	10.81	-0.02	0.20	-0.61
.25	A_rude	C_reliable	26.44	0.00	0.13	-0.71	11.74	0.02	0.17	1.60
.25	A_rude	E_outgoing	43.35	0.01	0.12	1.28	15.79	0.02	0.18	0.77
.25	A_rude	E_quiet	26.39	0.00	0.13	0.60	18.68	0.01	0.13	1.27
.25	A_rude	N_depressed	17.19	0.01	0.15	0.47	13.10	0.05	0.19	1.38
.25	A_rude	N_relaxed	19.67	-0.01	0.09	-0.19	18.62	-0.02	0.17	-0.53
.25	A_rude	N_worried	27.26	0.00	0.13	-0.64	14.50	-0.09	0.18	-1.62
.25	C_lazy	A_kind	11.34	0.04	0.18	1.59	11.32	-0.02	0.21	-1.03
.25	C_lazy	A_rude	7.33	-0.08	0.22	-0.54	6.96	0.02	0.24	1.11
.25	C_lazy	C_lazy	13.66	0.01	0.13	0.98	9.74	0.00	0.17	-0.65
.25	C_lazy	C_reliable	15.01	0.03	0.15	1.11	6.99	0.00	0.20	0.50
.25	C_lazy	E_outgoing	17.46	0.06	0.17	2.04	13.40	0.00	0.19	0.13
.25	C_lazy	E_quiet	21.76	-0.04	0.15	-1.26	14.97	-0.03	0.18	-1.73
.25	C_lazy	N_depressed	8.62	0.03	0.20	0.63	7.99	0.02	0.24	0.39
.25	C_lazy	N_relaxed	14.59	0.02	0.16	0.51	10.68	0.00	0.22	-0.14
.25	C_lazy	N_worried	13.40	-0.02	0.18	-0.91	9.30	0.16	0.22	1.64
.25	C_reliable	A_kind	9.06	0.02	0.17	0.75	11.53	-0.01	0.20	-0.87
.25	C_reliable	A_rude	8.17	0.00	0.21	-0.68	9.20	-0.10	0.21	-2.19
.25	C_reliable	C_lazy	16.25	0.00	0.14	0.23	8.47	-0.05	0.21	-1.68
.25	C_reliable	C_reliable	11.21	0.00	0.15	-0.34	6.41	0.01	0.21	0.24
.25	C_reliable	E_outgoing	19.68	0.01	0.16	-0.91	24.97	-0.01	0.14	-1.04
.25	C_reliable	E_quiet	17.47	0.00	0.15	0.66	19.28	-0.02	0.16	-0.93
.25	C_reliable	N_depressed	11.32	-0.03	0.19	0.05	9.42	-0.01	0.21	0.84
.25	C_reliable	N_relaxed	15.98	0.03	0.14	1.51	14.33	0.09	0.21	0.90
.25	C_reliable	N_worried	11.02	-0.02	0.16	-1.34	11.87	-0.13	0.22	-0.17
.25	E_outgoing	A_kind	10.65	0.02	0.17	0.47	8.15	0.06	0.21	0.87
.25	E_outgoing	A_rude	6.49	-0.10	0.22	-0.84	5.71	-0.08	0.24	-0.08
.25	E_outgoing	C_lazy	17.95	0.00	0.15	0.37	10.54	0.00	0.22	-0.46
.25	E_outgoing	C_reliable	13.46	-0.02	0.16	-1.65	6.96	-0.05	0.23	-0.61
.25	E_outgoing	E_outgoing	15.32	0.02	0.15	1.34	14.05	0.00	0.17	-0.97
.25	E_outgoing	E_quiet	17.73	-0.01	0.14	-0.89	19.38	0.00	0.17	-0.72
.25	E_outgoing	N_depressed	10.90	-0.02	0.18	0.24	6.70	0.19	0.25	1.16
.25	E_outgoing	N_relaxed	13.44	0.00	0.16	0.09	9.21	0.04	0.21	0.72
.25	E_outgoing	N_worried	10.55	0.01	0.17	1.09	7.97	0.17	0.24	0.18
.25	E_quiet	A_kind	8.93	0.02	0.18	0.70	12.26	-0.01	0.19	0.92
.25	E_quiet	A_rude	8.54	-0.07	0.20	-0.83	5.93	0.03	0.25	0.42
.25	E_quiet	C_lazy	14.52	0.00	0.15	-0.31	8.66	0.04	0.24	0.25
.25	E_quiet	C_reliable	15.59	-0.01	0.16	-1.60	7.30	-0.01	0.24	0.12
.25	E_quiet	E_outgoing	20.94	0.04	0.16	2.23	11.70	0.02	0.19	0.92
.25	E_quiet	E_quiet	15.35	-0.02	0.14	-1.79	12.04	-0.01	0.16	-2.09
.25	E_quiet	N_depressed	13.37	0.00	0.18	1.20	6.54	0.10	0.26	0.55
.25	E_quiet	N_relaxed	12.15	0.02	0.14	1.73	13.00	0.03	0.21	1.73
.25	E_quiet	N_worried	10.11	0.03	0.17	1.88	8.83	0.18	0.23	0.88
.25	N_depressed	A_kind	15.86	0.00	0.17	0.59	12.83	0.00	0.15	1.13
.25	N_depressed	A_rude	8.05	0.05	0.22	0.46	11.50	0.01	0.19	-0.26
.25	N_depressed	C_lazy	17.24	-0.01	0.15	-1.79	12.17	-0.02	0.18	-1.25
.25	N_depressed	C_reliable	18.39	-0.01	0.15	-0.83	9.42	0.03	0.16	0.57
.25	N_depressed	E_outgoing	22.08	-0.01	0.15	-1.26	17.25	0.03	0.16	2.16
.25	N_depressed	E_quiet	19.70	0.01	0.13	0.98	22.39	-0.02	0.14	-3.48
.25	N_depressed	N_depressed	8.45	0.01	0.18	-1.52	7.00	0.04	0.21	0.85
.25	N_depressed	N_relaxed	20.40	-0.02	0.13	-1.11	19.84	-0.04	0.13	-2.84
.25	N_depressed	N_worried	21.49	0.03	0.13	2.38	18.26	-0.09	0.15	-1.68
.25	N_relaxed	A_kind	12.88	-0.01	0.19	0.10	7.37	0.02	0.21	0.56
.25	N_relaxed	A_rude	6.47	-0.09	0.24	-0.69	6.48	0.05	0.22	0.45
.25	N_relaxed	C_lazy	14.44	0.02	0.16	-0.09	9.22	-0.02	0.21	-0.73
.25	N_relaxed	C_reliable	12.69	0.01	0.18	0.37	9.36	0.03	0.21	0.22
.25	N_relaxed	E_outgoing	18.76	-0.02	0.17	-0.42	12.12	0.02	0.20	1.72
.25	N_relaxed	E_quiet	16.30	0.00	0.16	-0.81	15.27	0.02	0.17	0.25
.25	N_relaxed	N_depressed	7.48	-0.04	0.21	-0.27	7.93	0.08	0.24	1.62
.25	N_relaxed	N_relaxed	11.97	0.00	0.14	0.50	11.45	-0.04	0.17	-2.41
.25	N_relaxed	N_worried	12.10	-0.01	0.16	0.77	9.23	-0.07	0.21	0.54
.25	N_worried	A_kind	13.59	0.00	0.17	1.12	7.07	0.06	0.20	1.65
.25	N_worried	A_rude	7.65	0.02	0.21	-0.57	7.07	0.01	0.23	-0.05
.25	N_worried	C_lazy	12.12	-0.01	0.17	-0.08	9.86	0.03	0.22	0.78
.25	N_worried	C_reliable	16.29	0.00	0.17	1.01	9.71	-0.02	0.20	-0.09
.25	N_worried	E_outgoing	21.97	-0.05	0.16	-1.36	9.96	-0.02	0.21	-0.24
.25	N_worried	E_quiet	18.52	0.03	0.16	1.55	21.13	0.06	0.16	3.98
.25	N_worried	N_depressed	10.52	0.01	0.19	0.48	7.48	0.07	0.23	0.86
.25	N_worried	N_relaxed	13.94	-0.04	0.16	-0.81	9.89	-0.07	0.19	-1.83
.25	N_worried	N_worried	8.91	0.03	0.15	0.94	7.03	0.05	0.19	-0.02
.5	A_kind	A_kind	14.45	-0.01	0.15	0.06	8.57	-0.03	0.15	-2.08
.5	A_kind	A_rude	11.69	0.06	0.20	1.04	8.93	0.07	0.22	0.12
.5	A_kind	C_lazy	20.20	0.00	0.15	0.10	11.34	0.10	0.20	2.87
.5	A_kind	C_reliable	18.06	0.02	0.15	2.05	8.65	-0.02	0.21	0.11
.5	A_kind	E_outgoing	21.57	-0.07	0.16	-2.48	11.68	-0.04	0.19	-2.23
.5	A_kind	E_quiet	19.48	0.01	0.16	0.79	14.73	0.07	0.16	2.92
.5	A_kind	N_depressed	11.42	-0.03	0.20	-1.23	7.42	0.05	0.25	0.46
.5	A_kind	N_relaxed	24.75	-0.01	0.14	-0.45	14.39	-0.05	0.18	0.73
.5	A_kind	N_worried	18.87	0.01	0.16	1.02	12.23	0.03	0.19	0.50
.5	A_rude	A_kind	16.29	0.00	0.13	0.02	9.90	0.03	0.18	1.28
.5	A_rude	A_rude	11.44	-0.01	0.16	-0.54	10.55	0.00	0.19	0.72
.5	A_rude	C_lazy	25.29	0.00	0.13	-0.29	12.29	-0.01	0.19	-0.15
.5	A_rude	C_reliable	30.67	0.00	0.13	-0.68	11.81	0.01	0.15	1.01
.5	A_rude	E_outgoing	44.52	0.01	0.12	1.32	15.92	0.02	0.18	0.81
.5	A_rude	E_quiet	26.77	0.00	0.13	0.62	18.54	0.01	0.13	1.26
.5	A_rude	N_depressed	18.89	0.01	0.14	0.25	15.60	0.01	0.16	0.91
.5	A_rude	N_relaxed	24.20	-0.01	0.09	-0.19	20.99	0.00	0.14	1.10
.5	A_rude	N_worried	31.57	0.00	0.12	-0.53	16.35	-0.09	0.17	-1.70
.5	C_lazy	A_kind	15.59	0.04	0.17	2.03	17.15	-0.05	0.18	-2.87
.5	C_lazy	A_rude	10.43	-0.08	0.20	-0.74	9.38	0.02	0.22	1.18
.5	C_lazy	C_lazy	20.86	0.00	0.12	1.49	12.14	0.00	0.16	-0.60
.5	C_lazy	C_reliable	22.64	0.04	0.13	1.54	7.75	-0.01	0.18	-0.13
.5	C_lazy	E_outgoing	20.47	0.07	0.17	2.30	13.80	0.00	0.19	0.11
.5	C_lazy	E_quiet	25.95	-0.04	0.14	-1.53	16.11	-0.03	0.18	-1.85
.5	C_lazy	N_depressed	10.76	0.02	0.19	0.57	9.99	-0.01	0.21	0.00
.5	C_lazy	N_relaxed	20.42	0.02	0.15	0.68	12.39	0.03	0.20	0.47
.5	C_lazy	N_worried	17.95	-0.02	0.17	-1.18	11.79	0.12	0.20	1.52
.5	C_reliable	A_kind	13.09	0.01	0.16	1.10	12.95	-0.04	0.18	-2.24
.5	C_reliable	A_rude	11.16	0.00	0.20	-0.83	10.40	-0.10	0.20	-2.34
.5	C_reliable	C_lazy	19.22	0.01	0.13	0.20	9.60	-0.05	0.21	-1.85
.5	C_reliable	C_reliable	14.65	0.00	0.14	-0.43	7.76	0.00	0.20	0.16
.5	C_reliable	E_outgoing	21.22	0.01	0.15	-0.94	25.40	-0.01	0.14	-1.06
.5	C_reliable	E_quiet	19.54	0.00	0.14	0.67	19.24	-0.02	0.17	-0.92
.5	C_reliable	N_depressed	13.07	-0.03	0.18	0.04	12.36	-0.04	0.18	0.26
.5	C_reliable	N_relaxed	20.27	0.02	0.13	1.65	18.01	0.06	0.19	0.42
.5	C_reliable	N_worried	14.32	-0.02	0.15	-1.75	14.26	-0.15	0.20	-0.92
.5	E_outgoing	A_kind	14.93	0.01	0.15	0.66	11.97	0.01	0.17	0.05
.5	E_outgoing	A_rude	8.32	-0.09	0.21	-1.09	6.90	-0.13	0.22	-0.56
.5	E_outgoing	C_lazy	23.55	0.00	0.14	0.48	11.72	-0.05	0.20	-1.30
.5	E_outgoing	C_reliable	15.43	-0.02	0.16	-1.87	8.00	0.00	0.20	-0.17
.5	E_outgoing	E_outgoing	17.04	0.01	0.15	1.51	16.71	0.00	0.17	-1.06
.5	E_outgoing	E_quiet	19.30	-0.01	0.14	-0.85	21.62	0.00	0.17	-0.77
.5	E_outgoing	N_depressed	13.25	-0.02	0.17	0.38	8.18	0.11	0.23	1.04
.5	E_outgoing	N_relaxed	16.38	0.00	0.15	0.09	9.93	0.10	0.18	2.05
.5	E_outgoing	N_worried	13.52	0.01	0.15	0.90	9.60	0.11	0.22	-0.27
.5	E_quiet	A_kind	12.75	0.02	0.17	0.92	17.35	0.05	0.16	3.11
.5	E_quiet	A_rude	11.16	-0.07	0.19	-1.11	7.29	-0.01	0.23	0.23
.5	E_quiet	C_lazy	17.79	0.00	0.15	-0.25	10.21	0.00	0.23	-0.15
.5	E_quiet	C_reliable	20.80	-0.01	0.15	-2.08	8.40	0.03	0.22	0.74
.5	E_quiet	E_outgoing	23.65	0.04	0.15	2.48	12.31	0.02	0.19	0.93
.5	E_quiet	E_quiet	17.28	-0.02	0.14	-1.93	14.46	-0.01	0.15	-2.43
.5	E_quiet	N_depressed	15.66	0.00	0.17	1.40	8.10	0.01	0.24	0.32
.5	E_quiet	N_relaxed	17.40	0.02	0.13	2.24	16.69	0.10	0.18	3.49
.5	E_quiet	N_worried	13.53	0.03	0.15	2.34	9.96	0.12	0.22	0.56
.5	N_depressed	A_kind	18.08	0.00	0.16	0.60	14.98	0.00	0.14	1.29
.5	N_depressed	A_rude	9.77	0.05	0.21	0.40	13.55	0.01	0.18	-0.22
.5	N_depressed	C_lazy	19.25	-0.02	0.15	-2.16	12.24	-0.01	0.18	-1.27
.5	N_depressed	C_reliable	25.73	-0.01	0.14	-1.02	10.16	0.02	0.16	0.59
.5	N_depressed	E_outgoing	23.37	-0.01	0.15	-1.37	18.08	0.03	0.16	2.23
.5	N_depressed	E_quiet	20.56	0.01	0.13	1.03	22.97	-0.02	0.14	-3.54
.5	N_depressed	N_depressed	11.33	-0.01	0.16	-1.86	9.02	0.02	0.20	1.03
.5	N_depressed	N_relaxed	23.62	-0.01	0.13	-1.12	22.31	-0.04	0.13	-2.99
.5	N_depressed	N_worried	25.28	0.02	0.13	2.68	19.23	-0.09	0.15	-1.69
.5	N_relaxed	A_kind	17.10	0.00	0.18	0.17	9.52	0.01	0.20	0.65
.5	N_relaxed	A_rude	8.04	-0.09	0.23	-0.80	8.19	0.04	0.21	0.58
.5	N_relaxed	C_lazy	17.00	0.01	0.15	-0.08	10.08	-0.02	0.21	-0.72
.5	N_relaxed	C_reliable	17.62	0.00	0.16	0.41	9.59	0.03	0.21	0.28
.5	N_relaxed	E_outgoing	19.53	-0.02	0.17	-0.42	12.81	0.02	0.20	1.79
.5	N_relaxed	E_quiet	18.10	-0.01	0.15	-0.89	16.10	0.02	0.17	0.25
.5	N_relaxed	N_depressed	8.72	-0.03	0.20	-0.30	10.18	0.08	0.22	1.89
.5	N_relaxed	N_relaxed	14.29	0.00	0.13	0.56	13.23	-0.04	0.16	-2.69
.5	N_relaxed	N_worried	15.22	0.00	0.15	0.97	10.38	-0.07	0.21	0.54
.5	N_worried	A_kind	20.35	0.00	0.15	1.51	12.14	0.05	0.17	1.95
.5	N_worried	A_rude	9.97	0.01	0.20	-0.63	8.38	0.01	0.22	-0.01
.5	N_worried	C_lazy	15.07	0.00	0.16	0.02	10.30	0.03	0.21	0.81
.5	N_worried	C_reliable	20.60	0.00	0.16	1.13	10.38	-0.02	0.20	-0.12
.5	N_worried	E_outgoing	23.08	-0.05	0.16	-1.41	10.22	-0.02	0.20	-0.24
.5	N_worried	E_quiet	20.28	0.03	0.16	1.77	21.37	0.06	0.16	4.03
.5	N_worried	N_depressed	12.84	0.01	0.18	0.67	9.03	0.10	0.20	2.02
.5	N_worried	N_relaxed	18.88	-0.03	0.15	-1.06	14.47	-0.07	0.17	-2.86
.5	N_worried	N_worried	13.43	0.01	0.14	1.46	8.82	0.02	0.19	0.25
association	A_kind	A_kind	-0.08	0.02	0.22	-0.32	0.70	0.03	0.22	-0.49
association	A_kind	A_rude	0.43	-0.01	0.21	-0.25	1.34	-0.02	0.24	-0.26
association	A_kind	C_lazy	0.45	0.00	0.22	-0.08	1.11	0.02	0.19	-0.36
association	A_kind	C_reliable	0.11	0.01	0.21	0.07	0.55	0.03	0.20	-0.29
association	A_kind	E_outgoing	0.56	-0.02	0.22	-0.11	0.08	-0.01	0.23	-0.55
association	A_kind	E_quiet	-0.03	0.01	0.21	0.19	0.59	0.02	0.24	0.10
association	A_kind	N_depressed	0.79	-0.02	0.22	-0.18	-0.09	-0.02	0.21	0.10
association	A_kind	N_relaxed	0.08	0.00	0.22	-0.15	0.77	0.03	0.21	-0.22
association	A_kind	N_worried	0.31	0.01	0.21	0.50	0.54	-0.06	0.22	-0.13
association	A_rude	A_kind	0.26	-0.01	0.21	-0.10	0.74	0.01	0.21	0.19
association	A_rude	A_rude	0.66	0.00	0.21	0.14	1.66	0.02	0.20	0.36
association	A_rude	C_lazy	0.14	0.00	0.21	0.18	0.32	0.03	0.20	-0.15
association	A_rude	C_reliable	0.61	-0.02	0.21	-0.06	0.83	0.01	0.20	0.47
association	A_rude	E_outgoing	0.54	-0.01	0.21	-0.12	0.52	-0.01	0.19	0.12
association	A_rude	E_quiet	0.16	0.00	0.20	-0.08	-0.44	0.01	0.19	0.00
association	A_rude	N_depressed	0.74	0.04	0.20	0.26	0.38	0.01	0.18	0.41
association	A_rude	N_relaxed	0.07	-0.05	0.21	-0.18	-0.39	-0.03	0.20	-0.21
association	A_rude	N_worried	0.05	0.04	0.21	0.23	2.22	0.07	0.21	0.48
association	C_lazy	A_kind	0.58	0.00	0.20	-0.02	0.53	0.00	0.23	-0.07
association	C_lazy	A_rude	0.65	0.00	0.21	0.05	0.70	0.05	0.21	0.01
association	C_lazy	C_lazy	0.01	0.04	0.21	-0.12	0.19	0.02	0.23	-0.03
association	C_lazy	C_reliable	0.25	-0.02	0.21	0.21	0.38	-0.03	0.24	0.37
association	C_lazy	E_outgoing	0.46	-0.01	0.21	0.14	0.20	0.00	0.20	-0.04
association	C_lazy	E_quiet	0.10	0.02	0.20	-0.14	0.68	-0.01	0.20	-0.28
association	C_lazy	N_depressed	0.18	0.01	0.22	0.15	0.09	0.01	0.22	0.33
association	C_lazy	N_relaxed	0.07	0.00	0.22	-0.01	0.91	0.00	0.19	-0.26
association	C_lazy	N_worried	0.04	0.00	0.21	0.16	0.25	0.00	0.21	0.03
association	C_reliable	A_kind	0.31	0.01	0.22	-0.37	1.50	-0.01	0.24	-0.19
association	C_reliable	A_rude	-0.13	-0.02	0.21	-0.14	2.61	-0.04	0.22	-0.46
association	C_reliable	C_lazy	0.31	0.00	0.21	-0.03	0.27	0.00	0.22	-0.09
association	C_reliable	C_reliable	0.68	0.03	0.21	-0.08	0.70	0.05	0.22	-0.21
association	C_reliable	E_outgoing	0.40	-0.01	0.21	-0.12	1.03	0.02	0.21	-0.46
association	C_reliable	E_quiet	-0.43	0.00	0.20	0.13	1.23	-0.01	0.20	-0.25
association	C_reliable	N_depressed	-0.15	-0.02	0.21	-0.13	0.30	-0.03	0.21	0.26
association	C_reliable	N_relaxed	0.55	0.01	0.22	0.01	-0.23	0.04	0.20	0.21
association	C_reliable	N_worried	0.38	-0.02	0.21	-0.09	-0.16	-0.04	0.22	0.06
association	E_outgoing	A_kind	0.23	0.03	0.21	-0.12	-0.19	0.01	0.22	-0.13
association	E_outgoing	A_rude	0.38	-0.01	0.22	0.07	0.97	-0.01	0.21	0.38
association	E_outgoing	C_lazy	0.01	0.02	0.20	-0.03	1.08	0.04	0.20	-0.51
association	E_outgoing	C_reliable	0.29	0.00	0.22	-0.22	0.25	0.00	0.22	0.11
association	E_outgoing	E_outgoing	0.19	0.02	0.21	-0.06	0.57	0.00	0.20	-0.36
association	E_outgoing	E_quiet	0.41	-0.02	0.20	0.00	0.95	-0.01	0.19	0.79
association	E_outgoing	N_depressed	0.34	-0.04	0.20	0.11	0.33	-0.01	0.22	0.49
association	E_outgoing	N_relaxed	1.03	0.03	0.22	-0.32	0.47	0.04	0.21	-0.15
association	E_outgoing	N_worried	0.63	-0.05	0.23	0.21	0.13	-0.05	0.21	0.31
association	E_quiet	A_kind	0.30	0.00	0.21	0.22	0.55	-0.01	0.22	-0.06
association	E_quiet	A_rude	0.31	0.01	0.22	-0.14	0.22	-0.01	0.21	-0.20
association	E_quiet	C_lazy	0.08	-0.02	0.22	-0.12	0.62	-0.02	0.21	0.14
association	E_quiet	C_reliable	0.36	0.00	0.22	0.21	0.78	0.00	0.20	0.38
association	E_quiet	E_outgoing	0.35	-0.01	0.21	0.26	1.65	0.00	0.18	0.73
association	E_quiet	E_quiet	0.15	0.00	0.20	-0.01	0.16	0.02	0.20	-0.23
association	E_quiet	N_depressed	0.95	0.02	0.21	-0.64	-0.06	-0.01	0.22	-0.48
association	E_quiet	N_relaxed	0.89	-0.01	0.22	0.28	0.34	-0.02	0.22	0.25
association	E_quiet	N_worried	0.40	0.04	0.22	-0.22	0.54	0.04	0.20	-0.29
association	N_depressed	A_kind	0.79	0.00	0.21	0.16	0.54	-0.02	0.22	0.08
association	N_depressed	A_rude	1.45	0.00	0.22	0.67	2.95	0.01	0.22	1.00
association	N_depressed	C_lazy	0.59	0.00	0.22	0.06	0.70	-0.02	0.20	0.10
association	N_depressed	C_reliable	1.16	-0.02	0.22	0.13	0.19	-0.02	0.21	0.27
association	N_depressed	E_outgoing	0.24	-0.01	0.22	0.14	0.84	-0.01	0.22	0.47
association	N_depressed	E_quiet	0.47	0.02	0.22	-0.27	0.95	0.02	0.22	-0.16
association	N_depressed	N_depressed	-0.25	0.10	0.24	0.13	0.22	0.12	0.24	0.34
association	N_depressed	N_relaxed	0.47	-0.05	0.21	0.28	0.30	-0.09	0.21	0.09
association	N_depressed	N_worried	-0.60	0.08	0.23	-0.10	-0.39	0.07	0.21	0.00
association	N_relaxed	A_kind	0.07	0.03	0.21	0.01	0.11	0.01	0.22	-0.47
association	N_relaxed	A_rude	0.06	-0.01	0.21	-0.07	0.65	0.02	0.22	0.29
association	N_relaxed	C_lazy	0.44	0.06	0.21	0.07	0.63	0.04	0.21	0.30
association	N_relaxed	C_reliable	0.39	0.01	0.22	-0.10	0.30	0.02	0.20	-0.10
association	N_relaxed	E_outgoing	0.44	0.03	0.21	-0.12	2.35	0.01	0.22	-0.52
association	N_relaxed	E_quiet	0.77	-0.04	0.20	0.25	0.58	0.00	0.23	0.26
association	N_relaxed	N_depressed	-0.10	-0.09	0.21	0.15	-0.13	-0.05	0.21	0.34
association	N_relaxed	N_relaxed	-0.29	0.08	0.22	-0.13	0.12	0.06	0.20	-0.23
association	N_relaxed	N_worried	0.60	-0.12	0.21	0.36	-0.10	-0.06	0.21	0.39
association	N_worried	A_kind	0.57	-0.03	0.22	0.23	-0.02	-0.04	0.22	0.25
association	N_worried	A_rude	0.21	0.00	0.22	0.01	-0.02	0.02	0.22	-0.04
association	N_worried	C_lazy	0.33	-0.06	0.22	-0.06	-0.10	-0.03	0.21	0.10
association	N_worried	C_reliable	0.96	-0.01	0.21	0.12	0.29	0.00	0.21	0.06
association	N_worried	E_outgoing	0.26	-0.05	0.21	0.10	0.85	-0.04	0.22	0.11
association	N_worried	E_quiet	0.14	0.04	0.20	-0.21	0.73	0.05	0.24	0.00
association	N_worried	N_depressed	0.25	0.11	0.23	-0.17	0.66	0.07	0.23	-0.45
association	N_worried	N_relaxed	0.40	-0.12	0.24	0.22	0.75	-0.09	0.22	0.53
association	N_worried	N_worried	0.43	0.16	0.25	-0.33	0.46	0.16	0.23	-0.26
gimme	A_kind	A_kind	5.02	0.00	0.31	-0.89	2.75	0.02	0.30	-1.20
gimme	A_kind	A_rude	82.37	-0.01	0.23	-7.55	11.81	0.04	0.19	2.90
gimme	A_kind	C_lazy	29.85	0.01	0.19	-0.63	11.38	0.00	0.19	0.29
gimme	A_kind	C_reliable	12.52	0.00	0.17	0.95	25.21	0.00	0.19	-1.29
gimme	A_kind	E_outgoing	11.90	0.01	0.14	1.09	33.01	0.03	0.26	-3.40
gimme	A_kind	E_quiet	21.95	-0.02	0.13	-2.61	12.03	0.00	0.20	-0.83
gimme	A_kind	N_depressed	25.35	-0.01	0.17	1.64	26.01	-0.01	0.22	-0.15
gimme	A_kind	N_relaxed	15.01	0.01	0.17	1.90	10.54	-0.01	0.14	-1.17
gimme	A_kind	N_worried	56.81	0.00	0.17	-3.88	8.03	-0.03	0.18	-0.97
gimme	A_rude	A_kind	10.53	0.00	0.17	0.48	16.08	-0.01	0.17	-0.88
gimme	A_rude	A_rude	302.67	0.00	1.85	17.26	0.97	0.01	0.26	0.18
gimme	A_rude	C_lazy	16.36	0.00	0.18	1.29	13.22	0.03	0.17	1.44
gimme	A_rude	C_reliable	39.40	-0.01	0.19	-3.99	10.97	-0.02	0.20	-1.55
gimme	A_rude	E_outgoing	15.37	-0.02	0.17	-2.50	14.38	0.01	0.19	1.56
gimme	A_rude	E_quiet	15.90	-0.02	0.16	-0.15	18.28	-0.01	0.13	-0.98
gimme	A_rude	N_depressed	21.41	0.01	0.21	0.46	11.54	0.01	0.15	0.46
gimme	A_rude	N_relaxed	17.59	-0.02	0.17	-2.19	16.34	0.02	0.12	2.45
gimme	A_rude	N_worried	17.94	0.00	0.16	0.52	17.21	0.04	0.17	3.55
gimme	C_lazy	A_kind	13.32	-0.02	0.18	-0.42	15.62	0.02	0.15	2.33
gimme	C_lazy	A_rude	16.87	0.00	0.15	1.30	18.24	0.02	0.16	3.02
gimme	C_lazy	C_lazy	1.89	0.05	0.28	-0.21	2.66	-0.02	0.30	0.14
gimme	C_lazy	C_reliable	10.41	0.02	0.16	1.59	28.21	0.00	0.20	-2.87
gimme	C_lazy	E_outgoing	20.70	0.00	0.21	-1.07	18.29	-0.02	0.16	-2.22
gimme	C_lazy	E_quiet	22.57	0.00	0.20	-0.86	12.80	-0.01	0.15	-2.37
gimme	C_lazy	N_depressed	10.55	0.02	0.16	1.29	41.59	0.01	0.09	3.28
gimme	C_lazy	N_relaxed	25.25	0.01	0.18	-0.96	16.22	0.04	0.17	1.44
gimme	C_lazy	N_worried	13.32	-0.01	0.19	1.58	17.85	0.00	0.15	-0.23
gimme	C_reliable	A_kind	72.38	0.01	0.20	-4.57	13.26	0.00	0.13	-0.75
gimme	C_reliable	A_rude	22.12	-0.02	0.22	-1.48	17.40	0.01	0.12	1.36
gimme	C_reliable	C_lazy	105.55	-0.01	0.21	7.36	17.73	0.00	0.14	-1.78
gimme	C_reliable	C_reliable	20.61	0.01	0.42	0.52	29.86	0.05	0.33	-3.85
gimme	C_reliable	E_outgoing	56.47	-0.02	0.34	-3.45	17.04	0.01	0.12	1.52
gimme	C_reliable	E_quiet	53.68	0.01	0.29	-4.56	14.51	0.00	0.13	-0.23
gimme	C_reliable	N_depressed	20.78	0.01	0.23	-1.13	11.32	0.01	0.14	1.33
gimme	C_reliable	N_relaxed	49.69	0.02	0.23	-1.28	21.29	0.00	0.15	0.01
gimme	C_reliable	N_worried	15.92	0.00	0.23	-1.64	27.64	0.01	0.17	-0.15
gimme	E_outgoing	A_kind	32.37	-0.02	0.19	-3.80	10.54	0.00	0.14	-1.01
gimme	E_outgoing	A_rude	119.97	-0.01	0.25	8.49	15.85	0.00	0.10	-1.84
gimme	E_outgoing	C_lazy	20.40	-0.01	0.19	-0.31	37.93	-0.04	0.17	1.41
gimme	E_outgoing	C_reliable	25.24	0.00	0.16	2.34	39.90	0.01	0.10	4.37
gimme	E_outgoing	E_outgoing	250.79	0.00	0.86	-14.97	37.03	0.01	0.39	-3.61
gimme	E_outgoing	E_quiet	16.17	-0.01	0.16	-1.12	64.50	0.02	0.23	-5.03
gimme	E_outgoing	N_depressed	25.77	0.01	0.16	3.63	13.32	0.01	0.13	0.89
gimme	E_outgoing	N_relaxed	83.71	-0.01	0.27	7.05	36.79	0.00	0.17	2.48
gimme	E_outgoing	N_worried	281.66	0.00	0.64	-16.35	26.21	0.02	0.13	2.81
gimme	E_quiet	A_kind	16.30	0.01	0.12	0.46	137.03	0.00	23.81	11.75
gimme	E_quiet	A_rude	31.68	0.01	0.15	3.08	136.99	0.00	7.47	11.75
gimme	E_quiet	C_lazy	13.79	0.01	0.14	0.87	137.04	0.00	35.94	-11.75
gimme	E_quiet	C_reliable	14.05	-0.02	0.13	-1.82	137.03	0.01	29.53	-11.75
gimme	E_quiet	E_outgoing	44.09	-0.01	0.13	-3.52	137.03	0.01	22.18	-11.75
gimme	E_quiet	E_quiet	5.77	-0.04	0.31	-1.39	134.14	0.00	2.77	-11.57
gimme	E_quiet	N_depressed	22.34	-0.02	0.16	0.65	16.56	0.01	0.17	2.63
gimme	E_quiet	N_relaxed	12.11	-0.01	0.15	-1.11	137.00	0.01	11.88	11.75
gimme	E_quiet	N_worried	13.98	0.02	0.14	1.42	137.03	-0.01	21.98	-11.75
gimme	N_depressed	A_kind	10.26	-0.01	0.16	-0.27	42.44	-0.01	0.15	3.57
gimme	N_depressed	A_rude	17.45	-0.01	0.16	-2.41	19.55	0.00	0.16	1.86
gimme	N_depressed	C_lazy	30.72	0.01	0.19	3.26	48.65	0.03	0.27	-4.22
gimme	N_depressed	C_reliable	13.07	0.00	0.16	0.90	47.91	0.00	0.13	5.22
gimme	N_depressed	E_outgoing	22.81	-0.01	0.16	-3.26	12.95	0.03	0.15	2.00
gimme	N_depressed	E_quiet	34.11	-0.01	0.13	-3.48	23.14	-0.03	0.12	-4.17
gimme	N_depressed	N_depressed	2.96	0.09	0.31	-0.46	1.57	0.08	0.30	-0.03
gimme	N_depressed	N_relaxed	31.39	-0.01	0.23	-2.97	61.45	0.02	0.18	-5.37
gimme	N_depressed	N_worried	23.54	0.01	0.18	0.69	21.27	0.01	0.21	-2.94
gimme	N_relaxed	A_kind	15.88	0.00	0.14	-0.26	27.65	0.00	0.16	-3.22
gimme	N_relaxed	A_rude	14.55	-0.02	0.12	-2.24	12.42	-0.01	0.10	-1.70
gimme	N_relaxed	C_lazy	13.81	0.00	0.14	-0.54	55.21	-0.01	0.19	5.26
gimme	N_relaxed	C_reliable	20.77	0.00	0.14	0.37	12.07	0.03	0.13	3.35
gimme	N_relaxed	E_outgoing	17.58	0.00	0.15	-0.90	46.90	0.00	0.12	4.50
gimme	N_relaxed	E_quiet	12.52	0.01	0.15	1.38	17.99	0.01	0.17	0.07
gimme	N_relaxed	N_depressed	19.82	-0.01	0.13	-0.89	31.87	0.01	0.25	-3.79
gimme	N_relaxed	N_relaxed	7.76	0.00	0.28	1.05	1.32	0.02	0.24	-0.02
gimme	N_relaxed	N_worried	37.74	-0.02	0.26	3.34	8.96	-0.02	0.15	-1.54
gimme	N_worried	A_kind	29.43	0.00	0.20	-3.10	21.64	-0.01	0.12	-1.01
gimme	N_worried	A_rude	17.51	0.01	0.16	-0.72	16.78	0.00	0.16	-2.22
gimme	N_worried	C_lazy	26.32	0.00	0.18	2.62	15.17	0.02	0.16	1.72
gimme	N_worried	C_reliable	44.45	-0.02	0.23	-3.30	19.51	-0.02	0.15	-1.60
gimme	N_worried	E_outgoing	21.52	0.01	0.19	-0.01	18.65	-0.02	0.19	0.35
gimme	N_worried	E_quiet	15.86	0.00	0.15	-0.18	21.32	-0.02	0.17	0.56
gimme	N_worried	N_depressed	45.06	0.01	0.18	4.28	22.70	0.00	0.14	0.01
gimme	N_worried	N_relaxed	79.93	-0.02	0.27	-7.02	23.60	0.03	0.19	-1.74
gimme	N_worried	N_worried	68.57	0.11	0.50	-5.39	117.23	0.13	1.24	10.46
mlVAR_EB	A_kind	A_kind	0.93	0.03	0.03	0.03	1.89	0.11	0.06	0.81
mlVAR_EB	A_kind	A_rude	2.78	-0.01	0.03	-0.19	2.08	0.00	0.00	-0.50
mlVAR_EB	A_kind	C_lazy	3.04	0.01	0.01	0.03	1.35	0.01	0.03	0.29
mlVAR_EB	A_kind	C_reliable	2.28	0.03	0.01	0.13	1.61	-0.01	0.00	0.06
mlVAR_EB	A_kind	E_outgoing	1.09	0.01	0.00	0.18	6.16	-0.01	0.01	1.32
mlVAR_EB	A_kind	E_quiet	NaN	0.00	0.00	NaN	0.85	0.00	0.01	-0.34
mlVAR_EB	A_kind	N_depressed	2.52	0.03	0.01	-0.49	5.79	-0.04	0.02	-1.13
mlVAR_EB	A_kind	N_relaxed	1.83	0.01	0.01	0.31	0.49	-0.02	0.00	0.10
mlVAR_EB	A_kind	N_worried	2.11	-0.02	0.00	-0.29	2.36	-0.05	0.00	0.44
mlVAR_EB	A_rude	A_kind	4.14	0.00	0.04	-0.55	1.76	0.01	0.05	0.30
mlVAR_EB	A_rude	A_rude	3.76	0.04	0.07	0.30	2.14	0.03	0.06	0.90
mlVAR_EB	A_rude	C_lazy	1.30	-0.01	0.02	-0.27	5.77	0.02	0.03	-0.91
mlVAR_EB	A_rude	C_reliable	2.35	-0.01	0.01	0.07	0.74	-0.01	0.01	0.07
mlVAR_EB	A_rude	E_outgoing	3.23	0.00	0.01	-0.42	NaN	0.00	0.00	NaN
mlVAR_EB	A_rude	E_quiet	1.13	0.01	0.00	0.38	NaN	-0.01	0.00	NaN
mlVAR_EB	A_rude	N_depressed	3.49	-0.02	0.01	0.07	NaN	0.00	0.00	NaN
mlVAR_EB	A_rude	N_relaxed	2.55	0.01	0.00	-0.37	NaN	0.01	0.00	NaN
mlVAR_EB	A_rude	N_worried	2.70	0.01	0.01	0.42	2.42	0.02	0.03	-0.07
mlVAR_EB	C_lazy	A_kind	3.28	-0.03	0.01	-0.64	3.27	0.00	0.06	0.85
mlVAR_EB	C_lazy	A_rude	2.29	0.02	0.03	-0.26	1.51	0.01	0.02	-0.49
mlVAR_EB	C_lazy	C_lazy	2.59	0.04	0.05	0.44	1.06	0.05	0.06	0.18
mlVAR_EB	C_lazy	C_reliable	NaN	0.01	0.00	NaN	NaN	0.02	0.00	NaN
mlVAR_EB	C_lazy	E_outgoing	1.85	0.01	0.02	-0.42	NaN	0.02	0.00	NaN
mlVAR_EB	C_lazy	E_quiet	0.71	0.00	0.00	0.15	NaN	-0.01	0.00	NaN
mlVAR_EB	C_lazy	N_depressed	1.93	0.03	0.02	0.05	2.44	0.04	0.02	0.19
mlVAR_EB	C_lazy	N_relaxed	1.68	0.05	0.00	0.25	3.05	0.01	0.01	0.38
mlVAR_EB	C_lazy	N_worried	2.59	-0.04	0.02	-0.52	2.68	-0.04	0.02	-0.32
mlVAR_EB	C_reliable	A_kind	1.65	0.03	0.03	0.35	2.38	0.04	0.04	0.45
mlVAR_EB	C_reliable	A_rude	1.41	-0.02	0.01	0.05	2.69	0.02	0.02	0.59
mlVAR_EB	C_reliable	C_lazy	1.32	-0.01	0.01	-0.15	3.32	-0.03	0.00	0.10
mlVAR_EB	C_reliable	C_reliable	2.08	0.02	0.03	0.47	1.18	0.05	0.02	0.10
mlVAR_EB	C_reliable	E_outgoing	0.88	0.00	0.00	0.04	2.83	-0.02	0.00	0.69
mlVAR_EB	C_reliable	E_quiet	NaN	0.01	0.00	NaN	5.91	-0.02	0.03	-1.46
mlVAR_EB	C_reliable	N_depressed	3.42	-0.01	0.02	0.21	2.38	-0.01	0.02	-0.58
mlVAR_EB	C_reliable	N_relaxed	NaN	-0.01	0.00	NaN	NaN	0.01	0.00	NaN
mlVAR_EB	C_reliable	N_worried	NaN	0.00	0.00	NaN	NaN	-0.01	0.00	NaN
mlVAR_EB	E_outgoing	A_kind	NaN	-0.03	0.00	NaN	2.59	0.02	0.02	0.64
mlVAR_EB	E_outgoing	A_rude	NaN	-0.03	0.00	NaN	NaN	-0.01	0.00	NaN
mlVAR_EB	E_outgoing	C_lazy	0.83	-0.03	0.02	-0.26	2.15	0.03	0.01	0.82
mlVAR_EB	E_outgoing	C_reliable	1.60	-0.02	0.04	0.14	2.89	0.04	0.02	-0.26
mlVAR_EB	E_outgoing	E_outgoing	0.90	0.01	0.02	0.41	2.27	0.02	0.01	0.40
mlVAR_EB	E_outgoing	E_quiet	1.76	0.01	0.00	-0.43	NaN	0.00	0.00	NaN
mlVAR_EB	E_outgoing	N_depressed	1.81	0.02	0.02	-0.64	0.64	-0.02	0.02	-0.24
mlVAR_EB	E_outgoing	N_relaxed	2.01	0.01	0.01	-0.19	NaN	0.01	0.00	NaN
mlVAR_EB	E_outgoing	N_worried	1.41	-0.06	0.03	0.06	2.76	-0.03	0.01	-0.91
mlVAR_EB	E_quiet	A_kind	2.11	0.01	0.03	-0.38	1.30	-0.02	0.00	0.14
mlVAR_EB	E_quiet	A_rude	NaN	0.03	0.00	NaN	NaN	0.01	0.00	NaN
mlVAR_EB	E_quiet	C_lazy	2.14	0.04	0.02	-0.11	2.39	-0.02	0.00	0.35
mlVAR_EB	E_quiet	C_reliable	1.14	0.00	0.02	-0.18	3.05	-0.03	0.00	0.52
mlVAR_EB	E_quiet	E_outgoing	NaN	-0.01	0.00	NaN	0.84	0.01	0.01	-0.10
mlVAR_EB	E_quiet	E_quiet	2.80	-0.01	0.00	0.35	1.14	0.03	0.01	0.18
mlVAR_EB	E_quiet	N_depressed	1.88	0.00	0.01	-0.04	NaN	0.00	0.00	NaN
mlVAR_EB	E_quiet	N_relaxed	2.37	-0.01	0.01	-0.42	0.97	0.04	0.05	-0.35
mlVAR_EB	E_quiet	N_worried	1.18	0.04	0.02	0.07	0.70	0.06	0.04	0.38
mlVAR_EB	N_depressed	A_kind	NaN	0.01	0.00	NaN	2.74	0.01	0.04	-0.16
mlVAR_EB	N_depressed	A_rude	3.53	0.03	0.02	0.71	NaN	0.03	0.00	NaN
mlVAR_EB	N_depressed	C_lazy	2.80	0.01	0.00	0.18	4.05	-0.01	0.01	-0.98
mlVAR_EB	N_depressed	C_reliable	2.22	-0.01	0.02	0.00	2.03	-0.03	0.01	0.38
mlVAR_EB	N_depressed	E_outgoing	NaN	0.01	0.00	NaN	2.04	0.00	0.02	0.33
mlVAR_EB	N_depressed	E_quiet	NaN	0.00	0.00	NaN	NaN	0.00	0.00	NaN
mlVAR_EB	N_depressed	N_depressed	3.33	0.07	0.08	1.08	2.40	0.11	0.10	0.76
mlVAR_EB	N_depressed	N_relaxed	NaN	-0.02	0.00	NaN	NaN	0.00	0.00	NaN
mlVAR_EB	N_depressed	N_worried	1.52	0.04	0.03	0.47	1.54	0.02	0.02	0.45
mlVAR_EB	N_relaxed	A_kind	3.64	0.00	0.01	0.36	1.70	0.02	0.02	-0.22
mlVAR_EB	N_relaxed	A_rude	2.53	-0.05	0.04	-0.35	NaN	-0.02	0.00	NaN
mlVAR_EB	N_relaxed	C_lazy	1.34	0.00	0.03	-0.02	3.04	0.03	0.00	0.64
mlVAR_EB	N_relaxed	C_reliable	1.91	0.00	0.04	0.16	NaN	0.02	0.00	NaN
mlVAR_EB	N_relaxed	E_outgoing	1.31	-0.02	0.02	0.07	0.89	-0.03	0.01	-0.61
mlVAR_EB	N_relaxed	E_quiet	0.54	0.00	0.01	0.02	0.26	-0.03	0.02	-0.27
mlVAR_EB	N_relaxed	N_depressed	NaN	-0.01	0.00	NaN	3.16	-0.05	0.02	-0.69
mlVAR_EB	N_relaxed	N_relaxed	2.03	0.02	0.01	0.27	3.92	0.04	0.03	0.89
mlVAR_EB	N_relaxed	N_worried	1.97	-0.10	0.05	-0.79	1.28	-0.08	0.01	-0.76
mlVAR_EB	N_worried	A_kind	NaN	0.02	0.00	NaN	1.49	0.01	0.02	0.36
mlVAR_EB	N_worried	A_rude	3.58	0.04	0.04	0.25	0.61	0.05	0.00	0.51
mlVAR_EB	N_worried	C_lazy	2.09	0.02	0.02	0.38	2.37	0.00	0.01	-0.28
mlVAR_EB	N_worried	C_reliable	1.74	0.01	0.05	0.03	0.37	-0.03	0.03	0.15
mlVAR_EB	N_worried	E_outgoing	1.47	0.00	0.02	0.24	1.84	0.04	0.01	0.90
mlVAR_EB	N_worried	E_quiet	NaN	0.01	0.00	NaN	2.47	0.02	0.02	-0.72
mlVAR_EB	N_worried	N_depressed	2.68	0.02	0.02	0.73	2.27	-0.01	0.02	0.54
mlVAR_EB	N_worried	N_relaxed	1.93	-0.05	0.00	-0.33	2.69	-0.01	0.03	-0.42
mlVAR_EB	N_worried	N_worried	1.89	0.11	0.10	0.71	2.75	0.18	0.07	0.81
scaled	A_kind	A_kind	12.35	-0.01	0.13	0.00	12.43	-0.03	0.15	-2.22
scaled	A_kind	A_rude	9.75	0.06	0.22	0.22	9.20	0.09	0.22	0.14
scaled	A_kind	C_lazy	25.51	0.00	0.12	0.38	14.46	0.06	0.18	3.46
scaled	A_kind	C_reliable	20.86	0.04	0.16	1.35	10.69	0.04	0.20	1.34
scaled	A_kind	E_outgoing	43.17	-0.05	0.12	-3.27	20.26	-0.02	0.15	-2.49
scaled	A_kind	E_quiet	54.46	0.02	0.09	5.68	16.62	0.03	0.16	1.43
scaled	A_kind	N_depressed	12.81	-0.06	0.19	-2.51	7.92	0.01	0.24	-0.20
scaled	A_kind	N_relaxed	23.84	-0.01	0.13	0.11	17.42	-0.07	0.14	-2.26
scaled	A_kind	N_worried	29.90	0.00	0.11	-1.31	15.10	0.06	0.16	2.91
scaled	A_rude	A_kind	16.90	-0.01	0.10	-1.83	12.66	0.05	0.18	2.42
scaled	A_rude	A_rude	10.60	-0.01	0.15	0.19	12.66	-0.01	0.16	-0.60
scaled	A_rude	C_lazy	60.80	-0.01	0.09	-4.02	13.85	0.00	0.14	0.01
scaled	A_rude	C_reliable	30.99	0.00	0.12	0.61	25.42	0.01	0.15	1.09
scaled	A_rude	E_outgoing	118.68	0.01	0.07	5.80	23.38	0.00	0.15	-0.44
scaled	A_rude	E_quiet	69.71	0.00	0.09	-0.81	22.94	0.01	0.13	0.19
scaled	A_rude	N_depressed	22.14	0.00	0.14	-0.09	17.86	0.02	0.16	1.65
scaled	A_rude	N_relaxed	48.61	0.00	0.07	3.44	37.12	0.00	0.11	3.68
scaled	A_rude	N_worried	47.35	0.00	0.08	3.44	19.86	-0.04	0.16	-0.96
scaled	C_lazy	A_kind	11.66	-0.01	0.15	-0.12	11.86	-0.04	0.19	-1.31
scaled	C_lazy	A_rude	6.57	-0.06	0.24	-0.57	6.67	-0.04	0.24	0.06
scaled	C_lazy	C_lazy	9.87	0.00	0.10	-0.74	12.89	0.00	0.14	1.39
scaled	C_lazy	C_reliable	14.18	0.04	0.15	1.28	11.20	0.00	0.18	0.67
scaled	C_lazy	E_outgoing	26.15	0.04	0.14	2.70	15.35	-0.03	0.16	-2.79
scaled	C_lazy	E_quiet	26.77	0.00	0.13	0.29	15.97	-0.02	0.17	-0.70
scaled	C_lazy	N_depressed	8.78	0.00	0.21	0.01	8.90	-0.02	0.22	-0.38
scaled	C_lazy	N_relaxed	16.04	0.00	0.16	-0.03	17.06	0.04	0.17	-0.17
scaled	C_lazy	N_worried	17.81	0.00	0.14	0.41	9.89	0.07	0.22	1.16
scaled	C_reliable	A_kind	12.24	-0.04	0.16	-1.02	11.59	-0.02	0.19	-1.35
scaled	C_reliable	A_rude	9.88	-0.03	0.21	-0.58	8.34	-0.12	0.23	-1.20
scaled	C_reliable	C_lazy	26.77	0.02	0.12	2.92	16.12	-0.04	0.18	-3.23
scaled	C_reliable	C_reliable	14.35	-0.01	0.12	-1.08	6.98	0.00	0.17	-0.68
scaled	C_reliable	E_outgoing	52.94	0.03	0.11	3.31	38.15	0.02	0.11	5.03
scaled	C_reliable	E_quiet	38.76	0.01	0.12	0.59	18.25	-0.04	0.15	-3.64
scaled	C_reliable	N_depressed	14.44	-0.04	0.18	-0.14	11.44	0.00	0.20	0.66
scaled	C_reliable	N_relaxed	27.33	-0.01	0.14	-0.69	28.24	0.08	0.13	3.05
scaled	C_reliable	N_worried	12.73	-0.01	0.14	-0.24	15.21	-0.11	0.19	-2.66
scaled	E_outgoing	A_kind	14.35	0.05	0.15	1.02	8.69	0.03	0.19	1.07
scaled	E_outgoing	A_rude	5.91	-0.03	0.25	-0.59	4.27	-0.08	0.25	-0.30
scaled	E_outgoing	C_lazy	25.24	-0.02	0.15	-0.94	16.43	0.01	0.18	-0.64
scaled	E_outgoing	C_reliable	16.42	-0.03	0.17	-1.81	7.41	0.04	0.24	0.59
scaled	E_outgoing	E_outgoing	24.38	0.01	0.10	2.69	23.61	-0.01	0.12	-2.42
scaled	E_outgoing	E_quiet	37.41	-0.02	0.10	-5.38	25.74	0.00	0.17	-0.33
scaled	E_outgoing	N_depressed	10.59	0.01	0.20	-0.03	7.10	0.05	0.24	0.84
scaled	E_outgoing	N_relaxed	16.71	-0.02	0.17	-0.56	11.67	0.10	0.18	2.59
scaled	E_outgoing	N_worried	12.17	0.01	0.16	1.10	7.30	0.01	0.23	-1.04
scaled	E_quiet	A_kind	12.60	0.04	0.16	0.14	12.97	0.02	0.18	0.56
scaled	E_quiet	A_rude	7.34	0.00	0.23	0.09	5.67	-0.07	0.25	-0.29
scaled	E_quiet	C_lazy	18.29	-0.03	0.16	-1.52	14.51	0.00	0.17	-0.05
scaled	E_quiet	C_reliable	15.45	-0.03	0.16	-1.05	7.11	0.06	0.25	1.37
scaled	E_quiet	E_outgoing	34.55	0.06	0.12	4.40	13.66	0.00	0.16	-0.28
scaled	E_quiet	E_quiet	21.47	-0.02	0.12	-2.50	12.94	-0.02	0.14	-2.25
scaled	E_quiet	N_depressed	13.51	0.05	0.19	1.07	8.94	0.01	0.23	0.84
scaled	E_quiet	N_relaxed	14.66	0.00	0.16	1.06	15.39	0.10	0.18	2.44
scaled	E_quiet	N_worried	15.37	0.03	0.16	3.01	13.66	0.04	0.21	-0.31
scaled	N_depressed	A_kind	21.89	0.00	0.13	0.30	18.11	0.01	0.17	0.95
scaled	N_depressed	A_rude	8.18	0.00	0.22	0.32	10.82	0.07	0.18	0.87
scaled	N_depressed	C_lazy	25.42	0.00	0.13	0.44	19.01	0.00	0.17	0.28
scaled	N_depressed	C_reliable	21.95	-0.06	0.14	-3.71	13.34	0.04	0.17	0.78
scaled	N_depressed	E_outgoing	62.10	-0.03	0.11	-6.64	27.92	0.01	0.15	1.39
scaled	N_depressed	E_quiet	45.37	0.01	0.09	5.77	23.57	-0.03	0.15	-3.22
scaled	N_depressed	N_depressed	11.09	0.01	0.14	-1.15	13.13	0.02	0.16	0.81
scaled	N_depressed	N_relaxed	31.42	-0.02	0.11	-2.15	20.62	-0.09	0.14	-2.97
scaled	N_depressed	N_worried	20.21	0.01	0.09	1.53	19.66	-0.04	0.16	-0.18
scaled	N_relaxed	A_kind	14.85	0.03	0.15	2.50	9.19	0.00	0.20	-0.09
scaled	N_relaxed	A_rude	5.98	-0.06	0.25	-0.44	3.68	0.07	0.22	1.07
scaled	N_relaxed	C_lazy	21.61	0.02	0.14	1.54	16.65	-0.02	0.18	-0.58
scaled	N_relaxed	C_reliable	15.81	0.03	0.17	0.61	9.56	0.03	0.22	0.40
scaled	N_relaxed	E_outgoing	35.04	-0.02	0.13	-1.15	13.67	0.01	0.19	0.35
scaled	N_relaxed	E_quiet	20.48	-0.01	0.12	-1.83	16.37	0.02	0.18	0.55
scaled	N_relaxed	N_depressed	9.74	-0.01	0.20	0.21	11.39	0.09	0.22	2.09
scaled	N_relaxed	N_relaxed	16.55	-0.01	0.13	0.06	8.57	-0.04	0.14	-2.76
scaled	N_relaxed	N_worried	17.51	0.00	0.14	1.30	8.85	-0.01	0.22	1.19
scaled	N_worried	A_kind	13.67	0.04	0.15	2.01	8.69	0.02	0.18	1.92
scaled	N_worried	A_rude	6.84	0.06	0.24	0.03	6.18	-0.09	0.22	-0.08
scaled	N_worried	C_lazy	19.79	-0.02	0.16	-0.86	13.39	-0.01	0.21	0.01
scaled	N_worried	C_reliable	17.49	-0.02	0.16	0.50	14.59	0.01	0.18	0.18
scaled	N_worried	E_outgoing	40.82	-0.05	0.12	-4.37	16.00	0.01	0.18	1.28
scaled	N_worried	E_quiet	28.87	0.00	0.13	0.66	23.44	0.02	0.14	1.63
scaled	N_worried	N_depressed	12.27	0.05	0.19	1.20	7.49	0.10	0.23	2.04
scaled	N_worried	N_relaxed	12.64	-0.04	0.17	-0.81	17.53	0.03	0.15	0.99
scaled	N_worried	N_worried	11.12	0.02	0.12	0.52	8.04	0.01	0.18	-0.14
unregularized	A_kind	A_kind	0.18	0.01	0.16	-0.31	0.40	0.03	0.16	-0.01
unregularized	A_kind	A_rude	0.17	0.00	0.15	-0.17	0.65	0.02	0.17	-0.41
unregularized	A_kind	C_lazy	0.58	0.00	0.16	-0.24	-0.61	0.01	0.14	-0.15
unregularized	A_kind	C_reliable	1.76	0.01	0.15	0.50	-0.45	0.01	0.13	0.03
unregularized	A_kind	E_outgoing	-0.21	-0.01	0.15	-0.02	0.41	0.00	0.14	0.26
unregularized	A_kind	E_quiet	-0.21	0.00	0.14	0.10	0.71	0.00	0.15	0.14
unregularized	A_kind	N_depressed	0.72	0.01	0.15	0.02	-0.03	0.02	0.15	-0.26
unregularized	A_kind	N_relaxed	0.27	-0.01	0.15	-0.03	0.48	0.01	0.13	-0.51
unregularized	A_kind	N_worried	-0.10	0.02	0.14	0.03	1.32	-0.01	0.14	0.35
unregularized	A_rude	A_kind	0.36	0.00	0.16	-0.01	1.10	0.00	0.14	-0.58
unregularized	A_rude	A_rude	2.63	0.00	0.15	-0.22	0.03	0.01	0.14	0.51
unregularized	A_rude	C_lazy	3.37	-0.01	0.18	-0.33	0.20	0.03	0.15	0.21
unregularized	A_rude	C_reliable	2.88	0.01	0.17	0.73	0.81	0.00	0.15	-0.11
unregularized	A_rude	E_outgoing	0.17	0.00	0.15	0.05	0.01	-0.01	0.13	-0.06
unregularized	A_rude	E_quiet	2.90	0.00	0.17	-0.36	-0.44	0.01	0.14	0.49
unregularized	A_rude	N_depressed	4.26	0.01	0.17	-0.07	0.46	0.01	0.13	-0.22
unregularized	A_rude	N_relaxed	3.72	-0.02	0.17	-0.65	0.16	0.01	0.14	0.25
unregularized	A_rude	N_worried	3.77	-0.01	0.17	-0.28	0.71	0.04	0.13	0.25
unregularized	C_lazy	A_kind	0.08	0.01	0.16	0.25	0.07	0.00	0.16	-0.08
unregularized	C_lazy	A_rude	0.45	0.00	0.15	-0.06	-0.10	0.02	0.14	0.11
unregularized	C_lazy	C_lazy	0.18	0.03	0.15	-0.08	-0.06	0.03	0.15	0.16
unregularized	C_lazy	C_reliable	-0.25	-0.03	0.16	-0.14	0.62	-0.01	0.14	0.44
unregularized	C_lazy	E_outgoing	0.55	-0.01	0.15	-0.15	0.18	-0.01	0.13	-0.08
unregularized	C_lazy	E_quiet	1.53	0.01	0.15	-0.26	-0.49	-0.01	0.14	-0.33
unregularized	C_lazy	N_depressed	0.04	0.01	0.15	0.05	0.34	0.01	0.16	0.19
unregularized	C_lazy	N_relaxed	-0.05	-0.01	0.15	0.18	-0.40	0.02	0.12	-0.31
unregularized	C_lazy	N_worried	-0.44	0.01	0.15	-0.01	-0.15	0.01	0.15	-0.12
unregularized	C_reliable	A_kind	0.33	0.02	0.16	-0.23	0.01	0.00	0.13	0.28
unregularized	C_reliable	A_rude	-0.18	0.00	0.16	-0.18	0.18	0.00	0.15	0.43
unregularized	C_reliable	C_lazy	0.31	0.00	0.16	-0.26	-0.34	-0.01	0.13	-0.05
unregularized	C_reliable	C_reliable	0.22	0.01	0.15	0.40	0.34	0.05	0.15	0.45
unregularized	C_reliable	E_outgoing	-0.13	0.00	0.17	-0.05	0.30	0.00	0.14	-0.29
unregularized	C_reliable	E_quiet	0.13	-0.02	0.16	-0.11	0.51	0.02	0.14	0.50
unregularized	C_reliable	N_depressed	-0.18	0.01	0.15	0.00	0.52	-0.01	0.15	0.08
unregularized	C_reliable	N_relaxed	0.12	0.00	0.15	-0.28	0.20	0.01	0.15	0.08
unregularized	C_reliable	N_worried	0.56	-0.01	0.16	-0.04	0.72	0.00	0.14	0.09
unregularized	E_outgoing	A_kind	0.18	0.01	0.15	0.13	1.40	0.01	0.18	0.16
unregularized	E_outgoing	A_rude	0.43	-0.01	0.15	0.08	0.98	-0.01	0.16	-0.22
unregularized	E_outgoing	C_lazy	0.14	0.02	0.17	-0.14	0.74	0.01	0.13	0.06
unregularized	E_outgoing	C_reliable	-0.43	-0.01	0.15	-0.27	-0.24	-0.02	0.15	-0.18
unregularized	E_outgoing	E_outgoing	-0.05	0.00	0.15	-0.01	-0.79	0.02	0.14	-0.03
unregularized	E_outgoing	E_quiet	0.55	0.00	0.15	0.30	0.17	-0.01	0.14	0.12
unregularized	E_outgoing	N_depressed	-0.08	0.00	0.14	-0.04	0.18	-0.01	0.15	0.07
unregularized	E_outgoing	N_relaxed	-0.03	0.00	0.16	-0.12	0.80	0.00	0.14	-0.40
unregularized	E_outgoing	N_worried	0.19	-0.01	0.14	-0.17	0.08	0.01	0.16	0.27
unregularized	E_quiet	A_kind	0.67	0.00	0.16	0.06	0.96	0.01	0.16	0.17
unregularized	E_quiet	A_rude	0.12	0.01	0.14	-0.13	1.44	0.00	0.15	-0.24
unregularized	E_quiet	C_lazy	0.27	0.01	0.16	-0.21	-0.10	-0.01	0.14	0.24
unregularized	E_quiet	C_reliable	0.29	0.00	0.15	-0.13	-0.23	0.00	0.14	-0.16
unregularized	E_quiet	E_outgoing	-0.22	0.00	0.14	0.13	0.09	0.02	0.14	0.13
unregularized	E_quiet	E_quiet	-0.71	0.00	0.15	0.00	0.04	-0.01	0.14	-0.14
unregularized	E_quiet	N_depressed	0.48	0.00	0.14	-0.10	1.04	0.00	0.14	0.21
unregularized	E_quiet	N_relaxed	0.68	0.01	0.15	0.08	0.21	-0.01	0.14	0.26
unregularized	E_quiet	N_worried	0.16	0.00	0.15	0.00	0.08	0.00	0.15	-0.23
unregularized	N_depressed	A_kind	0.06	0.01	0.15	0.18	0.38	-0.02	0.14	-0.16
unregularized	N_depressed	A_rude	2.56	-0.01	0.17	0.92	2.83	-0.02	0.17	-0.08
unregularized	N_depressed	C_lazy	0.91	0.01	0.17	-0.28	-0.38	0.01	0.14	-0.21
unregularized	N_depressed	C_reliable	1.96	0.00	0.16	-0.07	-0.15	-0.03	0.14	-0.16
unregularized	N_depressed	E_outgoing	1.27	0.00	0.16	-0.38	-0.19	-0.01	0.16	0.09
unregularized	N_depressed	E_quiet	0.31	0.00	0.16	0.12	0.26	0.00	0.14	-0.06
unregularized	N_depressed	N_depressed	-0.46	0.05	0.17	0.31	1.96	0.08	0.18	0.73
unregularized	N_depressed	N_relaxed	3.65	0.00	0.15	-0.49	0.28	-0.02	0.14	0.41
unregularized	N_depressed	N_worried	1.35	0.01	0.17	-0.15	0.96	0.01	0.14	-0.13
unregularized	N_relaxed	A_kind	0.66	-0.01	0.15	-0.27	0.05	0.01	0.16	0.21
unregularized	N_relaxed	A_rude	1.09	0.00	0.15	-0.28	-0.02	0.00	0.13	-0.24
unregularized	N_relaxed	C_lazy	1.04	0.01	0.16	0.24	1.91	0.01	0.14	-0.29
unregularized	N_relaxed	C_reliable	0.35	0.01	0.17	0.03	0.93	0.01	0.15	0.00
unregularized	N_relaxed	E_outgoing	0.55	0.00	0.15	-0.09	0.54	0.00	0.14	0.32
unregularized	N_relaxed	E_quiet	0.68	-0.01	0.16	-0.22	-0.32	0.03	0.14	0.01
unregularized	N_relaxed	N_depressed	1.39	-0.02	0.16	0.06	-0.16	-0.02	0.15	-0.11
unregularized	N_relaxed	N_relaxed	-0.24	0.02	0.14	0.05	-0.46	0.02	0.13	0.23
unregularized	N_relaxed	N_worried	0.20	-0.03	0.14	-0.07	0.96	-0.01	0.16	0.24
unregularized	N_worried	A_kind	0.61	-0.04	0.15	-0.04	1.54	-0.03	0.17	0.26
unregularized	N_worried	A_rude	1.03	0.01	0.15	-0.11	2.55	0.03	0.16	0.54
unregularized	N_worried	C_lazy	0.55	-0.02	0.17	0.33	0.48	-0.03	0.14	-0.05
unregularized	N_worried	C_reliable	0.26	-0.02	0.16	-0.22	0.02	0.01	0.15	0.15
unregularized	N_worried	E_outgoing	0.08	-0.03	0.16	0.14	-0.22	-0.02	0.14	-0.01
unregularized	N_worried	E_quiet	0.01	0.02	0.17	-0.14	0.11	0.04	0.15	-0.34
unregularized	N_worried	N_depressed	0.07	0.04	0.15	0.26	0.37	0.02	0.16	0.07
unregularized	N_worried	N_relaxed	0.29	-0.07	0.17	-0.14	0.63	-0.06	0.14	-0.14
unregularized	N_worried	N_worried	0.06	0.09	0.16	-0.26	-0.27	0.11	0.16	0.12

library(ggridges)
## figure  
temp_long %>% 
  filter(lambda_scale == ".25") %>%
  ggplot(aes(x = weight, y = from, fill = factor(wave))) +
  # tidybayes::geom_halfeyeh() +
  geom_density_ridges(rel_min_height = 0.025) +
  # geom_density() +
  labs(x = "From", y = "Edge Weight", 
       title = "Regularized Lagged Partial Correlations") +
  facet_grid(to~wave) +
  theme_classic() +
  theme(legend.position = "none")

Contemporaneous

#Contemporaneous: Partial Contemporaneous Correlations
contemp_long <- gVAR_fit %>%
  unnest(contemp, .drop = T) %>%
  mutate(Var1 = factor(Var1, levels = sort(unique(Var1))),
         Var2 = factor(Var2, levels = sort(unique(Var2))),
         type = "Contemporaneous") 

contemp_long %>%
  group_by(wave, lambda_scale, var) %>%
  summarise_at(vars(weight), funs(M = fisherz2r(mean(fisherz(.), na.rm = T)), 
                             sd(., na.rm = T), kurtosi, skew)) %>%
  gather(est, value, M:skew) %>%
  unite(tmp, wave, est, sep = ".") %>%
  # arrange(desc(r)) %>%
  spread(key = tmp, value = value) %>%
  kable(., "html", booktabs = T, longtable = T, digits = 2,
        caption = "Descriptives of Contemporaneous Edge Weights",
        col.names = c("Network Type", "Variable", rep(c("Kurtosi", "M", "SD", "Skew"),2))) %>%
  kable_styling(bootstrap_options = c("striped"),full_width = T) %>%
  add_header_above(c(" " = 2, "Wave 1" = 4, "Wave 2" = 4)) %>%
  scroll_box(width = "750px", height = "400px")

Descriptives of Contemporaneous Edge Weights

		Wave 1				Wave 2
Network Type	Variable	Kurtosi	M	SD	Skew	Kurtosi	M	SD	Skew
.25	A_kind.A_rude	15.00	-0.03	0.11	-3.29	4.81	-0.02	0.11	-0.75
.25	A_kind.C_lazy	7.84	-0.01	0.12	0.07	9.56	-0.02	0.13	-0.72
.25	A_kind.C_reliable	8.68	0.03	0.12	0.55	24.25	0.02	0.11	2.35
.25	A_kind.E_outgoing	4.49	0.07	0.14	1.67	5.09	0.06	0.12	1.86
.25	A_kind.E_quiet	4.07	-0.04	0.10	-1.16	14.98	-0.03	0.12	-2.10
.25	A_kind.N_depressed	15.41	-0.02	0.08	-3.25	17.81	-0.01	0.07	1.57
.25	A_kind.N_relaxed	6.88	0.03	0.10	1.48	11.55	0.02	0.12	0.31
.25	A_kind.N_worried	12.78	-0.02	0.09	-2.01	8.53	-0.01	0.12	0.15
.25	A_rude.C_lazy	18.09	0.01	0.09	0.48	11.73	0.01	0.10	0.71
.25	A_rude.C_reliable	19.96	-0.01	0.09	-0.64	19.40	0.00	0.07	1.45
.25	A_rude.E_outgoing	20.91	0.01	0.08	0.90	15.85	0.01	0.08	0.86
.25	A_rude.E_quiet	16.03	-0.01	0.08	-0.13	8.91	-0.02	0.10	-0.83
.25	A_rude.N_depressed	15.92	0.02	0.09	2.10	15.12	0.01	0.07	1.68
.25	A_rude.N_relaxed	23.76	-0.01	0.10	-1.42	18.73	-0.01	0.09	0.10
.25	A_rude.N_worried	21.22	0.00	0.08	-0.99	14.74	0.02	0.09	2.66
.25	C_lazy.C_reliable	0.96	-0.13	0.19	-0.36	0.26	-0.14	0.19	-0.89
.25	C_lazy.E_outgoing	3.73	-0.04	0.15	-0.06	4.05	-0.03	0.10	-1.20
.25	C_lazy.E_quiet	6.14	0.04	0.14	0.81	5.94	0.02	0.13	0.01
.25	C_lazy.N_depressed	7.99	0.03	0.12	1.52	13.90	-0.01	0.09	-2.22
.25	C_lazy.N_relaxed	0.63	0.11	0.16	0.58	1.36	0.10	0.16	1.14
.25	C_lazy.N_worried	8.70	-0.02	0.13	-0.49	5.35	-0.01	0.11	0.66
.25	C_reliable.E_outgoing	7.82	0.03	0.13	-0.51	9.89	0.04	0.12	-0.34
.25	C_reliable.E_quiet	4.66	-0.01	0.10	0.02	5.76	-0.02	0.12	-0.09
.25	C_reliable.N_depressed	10.75	-0.03	0.12	-0.98	8.09	-0.03	0.12	-1.71
.25	C_reliable.N_relaxed	8.72	0.02	0.13	0.84	6.75	0.01	0.08	0.61
.25	C_reliable.N_worried	7.77	-0.01	0.11	-0.14	15.75	-0.01	0.11	-0.06
.25	E_outgoing.E_quiet	-0.17	-0.51	0.22	0.66	-0.50	-0.50	0.25	0.52
.25	E_outgoing.N_depressed	6.48	-0.04	0.10	-1.87	11.54	-0.04	0.11	-2.60
.25	E_outgoing.N_relaxed	3.65	0.04	0.12	1.09	4.91	0.03	0.12	0.61
.25	E_outgoing.N_worried	9.61	-0.03	0.12	0.17	6.09	-0.04	0.10	-1.55
.25	E_quiet.N_depressed	14.14	0.02	0.08	1.26	7.13	0.03	0.10	1.27
.25	E_quiet.N_relaxed	6.50	-0.02	0.14	0.41	4.81	-0.02	0.13	-1.05
.25	E_quiet.N_worried	9.09	0.02	0.13	0.86	5.63	0.02	0.12	0.64
.25	N_depressed.N_relaxed	2.46	-0.07	0.12	-1.49	6.87	-0.05	0.12	-2.56
.25	N_depressed.N_worried	2.22	0.11	0.16	1.27	6.21	0.09	0.15	0.38
.25	N_relaxed.N_worried	-0.42	-0.24	0.21	-0.44	-0.59	-0.21	0.19	-0.27
.5	A_kind.A_rude	20.68	-0.02	0.10	-3.37	5.78	-0.02	0.11	-0.77
.5	A_kind.C_lazy	10.47	0.00	0.11	0.15	10.92	-0.01	0.12	-0.78
.5	A_kind.C_reliable	11.44	0.02	0.11	0.39	28.36	0.02	0.11	2.53
.5	A_kind.E_outgoing	6.47	0.06	0.12	1.92	6.42	0.05	0.11	2.08
.5	A_kind.E_quiet	4.75	-0.03	0.09	-1.07	18.55	-0.02	0.11	-2.36
.5	A_kind.N_depressed	21.08	-0.02	0.08	-3.74	15.42	-0.01	0.08	1.18
.5	A_kind.N_relaxed	9.49	0.03	0.10	1.78	12.08	0.02	0.12	0.31
.5	A_kind.N_worried	14.42	-0.02	0.09	-2.09	9.97	-0.01	0.12	0.22
.5	A_rude.C_lazy	22.38	0.01	0.08	-0.01	13.04	0.01	0.11	1.26
.5	A_rude.C_reliable	28.56	-0.01	0.08	-0.70	23.50	0.00	0.06	1.51
.5	A_rude.E_outgoing	22.91	0.01	0.08	1.04	16.21	0.01	0.08	0.87
.5	A_rude.E_quiet	19.95	0.00	0.07	0.37	10.29	-0.02	0.09	-1.07
.5	A_rude.N_depressed	18.25	0.02	0.09	2.28	22.86	0.01	0.06	1.73
.5	A_rude.N_relaxed	30.21	-0.01	0.09	-1.20	24.71	-0.01	0.08	0.54
.5	A_rude.N_worried	24.53	0.00	0.08	-1.52	14.00	0.01	0.09	2.33
.5	C_lazy.C_reliable	1.54	-0.11	0.19	-0.57	0.59	-0.13	0.17	-0.95
.5	C_lazy.E_outgoing	4.29	-0.04	0.14	-0.22	4.79	-0.03	0.09	-1.36
.5	C_lazy.E_quiet	7.80	0.03	0.14	0.97	6.42	0.02	0.13	-0.07
.5	C_lazy.N_depressed	10.52	0.03	0.11	2.36	18.28	-0.01	0.08	-2.53
.5	C_lazy.N_relaxed	1.74	0.10	0.15	0.96	1.49	0.10	0.16	1.20
.5	C_lazy.N_worried	11.14	-0.01	0.12	-0.54	5.88	-0.01	0.10	0.92
.5	C_reliable.E_outgoing	9.44	0.02	0.13	-0.47	10.62	0.04	0.12	-0.33
.5	C_reliable.E_quiet	5.58	-0.01	0.09	0.12	7.38	-0.01	0.12	0.13
.5	C_reliable.N_depressed	14.35	-0.03	0.11	-1.43	8.38	-0.03	0.12	-1.68
.5	C_reliable.N_relaxed	11.91	0.02	0.12	1.05	7.59	0.01	0.08	0.76
.5	C_reliable.N_worried	9.37	-0.01	0.10	-0.17	16.50	-0.01	0.10	0.04
.5	E_outgoing.E_quiet	-0.39	-0.48	0.22	0.51	-0.57	-0.47	0.24	0.43
.5	E_outgoing.N_depressed	8.96	-0.03	0.10	-2.26	13.85	-0.03	0.11	-2.90
.5	E_outgoing.N_relaxed	4.43	0.04	0.11	0.86	6.51	0.03	0.11	0.84
.5	E_outgoing.N_worried	11.66	-0.03	0.11	0.08	4.70	-0.03	0.09	-1.31
.5	E_quiet.N_depressed	17.18	0.02	0.08	1.46	9.77	0.02	0.09	1.53
.5	E_quiet.N_relaxed	7.15	-0.02	0.13	0.50	5.34	-0.02	0.12	-0.92
.5	E_quiet.N_worried	10.83	0.02	0.12	1.15	6.35	0.02	0.12	0.69
.5	N_depressed.N_relaxed	3.17	-0.06	0.11	-1.58	6.50	-0.05	0.10	-2.41
.5	N_depressed.N_worried	3.34	0.10	0.15	1.50	7.31	0.08	0.15	0.62
.5	N_relaxed.N_worried	-0.21	-0.21	0.20	-0.57	-0.66	-0.19	0.19	-0.30
association	A_kind.A_rude	-0.05	-1.00	0.29	-0.06	-0.42	-0.18	0.33	0.21
association	A_kind.C_lazy	0.07	-0.07	0.23	0.19	0.44	-0.07	0.22	0.00
association	A_kind.C_reliable	-0.14	0.21	0.22	-0.28	-0.18	0.19	0.23	-0.23
association	A_kind.E_outgoing	0.36	0.30	0.21	-0.59	0.26	0.27	0.21	-0.49
association	A_kind.E_quiet	0.43	-0.26	0.22	0.40	0.04	-0.22	0.22	0.35
association	A_kind.N_depressed	0.22	-0.15	0.24	0.12	0.39	-0.17	0.22	0.43
association	A_kind.N_relaxed	0.18	0.18	0.23	-0.42	0.34	0.18	0.22	-0.26
association	A_kind.N_worried	-0.11	-0.14	0.22	0.32	0.65	-0.15	0.24	0.54
association	A_rude.C_lazy	0.40	0.06	0.21	0.06	0.35	0.05	0.17	-0.04
association	A_rude.C_reliable	0.24	-0.10	0.22	0.02	0.11	-0.07	0.21	0.15
association	A_rude.E_outgoing	0.14	0.03	0.22	0.17	1.32	0.01	0.21	0.55
association	A_rude.E_quiet	0.38	-0.07	0.22	0.00	0.09	-0.04	0.20	0.05
association	A_rude.N_depressed	-0.32	0.12	0.24	0.31	-0.22	0.12	0.23	0.21
association	A_rude.N_relaxed	-0.02	-0.09	0.23	0.11	1.42	-0.09	0.21	0.13
association	A_rude.N_worried	-0.12	0.09	0.23	-0.06	0.74	0.09	0.21	0.20
association	C_lazy.C_reliable	-0.39	-0.32	0.25	0.35	-0.67	-0.35	0.26	0.32
association	C_lazy.E_outgoing	-0.22	-0.15	0.27	0.13	-0.33	-0.15	0.24	-0.19
association	C_lazy.E_quiet	-0.24	0.13	0.27	-0.25	-0.16	0.12	0.25	0.10
association	C_lazy.N_depressed	0.19	0.09	0.25	-0.22	0.06	0.08	0.21	0.19
association	C_lazy.N_relaxed	-0.30	0.23	0.26	-0.32	-0.30	0.22	0.26	-0.42
association	C_lazy.N_worried	0.19	-0.08	0.24	0.10	0.10	-0.05	0.24	0.55
association	C_reliable.E_outgoing	0.47	0.22	0.25	-0.32	-0.25	0.24	0.22	-0.25
association	C_reliable.E_quiet	0.28	-0.18	0.24	0.34	0.08	-0.17	0.21	0.27
association	C_reliable.N_depressed	-0.13	-0.18	0.25	-0.08	0.38	-0.19	0.25	0.18
association	C_reliable.N_relaxed	-0.29	0.08	0.26	0.13	-0.33	0.06	0.26	-0.21
association	C_reliable.N_worried	0.11	-0.10	0.25	0.27	-0.42	-0.11	0.23	0.10
association	E_outgoing.E_quiet	1.77	-0.72	0.16	1.19	2.31	-0.73	0.17	1.28
association	E_outgoing.N_depressed	0.53	-0.23	0.23	0.35	-0.25	-0.22	0.20	0.26
association	E_outgoing.N_relaxed	-0.31	0.18	0.24	-0.17	0.06	0.16	0.28	-0.51
association	E_outgoing.N_worried	-0.31	-0.20	0.22	0.16	-0.39	-0.21	0.24	0.36
association	E_quiet.N_depressed	-0.14	0.18	0.22	-0.11	-0.20	0.16	0.20	0.27
association	E_quiet.N_relaxed	-0.10	-0.13	0.25	0.14	-0.29	-0.10	0.27	0.20
association	E_quiet.N_worried	-0.18	0.16	0.22	-0.20	-0.18	0.17	0.25	-0.29
association	N_depressed.N_relaxed	0.08	-0.32	0.24	0.46	0.11	-0.30	0.22	0.57
association	N_depressed.N_worried	-0.04	0.39	0.24	-0.49	1.04	0.34	0.25	-0.85
association	N_relaxed.N_worried	0.28	-0.53	0.22	0.63	1.37	-0.48	0.24	1.17
gimme	A_kind.A_rude	2.53	-0.06	0.15	-0.96	0.41	-0.06	0.15	-0.26
gimme	A_kind.C_lazy	11.82	0.01	0.11	0.98	14.73	-0.01	0.13	1.35
gimme	A_kind.C_reliable	13.50	0.05	0.17	0.69	7.85	0.03	0.12	0.65
gimme	A_kind.E_outgoing	8.23	0.06	0.14	-0.17	0.08	0.06	0.11	0.98
gimme	A_kind.E_quiet	69.49	-0.04	0.16	5.25	136.49	-0.03	2.38	-11.72
gimme	A_kind.N_depressed	8.05	-0.04	0.13	-1.86	5.40	-0.03	0.10	-1.68
gimme	A_kind.N_relaxed	5.05	0.03	0.12	0.21	4.55	0.03	0.09	1.94
gimme	A_kind.N_worried	23.60	-0.03	0.12	-3.61	6.49	-0.02	0.12	-0.72
gimme	A_rude.C_lazy	20.19	0.02	0.12	-1.23	12.88	0.01	0.06	2.70
gimme	A_rude.C_reliable	41.72	-0.02	0.14	-3.22	7.36	0.00	0.09	-0.27
gimme	A_rude.E_outgoing	308.85	0.01	1.05	17.52	14.65	0.01	0.10	2.46
gimme	A_rude.E_quiet	9.82	-0.02	0.10	-0.16	136.99	-0.02	6.23	-11.75
gimme	A_rude.N_depressed	311.92	0.04	2.03	17.65	11.85	0.02	0.11	-0.62
gimme	A_rude.N_relaxed	58.68	-0.01	0.13	3.90	9.66	-0.02	0.10	-1.70
gimme	A_rude.N_worried	310.66	0.01	1.23	-17.59	110.42	0.01	0.36	10.05
gimme	C_lazy.C_reliable	14.53	-0.13	0.19	-2.49	1.42	-0.13	0.18	-0.56
gimme	C_lazy.E_outgoing	7.63	-0.04	0.15	-1.08	28.01	-0.01	0.14	-3.25
gimme	C_lazy.E_quiet	77.98	0.02	0.19	-5.24	136.49	0.04	3.42	11.72
gimme	C_lazy.N_depressed	9.23	0.02	0.12	0.26	15.40	0.01	0.12	2.30
gimme	C_lazy.N_relaxed	4.43	0.08	0.15	-0.47	12.82	0.07	0.16	-1.31
gimme	C_lazy.N_worried	33.01	-0.02	0.15	-4.27	122.44	-0.02	0.58	-10.81
gimme	C_reliable.E_outgoing	154.20	0.04	0.57	-11.31	33.33	0.04	0.15	-3.62
gimme	C_reliable.E_quiet	105.11	-0.03	0.32	-6.70	130.92	-0.01	0.64	-11.37
gimme	C_reliable.N_depressed	13.08	-0.05	0.14	0.35	4.55	-0.04	0.13	-0.71
gimme	C_reliable.N_relaxed	13.33	0.02	0.12	0.56	4.30	0.01	0.10	0.46
gimme	C_reliable.N_worried	23.66	-0.01	0.14	-2.27	9.57	-0.03	0.11	-2.37
gimme	E_outgoing.E_quiet	6.88	-0.33	0.17	1.53	133.90	-0.34	4.77	11.56
gimme	E_outgoing.N_depressed	143.79	-0.05	0.28	-10.10	106.27	-0.02	0.33	-9.68
gimme	E_outgoing.N_relaxed	270.84	0.03	0.47	15.88	2.97	0.03	0.12	0.08
gimme	E_outgoing.N_worried	270.22	-0.02	0.47	15.75	57.96	-0.02	0.19	-6.24
gimme	E_quiet.N_depressed	22.01	0.01	0.13	0.26	136.93	0.04	6.60	-11.74
gimme	E_quiet.N_relaxed	9.16	-0.01	0.12	-0.32	48.77	-0.01	0.17	5.23
gimme	E_quiet.N_worried	21.68	0.01	0.12	-0.60	136.97	0.01	11.47	11.75
gimme	N_depressed.N_relaxed	5.04	-0.07	0.18	0.04	10.04	-0.04	0.12	0.41
gimme	N_depressed.N_worried	3.86	0.12	0.18	0.15	6.29	0.10	0.18	1.44
gimme	N_relaxed.N_worried	17.98	-0.20	0.20	2.18	131.06	-0.18	1.13	11.37
mlVAR_EB	A_kind.A_rude	0.95	-0.14	0.12	-0.39	0.51	-0.18	0.19	0.10
mlVAR_EB	A_kind.C_lazy	0.58	-0.03	0.04	-0.13	0.62	-0.03	0.04	0.23
mlVAR_EB	A_kind.C_reliable	1.41	0.13	0.05	-0.05	0.27	0.13	0.06	-0.15
mlVAR_EB	A_kind.E_outgoing	1.47	0.16	0.06	-0.18	0.18	0.16	0.06	0.28
mlVAR_EB	A_kind.E_quiet	0.78	-0.10	0.03	0.02	1.38	-0.07	0.01	0.00
mlVAR_EB	A_kind.N_depressed	1.56	-0.03	0.05	-0.41	0.64	-0.05	0.04	0.20
mlVAR_EB	A_kind.N_relaxed	0.95	0.10	0.03	0.30	1.55	0.08	0.01	0.53
mlVAR_EB	A_kind.N_worried	1.34	0.00	0.00	0.11	1.32	-0.02	0.05	-0.19
mlVAR_EB	A_rude.C_lazy	2.65	0.05	0.05	0.33	1.71	0.05	0.00	0.16
mlVAR_EB	A_rude.C_reliable	1.98	-0.07	0.05	-0.06	1.52	-0.03	0.00	-0.26
mlVAR_EB	A_rude.E_outgoing	1.81	0.03	0.01	0.33	10.37	0.02	0.00	1.78
mlVAR_EB	A_rude.E_quiet	2.30	-0.10	0.03	-0.09	0.84	-0.08	0.02	-0.13
mlVAR_EB	A_rude.N_depressed	1.64	0.09	0.08	0.00	1.15	0.08	0.07	0.42
mlVAR_EB	A_rude.N_relaxed	2.98	-0.05	0.04	-0.35	0.90	-0.05	0.01	-0.26
mlVAR_EB	A_rude.N_worried	1.91	0.04	0.02	0.50	2.12	0.03	0.02	-0.09
mlVAR_EB	C_lazy.C_reliable	0.69	-0.25	0.08	-0.30	0.33	-0.29	0.10	-0.02
mlVAR_EB	C_lazy.E_outgoing	0.50	-0.05	0.06	-0.14	1.93	-0.03	0.01	0.04
mlVAR_EB	C_lazy.E_quiet	0.34	0.02	0.06	0.05	0.78	0.00	0.02	-0.29
mlVAR_EB	C_lazy.N_depressed	2.32	0.10	0.06	0.49	6.68	0.05	0.00	1.09
mlVAR_EB	C_lazy.N_relaxed	-0.18	0.23	0.06	-0.04	0.24	0.24	0.06	-0.35
mlVAR_EB	C_lazy.N_worried	2.28	-0.02	0.03	-0.07	1.83	-0.01	0.02	-0.15
mlVAR_EB	C_reliable.E_outgoing	3.68	0.05	0.05	0.29	1.08	0.06	0.02	0.05
mlVAR_EB	C_reliable.E_quiet	0.78	-0.04	0.02	-0.03	0.51	0.00	0.04	-0.37
mlVAR_EB	C_reliable.N_depressed	0.80	-0.08	0.06	-0.26	3.80	-0.10	0.02	-0.80
mlVAR_EB	C_reliable.N_relaxed	1.57	0.04	0.04	-0.30	2.43	0.03	0.05	0.17
mlVAR_EB	C_reliable.N_worried	0.79	-0.02	0.03	-0.11	0.11	-0.06	0.01	-0.02
mlVAR_EB	E_outgoing.E_quiet	0.85	-0.56	0.09	0.64	-0.21	-0.60	0.08	0.29
mlVAR_EB	E_outgoing.N_depressed	0.83	-0.08	0.03	-0.53	0.11	-0.07	0.01	-0.08
mlVAR_EB	E_outgoing.N_relaxed	0.53	0.09	0.04	0.02	0.34	0.09	0.06	-0.03
mlVAR_EB	E_outgoing.N_worried	0.83	-0.07	0.00	0.14	1.44	-0.07	0.03	-0.32
mlVAR_EB	E_quiet.N_depressed	0.68	0.03	0.01	0.25	2.15	0.03	0.03	0.71
mlVAR_EB	E_quiet.N_relaxed	0.56	-0.03	0.02	0.11	2.46	0.01	0.03	-0.11
mlVAR_EB	E_quiet.N_worried	0.57	0.02	0.02	0.00	0.46	0.02	0.01	-0.50
mlVAR_EB	N_depressed.N_relaxed	0.88	-0.15	0.05	-0.37	1.34	-0.15	0.05	-0.32
mlVAR_EB	N_depressed.N_worried	1.04	0.23	0.09	0.54	-0.39	0.22	0.07	0.27
mlVAR_EB	N_relaxed.N_worried	-0.05	-0.37	0.09	0.06	-0.04	-0.38	0.07	-0.05
scaled	A_kind.A_rude	109.70	-0.01	0.04	-9.92	45.86	0.00	0.04	-0.07
scaled	A_kind.C_lazy	33.67	-0.01	0.07	0.55	33.14	-0.01	0.05	-4.98
scaled	A_kind.C_reliable	38.59	0.00	0.04	3.75	58.57	0.01	0.07	7.41
scaled	A_kind.E_outgoing	7.15	0.03	0.07	2.51	14.06	0.03	0.07	3.53
scaled	A_kind.E_quiet	13.79	-0.02	0.06	-3.65	27.08	-0.01	0.07	-2.82
scaled	A_kind.N_depressed	50.36	-0.01	0.06	-0.96	72.04	0.00	0.02	-8.23
scaled	A_kind.N_relaxed	38.06	0.01	0.05	5.16	20.06	0.01	0.05	2.02
scaled	A_kind.N_worried	38.79	0.00	0.06	0.15	20.31	0.00	0.07	2.05
scaled	A_rude.C_lazy	45.10	0.01	0.05	5.18	51.63	0.00	0.02	7.20
scaled	A_rude.C_reliable	82.79	0.00	0.02	-7.51	133.39	0.00	0.01	-11.24
scaled	A_rude.E_outgoing	82.66	0.00	0.02	4.93	69.97	0.00	0.02	8.27
scaled	A_rude.E_quiet	50.18	0.00	0.03	-1.32	50.13	-0.01	0.04	-6.87
scaled	A_rude.N_depressed	54.46	0.00	0.04	6.04	109.23	0.01	0.05	10.05
scaled	A_rude.N_relaxed	65.59	0.00	0.03	-5.20	97.02	-0.01	0.06	-9.67
scaled	A_rude.N_worried	50.74	0.00	0.04	1.99	73.38	0.01	0.04	8.23
scaled	C_lazy.C_reliable	2.93	-0.07	0.13	-1.87	5.96	-0.06	0.13	-2.60
scaled	C_lazy.E_outgoing	9.90	-0.03	0.10	-1.29	18.34	-0.02	0.07	-4.02
scaled	C_lazy.E_quiet	10.59	0.02	0.10	0.71	15.67	0.01	0.09	-2.01
scaled	C_lazy.N_depressed	28.18	0.01	0.06	3.61	29.35	0.00	0.05	3.10
scaled	C_lazy.N_relaxed	4.34	0.05	0.10	1.92	9.44	0.04	0.09	2.95
scaled	C_lazy.N_worried	44.17	-0.01	0.07	-4.08	26.26	0.00	0.07	2.10
scaled	C_reliable.E_outgoing	12.46	0.02	0.08	1.49	14.37	0.02	0.04	3.44
scaled	C_reliable.E_quiet	21.43	-0.02	0.06	-3.79	16.97	-0.01	0.05	-2.62
scaled	C_reliable.N_depressed	23.77	-0.02	0.07	-4.73	39.68	-0.01	0.06	-6.10
scaled	C_reliable.N_relaxed	30.45	0.01	0.07	1.39	19.08	0.01	0.05	2.96
scaled	C_reliable.N_worried	28.31	-0.01	0.06	-1.54	56.66	0.00	0.07	3.50
scaled	E_outgoing.E_quiet	-0.94	-0.41	0.23	0.31	-1.38	-0.33	0.25	-0.13
scaled	E_outgoing.N_depressed	11.91	-0.02	0.06	-3.38	51.90	-0.01	0.06	-6.26
scaled	E_outgoing.N_relaxed	15.49	0.02	0.07	3.49	10.32	0.01	0.07	2.37
scaled	E_outgoing.N_worried	11.69	-0.02	0.07	-2.54	26.52	-0.02	0.06	-4.36
scaled	E_quiet.N_depressed	23.48	0.01	0.05	4.44	17.25	0.01	0.06	-0.72
scaled	E_quiet.N_relaxed	15.64	-0.02	0.08	-2.44	14.96	0.00	0.07	-0.52
scaled	E_quiet.N_worried	21.92	0.02	0.09	2.12	16.45	0.01	0.05	2.13
scaled	N_depressed.N_relaxed	9.45	-0.03	0.08	-2.94	21.67	-0.02	0.08	-4.48
scaled	N_depressed.N_worried	6.11	0.05	0.10	2.57	11.43	0.03	0.09	2.83
scaled	N_relaxed.N_worried	0.70	-0.13	0.17	-1.28	4.43	-0.08	0.14	-2.16
unregularized	A_kind.A_rude	-0.06	-0.13	0.37	0.27	-0.50	-0.22	0.34	0.01
unregularized	A_kind.C_lazy	0.76	0.01	0.32	0.03	0.66	-0.01	0.26	0.12
unregularized	A_kind.C_reliable	0.66	0.14	0.33	-0.19	0.62	0.08	0.25	0.02
unregularized	A_kind.E_outgoing	0.69	0.12	0.32	0.07	1.96	0.12	0.25	-0.06
unregularized	A_kind.E_quiet	1.42	-0.02	0.32	0.70	0.31	-0.09	0.21	-0.07
unregularized	A_kind.N_depressed	0.31	-0.06	0.33	-0.26	1.21	-0.06	0.28	-0.69
unregularized	A_kind.N_relaxed	1.21	0.10	0.32	-0.54	0.06	0.10	0.22	0.17
unregularized	A_kind.N_worried	0.70	-0.06	0.31	-0.34	1.63	0.02	0.29	0.88
unregularized	A_rude.C_lazy	0.88	0.01	0.29	-0.22	1.40	0.05	0.26	-0.04
unregularized	A_rude.C_reliable	0.33	-0.11	0.31	-0.27	1.56	0.02	0.27	0.40
unregularized	A_rude.E_outgoing	1.08	0.07	0.29	0.41	2.01	0.01	0.28	0.46
unregularized	A_rude.E_quiet	0.95	-0.06	0.29	0.41	1.38	-0.08	0.26	0.39
unregularized	A_rude.N_depressed	0.39	0.07	0.31	0.00	2.28	0.02	0.28	-0.85
unregularized	A_rude.N_relaxed	0.45	-0.05	0.30	-0.21	0.97	0.00	0.26	0.09
unregularized	A_rude.N_worried	0.99	0.07	0.31	-0.15	1.80	0.10	0.30	0.67
unregularized	C_lazy.C_reliable	0.55	-0.31	0.34	0.50	-0.51	-0.30	0.27	-0.03
unregularized	C_lazy.E_outgoing	1.34	-0.03	0.31	0.67	0.83	-0.06	0.27	-0.25
unregularized	C_lazy.E_quiet	0.69	0.05	0.32	0.20	-0.02	0.03	0.23	0.40
unregularized	C_lazy.N_depressed	0.80	0.11	0.31	-0.17	1.62	0.05	0.26	-0.13
unregularized	C_lazy.N_relaxed	1.18	0.27	0.32	-0.57	0.07	0.23	0.24	-0.27
unregularized	C_lazy.N_worried	1.18	0.00	0.30	0.28	0.51	-0.04	0.29	-0.42
unregularized	C_reliable.E_outgoing	1.07	0.08	0.30	0.10	2.03	0.11	0.24	0.48
unregularized	C_reliable.E_quiet	1.44	-0.05	0.31	0.30	2.21	-0.01	0.24	0.32
unregularized	C_reliable.N_depressed	0.38	-0.06	0.33	0.35	1.61	-0.06	0.26	0.16
unregularized	C_reliable.N_relaxed	1.20	0.04	0.31	-0.19	1.64	0.05	0.23	0.22
unregularized	C_reliable.N_worried	0.95	0.02	0.32	-0.15	0.89	-0.06	0.27	-0.59
unregularized	E_outgoing.E_quiet	4.99	-0.65	0.27	1.76	3.47	-0.68	0.22	1.48
unregularized	E_outgoing.N_depressed	0.96	-0.09	0.30	0.26	2.43	-0.03	0.26	0.55
unregularized	E_outgoing.N_relaxed	0.32	0.06	0.33	-0.09	1.15	0.09	0.25	0.52
unregularized	E_outgoing.N_worried	0.95	-0.10	0.30	-0.10	2.93	-0.07	0.25	0.01
unregularized	E_quiet.N_depressed	0.80	0.00	0.30	0.14	1.18	0.07	0.22	0.64
unregularized	E_quiet.N_relaxed	0.55	-0.03	0.33	-0.22	1.48	0.07	0.22	0.71
unregularized	E_quiet.N_worried	0.87	0.04	0.32	0.16	0.91	0.02	0.23	-0.09
unregularized	N_depressed.N_relaxed	1.34	-0.15	0.31	0.51	1.53	-0.11	0.27	0.90
unregularized	N_depressed.N_worried	1.53	0.22	0.32	-0.71	0.40	0.25	0.29	-0.10
unregularized	N_relaxed.N_worried	2.61	-0.39	0.31	1.13	0.05	-0.40	0.22	0.32

# get SIDs from models
SID_w1 <- as.character(unique((gVAR_fit %>% filter(wave == "1"))$SID))
SID_w2 <- as.character(unique((gVAR_fit %>% filter(wave == "2"))$SID))

# find subjects in both waves
w1w2_subs <- unique(SID_w1)[unique(SID_w1) %in% unique(SID_w2)]

## figure  
contemp_long %>% 
  filter(lambda_scale == ".25") %>%
  ggplot(aes(x = weight, y = var, fill = factor(wave))) +
  # tidybayes::geom_halfeyeh() +
  geom_density_ridges(rel_min_height = 0.025) +
  # geom_density() +
  labs(x = "From", y = "Edge Weight", 
       title = "Regularized Contemporaneous Partial Correlations") +
  facet_grid(~wave) +
  theme_classic() +
  theme(legend.position = "none")

Plots (Figures 2 & 3)

It’s easy to create simple plots of graphivalVAR objects in R. The code below exactly reproduces Figures 2 and 3 in the manuscript.

edge_colors <- RColorBrewer::brewer.pal(8, "Purples")[c(3,4,6)]
idio_plot_fun <- function(data, subject, wave, type){
  if(type == "Lagged"){data_mod <- data$PDC}
  else{data_mod <- data$PCC}
  b5_groups <- list(A = c(1,7), E = c(2, 6), C = c(3,8), N = c(4,5,9))
  subject <- ifelse(subject == "22652", "1", 
             ifelse(subject == "91339", "2", subject))
  plot <- 
    qgraph(data_mod, layout = "spring", loop = .7, node.width = 1.85, edge.width = 1,
           esize = 7, title = sprintf("%s Wave %s for S%s", type, wave, subject),
           label.font = 2, repulsion = .8, label.fill.vertical = 1, 
           label.fill.horizontal = 1, edge.color = "black", groups = b5_groups, 
           color = rev(t(RColorBrewer::brewer.pal(9, "Purples")[seq(1,7,2)])),
           legend = F, DoNotPlot = TRUE, mar = c(4,4,4,4))
  #change lines to dashed
  plot$graphAttributes$Edges$lty[plot$Edgelist$weight < 0] <- 2
  #change line colors
  plot$graphAttributes$Edges$color <-
    ifelse(abs(plot$Edgelist$weight) <.1, edge_colors[1],
    ifelse(abs(plot$Edgelist$weight) <.2, edge_colors[2], edge_colors[3]))
  # change labels of dark nodes to white
  dark_colors <- c("#9E9AC8", "#807DBA", "#6A51A3", "#54278F", "#3F007D")
  plot$graphAttributes$Nodes$label.color[plot$graphAttributes$Nodes$color %in% dark_colors] <- "white"
  #change variable names
  plot$graphAttributes$Nodes$labels <- gsub("_", "\n", names(plot$graphAttributes$Nodes$labels))
  return(plot)
}

gVAR_fit <- gVAR_fit %>%
  mutate(temp_plot = pmap(list(gVAR_fit, SID, wave, "Lagged"),
                          possibly(idio_plot_fun, NA_real_)),
         contemp_plot = pmap(list(gVAR_fit, SID, wave, "Contemporaneous"),
                          possibly(idio_plot_fun, NA_real_)))

Below we will print the two example subjects (22652 = 1; 91339 = 2) from the manuscript.

Subject 22652

par(mfrow = c(2,2))
gVAR_fit %>% filter(SID %in% c("22652") & lambda_scale == ".25") %>% 
  mutate(map(temp_plot, plot), map(contemp_plot, plot))

Subject 91339

par(mfrow = c(2,2))
gVAR_fit %>% filter(SID %in% c("91339") & lambda_scale == ".25") %>% 
  mutate(map(temp_plot, plot), map(contemp_plot, plot))

Tables

We might also want to see a table of the individual edge weights over time, so below we produce a table for both contemporaneous and lagged network edges.

Lagged

gVAR_fit %>% filter(SID %in% c("22652", "91339")) %>%
  unnest(temp) %>% select(SID, wave, lambda_scale, from, to, weight) %>%
  mutate_at(vars(from,to), funs(str_replace(., "_", " "))) %>%
  unite(temp, SID, wave, sep = ".") %>%
  spread(key = temp, value = weight) %>%
  kable(., "html", booktabs = T, escape = F, digits = 2,
        col.names = c("Network Type", "From", "To", "W1", "W2", "W1", "W2"),
        align = c("l", "l", "l", rep("c", 4)),
        caption = "Lagged Networks") %>%
  add_header_above(c(" " = 3, "S1" = 2,"S2" = 2)) %>%
  kable_styling(bootstrap_options = c("striped"), full_width = T) %>%
  scroll_box(width = "750px", height = "400px")

Lagged Networks

			S1		S2
Network Type	From	To	W1	W2	W1	W2
.25	A kind	A kind	0.00	0.00	0.00	0.22
.25	A kind	A rude	0.12	0.00	0.00	0.13
.25	A kind	C lazy	0.00	0.00	0.00	0.00
.25	A kind	C reliable	0.00	0.00	0.00	0.00
.25	A kind	E outgoing	0.00	0.00	0.00	0.11
.25	A kind	E quiet	0.00	0.00	0.00	0.00
.25	A kind	N depressed	0.00	0.00	0.00	0.00
.25	A kind	N relaxed	0.00	0.00	0.00	0.00
.25	A kind	N worried	0.00	0.00	0.00	0.00
.25	A rude	A kind	0.00	0.00	0.00	-0.07
.25	A rude	A rude	0.00	0.12	-0.04	0.00
.25	A rude	C lazy	0.00	0.00	0.00	0.00
.25	A rude	C reliable	0.00	0.00	0.00	0.00
.25	A rude	E outgoing	0.00	0.00	0.00	0.00
.25	A rude	E quiet	0.00	0.00	0.00	0.00
.25	A rude	N depressed	0.00	0.00	0.00	0.00
.25	A rude	N relaxed	0.00	0.00	0.00	0.05
.25	A rude	N worried	0.00	0.00	0.00	0.00
.25	C lazy	A kind	0.00	0.20	0.20	0.00
.25	C lazy	A rude	0.31	0.00	0.37	0.00
.25	C lazy	C lazy	0.05	0.00	0.18	-0.15
.25	C lazy	C reliable	0.00	0.00	-0.26	-0.15
.25	C lazy	E outgoing	0.00	0.00	0.00	0.11
.25	C lazy	E quiet	0.00	0.00	-0.12	0.02
.25	C lazy	N depressed	0.00	0.00	0.07	0.00
.25	C lazy	N relaxed	0.00	0.00	0.00	0.00
.25	C lazy	N worried	0.00	0.00	0.00	0.00
.25	C reliable	A kind	0.00	0.00	0.00	-0.05
.25	C reliable	A rude	0.54	0.00	0.23	0.00
.25	C reliable	C lazy	0.00	0.00	0.00	0.32
.25	C reliable	C reliable	0.00	0.00	-0.32	0.00
.25	C reliable	E outgoing	0.00	0.00	0.00	0.00
.25	C reliable	E quiet	0.00	0.00	0.00	0.00
.25	C reliable	N depressed	0.00	0.00	0.00	0.04
.25	C reliable	N relaxed	0.00	0.00	0.00	0.00
.25	C reliable	N worried	0.00	0.00	0.00	0.00
.25	E outgoing	A kind	0.00	-0.20	-0.19	-0.24
.25	E outgoing	A rude	-0.31	0.00	-0.01	0.00
.25	E outgoing	C lazy	0.00	0.02	0.00	0.00
.25	E outgoing	C reliable	0.00	0.00	0.00	0.00
.25	E outgoing	E outgoing	0.00	0.00	0.00	-0.08
.25	E outgoing	E quiet	0.00	0.00	0.00	0.00
.25	E outgoing	N depressed	0.00	0.00	-0.18	0.00
.25	E outgoing	N relaxed	0.00	0.00	0.00	0.00
.25	E outgoing	N worried	0.00	0.00	0.00	0.00
.25	E quiet	A kind	0.00	0.00	0.19	0.00
.25	E quiet	A rude	0.17	-0.09	0.14	0.00
.25	E quiet	C lazy	0.00	0.00	0.00	0.02
.25	E quiet	C reliable	0.00	-0.02	0.42	0.00
.25	E quiet	E outgoing	0.00	0.00	0.00	0.00
.25	E quiet	E quiet	0.00	0.00	0.00	0.00
.25	E quiet	N depressed	-0.03	0.00	0.02	0.00
.25	E quiet	N relaxed	0.00	0.00	-0.16	-0.23
.25	E quiet	N worried	0.00	0.01	0.00	-0.08
.25	N depressed	A kind	0.00	-0.05	-0.04	0.00
.25	N depressed	A rude	0.32	0.32	0.00	0.00
.25	N depressed	C lazy	0.00	0.00	0.00	0.00
.25	N depressed	C reliable	0.00	0.00	0.00	0.00
.25	N depressed	E outgoing	0.00	-0.07	-0.01	0.18
.25	N depressed	E quiet	0.00	0.00	0.00	0.00
.25	N depressed	N depressed	0.07	0.00	0.00	0.28
.25	N depressed	N relaxed	0.00	0.00	0.00	0.00
.25	N depressed	N worried	0.00	0.00	0.04	0.22
.25	N relaxed	A kind	0.00	0.00	0.00	0.00
.25	N relaxed	A rude	-0.61	0.00	0.00	0.00
.25	N relaxed	C lazy	0.00	0.00	0.08	0.00
.25	N relaxed	C reliable	0.00	0.00	0.03	-0.03
.25	N relaxed	E outgoing	0.00	0.00	0.00	0.00
.25	N relaxed	E quiet	0.00	0.00	0.00	0.00
.25	N relaxed	N depressed	0.00	0.00	0.00	0.00
.25	N relaxed	N relaxed	0.00	0.00	0.00	0.00
.25	N relaxed	N worried	0.00	0.00	-0.06	-0.06
.25	N worried	A kind	-0.11	-0.45	0.16	-0.06
.25	N worried	A rude	-0.18	0.00	0.31	0.00
.25	N worried	C lazy	0.00	0.00	0.00	0.00
.25	N worried	C reliable	0.00	0.00	0.00	0.00
.25	N worried	E outgoing	0.00	0.00	0.00	0.00
.25	N worried	E quiet	0.00	0.00	0.00	0.00
.25	N worried	N depressed	0.00	0.00	0.30	0.06
.25	N worried	N relaxed	0.00	0.00	-0.26	-0.08
.25	N worried	N worried	0.00	0.00	0.00	0.00
.5	A kind	A kind	0.00	0.00	0.00	0.00
.5	A kind	A rude	0.13	0.00	0.00	0.00
.5	A kind	C lazy	0.00	0.00	0.00	0.00
.5	A kind	C reliable	0.00	0.00	0.00	0.00
.5	A kind	E outgoing	0.00	0.00	0.00	0.00
.5	A kind	E quiet	0.00	0.00	0.00	0.00
.5	A kind	N depressed	0.00	0.00	0.00	0.00
.5	A kind	N relaxed	0.00	0.00	0.00	0.00
.5	A kind	N worried	0.00	0.00	0.00	0.00
.5	A rude	A kind	0.00	0.00	0.00	0.00
.5	A rude	A rude	0.00	0.00	0.00	0.00
.5	A rude	C lazy	0.00	0.00	0.00	0.00
.5	A rude	C reliable	0.00	0.00	0.00	0.00
.5	A rude	E outgoing	0.00	0.00	0.00	0.00
.5	A rude	E quiet	0.00	0.00	0.00	0.00
.5	A rude	N depressed	0.00	0.00	0.00	0.00
.5	A rude	N relaxed	0.00	0.00	0.00	0.00
.5	A rude	N worried	0.00	0.00	0.00	0.00
.5	C lazy	A kind	0.00	0.14	0.12	0.00
.5	C lazy	A rude	0.28	0.00	0.32	0.00
.5	C lazy	C lazy	0.00	0.00	0.00	0.00
.5	C lazy	C reliable	0.00	0.00	0.00	0.00
.5	C lazy	E outgoing	0.00	0.00	0.00	0.00
.5	C lazy	E quiet	0.00	0.00	0.00	0.00
.5	C lazy	N depressed	0.00	0.00	0.00	0.00
.5	C lazy	N relaxed	0.00	0.00	0.00	0.00
.5	C lazy	N worried	0.00	0.00	0.00	0.00
.5	C reliable	A kind	0.00	0.00	0.00	0.00
.5	C reliable	A rude	0.50	0.00	0.16	0.00
.5	C reliable	C lazy	0.00	0.00	0.00	0.00
.5	C reliable	C reliable	0.00	0.00	0.00	0.00
.5	C reliable	E outgoing	0.00	0.00	0.00	0.00
.5	C reliable	E quiet	0.00	0.00	0.00	0.00
.5	C reliable	N depressed	0.00	0.00	0.00	0.00
.5	C reliable	N relaxed	0.00	0.00	0.00	0.00
.5	C reliable	N worried	0.00	0.00	0.00	0.00
.5	E outgoing	A kind	0.00	-0.13	-0.16	0.00
.5	E outgoing	A rude	-0.29	0.00	0.00	0.00
.5	E outgoing	C lazy	0.00	0.00	0.00	0.00
.5	E outgoing	C reliable	0.00	0.00	0.00	0.00
.5	E outgoing	E outgoing	0.00	0.00	0.00	0.00
.5	E outgoing	E quiet	0.00	0.00	0.00	0.00
.5	E outgoing	N depressed	0.00	0.00	0.00	0.00
.5	E outgoing	N relaxed	0.00	0.00	0.00	0.00
.5	E outgoing	N worried	0.00	0.00	0.00	0.00
.5	E quiet	A kind	0.00	0.00	0.14	0.00
.5	E quiet	A rude	0.17	-0.04	0.18	0.00
.5	E quiet	C lazy	0.00	0.00	0.00	0.00
.5	E quiet	C reliable	0.00	0.00	0.00	0.00
.5	E quiet	E outgoing	0.00	0.00	0.00	0.00
.5	E quiet	E quiet	0.00	0.00	0.00	0.00
.5	E quiet	N depressed	0.00	0.00	0.00	0.00
.5	E quiet	N relaxed	0.00	0.00	0.00	0.00
.5	E quiet	N worried	0.00	0.00	0.00	0.00
.5	N depressed	A kind	0.00	0.00	0.00	0.00
.5	N depressed	A rude	0.29	0.22	0.00	0.00
.5	N depressed	C lazy	0.00	0.00	0.00	0.00
.5	N depressed	C reliable	0.00	0.00	0.00	0.00
.5	N depressed	E outgoing	0.00	0.00	0.00	0.00
.5	N depressed	E quiet	0.00	0.00	0.00	0.00
.5	N depressed	N depressed	0.00	0.00	0.00	0.00
.5	N depressed	N relaxed	0.00	0.00	0.00	0.00
.5	N depressed	N worried	0.00	0.00	0.00	0.00
.5	N relaxed	A kind	0.00	0.00	0.00	0.00
.5	N relaxed	A rude	-0.59	0.00	0.00	0.00
.5	N relaxed	C lazy	0.00	0.00	0.00	0.00
.5	N relaxed	C reliable	0.00	0.00	0.00	0.00
.5	N relaxed	E outgoing	0.00	0.00	0.00	0.00
.5	N relaxed	E quiet	0.00	0.00	0.00	0.00
.5	N relaxed	N depressed	0.00	0.00	0.00	0.00
.5	N relaxed	N relaxed	0.00	0.00	0.00	0.00
.5	N relaxed	N worried	0.00	0.00	0.00	0.00
.5	N worried	A kind	0.00	-0.38	0.03	0.00
.5	N worried	A rude	-0.16	0.00	0.22	0.00
.5	N worried	C lazy	0.00	0.00	0.00	0.00
.5	N worried	C reliable	0.00	0.00	0.00	0.00
.5	N worried	E outgoing	0.00	0.00	0.00	0.00
.5	N worried	E quiet	0.00	0.00	0.00	0.00
.5	N worried	N depressed	0.00	0.00	0.00	0.00
.5	N worried	N relaxed	0.00	0.00	0.00	0.00
.5	N worried	N worried	0.00	0.00	0.00	0.00
association	A kind	A kind	0.08	-0.05	0.25	0.40
association	A kind	A rude	0.31	0.04	0.01	0.30
association	A kind	C lazy	0.16	0.27	-0.04	0.33
association	A kind	C reliable	-0.14	-0.34	-0.02	-0.07
association	A kind	E outgoing	-0.11	-0.01	-0.41	0.24
association	A kind	E quiet	0.31	0.13	0.33	-0.30
association	A kind	N depressed	-0.05	-0.03	0.56	-0.03
association	A kind	N relaxed	0.07	0.35	-0.29	0.35
association	A kind	N worried	-0.04	-0.23	0.47	-0.18
association	A rude	A kind	0.07	-0.30	-0.02	-0.34
association	A rude	A rude	-0.15	0.29	-0.14	-0.04
association	A rude	C lazy	-0.01	-0.04	-0.21	0.41
association	A rude	C reliable	0.16	0.07	-0.23	0.02
association	A rude	E outgoing	0.20	-0.17	-0.03	-0.11
association	A rude	E quiet	-0.23	0.12	-0.28	0.19
association	A rude	N depressed	-0.24	0.24	0.14	0.02
association	A rude	N relaxed	0.32	-0.02	-0.11	0.37
association	A rude	N worried	-0.24	0.28	0.17	-0.05
association	C lazy	A kind	-0.05	0.23	0.27	-0.07
association	C lazy	A rude	0.04	0.01	0.29	0.00
association	C lazy	C lazy	0.33	0.11	0.50	-0.38
association	C lazy	C reliable	-0.21	-0.17	-0.34	-0.35
association	C lazy	E outgoing	-0.05	-0.12	0.07	0.13
association	C lazy	E quiet	0.02	0.09	-0.39	0.22
association	C lazy	N depressed	0.15	-0.10	-0.02	-0.20
association	C lazy	N relaxed	0.07	-0.05	0.26	0.02
association	C lazy	N worried	-0.02	-0.04	-0.21	-0.02
association	C reliable	A kind	0.05	-0.10	0.09	-0.17
association	C reliable	A rude	0.06	0.12	0.31	-0.06
association	C reliable	C lazy	-0.22	-0.17	-0.31	0.43
association	C reliable	C reliable	0.16	0.25	-0.33	-0.15
association	C reliable	E outgoing	0.01	0.15	0.20	0.01
association	C reliable	E quiet	0.07	-0.13	-0.31	-0.14
association	C reliable	N depressed	-0.16	0.02	-0.15	0.22
association	C reliable	N relaxed	-0.16	0.01	-0.14	-0.02
association	C reliable	N worried	0.06	0.03	0.01	0.16
association	E outgoing	A kind	0.08	-0.20	-0.41	-0.39
association	E outgoing	A rude	0.17	-0.01	-0.24	0.17
association	E outgoing	C lazy	-0.16	0.28	-0.06	0.03
association	E outgoing	C reliable	0.10	-0.18	-0.22	-0.21
association	E outgoing	E outgoing	0.03	0.01	0.17	-0.28
association	E outgoing	E quiet	-0.08	-0.04	-0.16	0.00
association	E outgoing	N depressed	-0.04	0.09	-0.43	0.04
association	E outgoing	N relaxed	-0.02	0.05	0.12	0.19
association	E outgoing	N worried	-0.01	0.10	-0.05	-0.12
association	E quiet	A kind	-0.10	0.09	0.34	0.12
association	E quiet	A rude	-0.03	-0.15	0.36	-0.02
association	E quiet	C lazy	-0.04	-0.02	-0.27	0.18
association	E quiet	C reliable	0.02	-0.13	0.32	0.13
association	E quiet	E outgoing	-0.05	-0.06	0.13	0.12
association	E quiet	E quiet	-0.03	0.11	0.01	-0.04
association	E quiet	N depressed	-0.19	0.11	0.30	-0.03
association	E quiet	N relaxed	-0.01	-0.09	-0.46	-0.17
association	E quiet	N worried	-0.08	0.17	0.18	-0.23
association	N depressed	A kind	-0.03	-0.30	0.01	-0.19
association	N depressed	A rude	-0.16	0.40	0.11	-0.14
association	N depressed	C lazy	0.12	-0.15	-0.42	-0.11
association	N depressed	C reliable	-0.13	0.22	0.16	0.11
association	N depressed	E outgoing	0.12	-0.36	-0.24	0.24
association	N depressed	E quiet	-0.20	0.09	0.17	0.17
association	N depressed	N depressed	0.34	-0.05	0.46	0.55
association	N depressed	N relaxed	0.14	-0.26	-0.41	-0.25
association	N depressed	N worried	0.13	-0.10	0.43	0.50
association	N relaxed	A kind	0.06	0.12	-0.03	0.10
association	N relaxed	A rude	0.25	-0.04	-0.01	0.06
association	N relaxed	C lazy	0.19	0.15	0.51	-0.02
association	N relaxed	C reliable	-0.14	-0.11	0.09	-0.31
association	N relaxed	E outgoing	0.17	-0.15	0.19	0.04
association	N relaxed	E quiet	-0.09	0.20	-0.11	0.22
association	N relaxed	N depressed	0.09	-0.01	-0.41	-0.20
association	N relaxed	N relaxed	0.06	-0.02	0.38	0.23
association	N relaxed	N worried	-0.13	-0.03	-0.39	-0.42
association	N worried	A kind	-0.24	-0.47	0.28	-0.31
association	N worried	A rude	-0.35	0.23	0.30	-0.03
association	N worried	C lazy	-0.24	0.00	-0.37	-0.01
association	N worried	C reliable	0.14	-0.02	0.17	0.12
association	N worried	E outgoing	0.21	-0.38	-0.21	-0.07
association	N worried	E quiet	-0.12	0.25	-0.02	0.07
association	N worried	N depressed	0.06	0.08	0.65	0.42
association	N worried	N relaxed	-0.21	-0.03	-0.67	-0.36
association	N worried	N worried	0.22	0.03	0.44	0.31
gimme	A kind	A kind	0.19	-0.13	0.24	0.46
gimme	A kind	A rude	0.00	0.00	0.00	-0.33
gimme	A kind	C lazy	0.00	0.34	0.00	0.00
gimme	A kind	C reliable	0.00	0.00	0.00	0.00
gimme	A kind	E outgoing	0.00	0.00	0.00	-0.33
gimme	A kind	E quiet	0.00	0.00	0.00	0.00
gimme	A kind	N depressed	0.00	0.00	0.00	0.00
gimme	A kind	N relaxed	0.00	0.00	0.00	0.00
gimme	A kind	N worried	0.00	-0.56	0.00	0.00
gimme	A rude	A kind	0.00	0.00	-0.40	0.33
gimme	A rude	A rude	-0.10	0.33	0.07	-0.03
gimme	A rude	C lazy	0.00	0.00	0.97	0.00
gimme	A rude	C reliable	0.00	0.00	0.95	0.00
gimme	A rude	E outgoing	0.00	0.00	0.00	0.00
gimme	A rude	E quiet	0.00	0.00	0.00	0.00
gimme	A rude	N depressed	0.00	0.43	0.00	0.00
gimme	A rude	N relaxed	0.00	0.00	0.00	0.00
gimme	A rude	N worried	0.00	0.00	0.90	0.00
gimme	C lazy	A kind	0.00	0.00	0.00	0.00
gimme	C lazy	A rude	0.00	0.00	0.00	0.00
gimme	C lazy	C lazy	0.33	-0.05	0.36	-0.30
gimme	C lazy	C reliable	0.00	0.00	0.00	0.46
gimme	C lazy	E outgoing	0.00	0.00	0.00	0.00
gimme	C lazy	E quiet	0.00	0.00	0.00	0.32
gimme	C lazy	N depressed	0.00	0.00	0.00	0.00
gimme	C lazy	N relaxed	0.00	0.00	0.00	0.00
gimme	C lazy	N worried	0.00	0.00	0.00	0.00
gimme	C reliable	A kind	0.00	-0.54	0.00	0.00
gimme	C reliable	A rude	0.00	0.00	-0.44	0.00
gimme	C reliable	C lazy	0.00	0.00	0.00	-0.41
gimme	C reliable	C reliable	0.00	0.22	-0.56	-0.25
gimme	C reliable	E outgoing	0.00	0.00	-0.51	0.00
gimme	C reliable	E quiet	0.00	0.00	0.43	0.00
gimme	C reliable	N depressed	0.00	0.00	0.00	0.00
gimme	C reliable	N relaxed	0.00	0.00	0.00	0.00
gimme	C reliable	N worried	0.00	0.00	0.00	0.00
gimme	E outgoing	A kind	0.00	0.00	-0.39	0.00
gimme	E outgoing	A rude	0.00	0.00	0.00	0.00
gimme	E outgoing	C lazy	0.00	0.00	0.00	0.00
gimme	E outgoing	C reliable	0.00	0.00	0.00	0.00
gimme	E outgoing	E outgoing	-0.03	-0.08	0.08	-0.30
gimme	E outgoing	E quiet	0.00	0.00	0.00	0.00
gimme	E outgoing	N depressed	0.00	-0.32	0.00	0.41
gimme	E outgoing	N relaxed	0.00	0.00	0.00	0.00
gimme	E outgoing	N worried	0.00	0.00	0.00	0.00
gimme	E quiet	A kind	0.00	0.00	0.00	0.00
gimme	E quiet	A rude	0.00	0.00	0.00	0.00
gimme	E quiet	C lazy	0.00	0.00	0.00	0.00
gimme	E quiet	C reliable	0.00	0.00	0.00	0.00
gimme	E quiet	E outgoing	0.00	0.00	0.00	0.00
gimme	E quiet	E quiet	-0.06	0.15	-0.24	0.02
gimme	E quiet	N depressed	0.00	0.00	0.00	0.00
gimme	E quiet	N relaxed	0.00	0.00	0.00	0.00
gimme	E quiet	N worried	0.00	0.00	0.00	0.00
gimme	N depressed	A kind	0.00	0.00	0.00	0.00
gimme	N depressed	A rude	0.00	0.00	0.00	0.00
gimme	N depressed	C lazy	0.00	0.00	0.88	0.00
gimme	N depressed	C reliable	0.00	0.00	0.00	0.00
gimme	N depressed	E outgoing	0.00	0.00	0.00	0.00
gimme	N depressed	E quiet	0.00	0.00	0.00	0.00
gimme	N depressed	N depressed	0.27	-0.05	-0.21	0.55
gimme	N depressed	N relaxed	0.00	0.00	-0.80	0.00
gimme	N depressed	N worried	0.00	0.00	0.00	0.00
gimme	N relaxed	A kind	0.00	0.00	0.00	0.00
gimme	N relaxed	A rude	0.00	0.00	0.00	0.33
gimme	N relaxed	C lazy	0.00	0.00	0.00	0.00
gimme	N relaxed	C reliable	0.00	0.00	0.00	0.00
gimme	N relaxed	E outgoing	0.00	0.00	0.00	0.00
gimme	N relaxed	E quiet	0.00	0.00	-0.34	-0.47
gimme	N relaxed	N depressed	0.00	0.00	0.00	0.30
gimme	N relaxed	N relaxed	-0.09	-0.15	-0.15	-0.06
gimme	N relaxed	N worried	0.00	0.00	-0.77	0.00
gimme	N worried	A kind	0.00	0.00	0.39	-0.37
gimme	N worried	A rude	0.00	0.00	0.00	0.00
gimme	N worried	C lazy	0.00	0.00	0.00	0.00
gimme	N worried	C reliable	0.00	0.00	0.00	0.00
gimme	N worried	E outgoing	0.00	0.00	0.00	0.00
gimme	N worried	E quiet	0.00	0.00	0.00	-0.32
gimme	N worried	N depressed	0.00	0.00	0.00	0.00
gimme	N worried	N relaxed	0.00	0.00	0.00	-0.50
gimme	N worried	N worried	0.20	-0.09	0.35	-0.28
mlVAR_EB	A kind	A kind	0.02	0.07	0.05	0.14
mlVAR_EB	A kind	A rude	-0.01	0.00	-0.02	0.00
mlVAR_EB	A kind	C lazy	0.00	0.02	0.02	0.02
mlVAR_EB	A kind	C reliable	0.03	-0.01	0.03	-0.01
mlVAR_EB	A kind	E outgoing	0.01	-0.02	0.00	-0.02
mlVAR_EB	A kind	E quiet	0.00	0.01	0.00	0.01
mlVAR_EB	A kind	N depressed	0.02	-0.06	0.03	-0.04
mlVAR_EB	A kind	N relaxed	0.00	-0.02	0.00	-0.01
mlVAR_EB	A kind	N worried	-0.02	-0.05	-0.01	-0.05
mlVAR_EB	A rude	A kind	0.00	0.02	0.00	0.07
mlVAR_EB	A rude	A rude	0.04	0.09	0.00	0.01
mlVAR_EB	A rude	C lazy	-0.01	0.04	0.01	0.01
mlVAR_EB	A rude	C reliable	-0.01	0.00	0.01	-0.01
mlVAR_EB	A rude	E outgoing	0.00	0.00	-0.01	0.00
mlVAR_EB	A rude	E quiet	0.01	-0.01	0.01	-0.01
mlVAR_EB	A rude	N depressed	-0.02	0.00	-0.01	0.00
mlVAR_EB	A rude	N relaxed	0.01	0.01	0.01	0.01
mlVAR_EB	A rude	N worried	0.00	0.05	0.01	0.02
mlVAR_EB	C lazy	A kind	-0.04	0.09	-0.03	0.05
mlVAR_EB	C lazy	A rude	0.02	-0.01	-0.01	0.06
mlVAR_EB	C lazy	C lazy	0.05	0.06	0.22	-0.05
mlVAR_EB	C lazy	C reliable	0.01	0.02	0.01	0.02
mlVAR_EB	C lazy	E outgoing	-0.04	0.02	0.00	0.02
mlVAR_EB	C lazy	E quiet	0.00	-0.01	-0.01	-0.01
mlVAR_EB	C lazy	N depressed	0.03	0.02	-0.02	0.01
mlVAR_EB	C lazy	N relaxed	0.05	0.02	0.05	0.01
mlVAR_EB	C lazy	N worried	-0.02	-0.04	-0.07	-0.05
mlVAR_EB	C reliable	A kind	0.05	-0.04	0.00	0.00
mlVAR_EB	C reliable	A rude	-0.02	0.03	-0.02	0.01
mlVAR_EB	C reliable	C lazy	-0.02	-0.03	-0.02	-0.03
mlVAR_EB	C reliable	C reliable	0.03	0.09	-0.05	0.01
mlVAR_EB	C reliable	E outgoing	0.01	-0.02	0.00	-0.02
mlVAR_EB	C reliable	E quiet	0.01	0.00	0.01	0.00
mlVAR_EB	C reliable	N depressed	-0.02	0.04	0.00	0.01
mlVAR_EB	C reliable	N relaxed	-0.01	0.01	-0.01	0.01
mlVAR_EB	C reliable	N worried	0.00	-0.01	0.00	-0.01
mlVAR_EB	E outgoing	A kind	-0.03	0.02	-0.03	0.02
mlVAR_EB	E outgoing	A rude	-0.03	-0.01	-0.03	-0.01
mlVAR_EB	E outgoing	C lazy	-0.03	0.03	-0.02	0.04
mlVAR_EB	E outgoing	C reliable	-0.03	0.06	0.03	0.05
mlVAR_EB	E outgoing	E outgoing	0.05	0.02	0.01	0.02
mlVAR_EB	E outgoing	E quiet	0.01	0.00	0.01	0.00
mlVAR_EB	E outgoing	N depressed	0.03	-0.08	0.00	0.00
mlVAR_EB	E outgoing	N relaxed	0.03	0.01	0.02	0.01
mlVAR_EB	E outgoing	N worried	-0.08	-0.05	-0.07	-0.03
mlVAR_EB	E quiet	A kind	0.04	-0.02	0.02	-0.02
mlVAR_EB	E quiet	A rude	0.03	0.01	0.03	0.01
mlVAR_EB	E quiet	C lazy	0.04	-0.02	0.00	-0.02
mlVAR_EB	E quiet	C reliable	0.01	-0.03	-0.04	-0.04
mlVAR_EB	E quiet	E outgoing	-0.01	0.01	-0.01	0.01
mlVAR_EB	E quiet	E quiet	-0.01	0.02	-0.01	0.03
mlVAR_EB	E quiet	N depressed	-0.01	0.00	0.00	0.00
mlVAR_EB	E quiet	N relaxed	-0.01	0.11	-0.01	0.04
mlVAR_EB	E quiet	N worried	0.05	0.17	0.03	0.09
mlVAR_EB	N depressed	A kind	0.01	0.03	0.01	0.03
mlVAR_EB	N depressed	A rude	0.03	0.03	0.03	0.03
mlVAR_EB	N depressed	C lazy	0.01	-0.02	0.02	-0.01
mlVAR_EB	N depressed	C reliable	0.00	-0.02	-0.04	-0.01
mlVAR_EB	N depressed	E outgoing	0.01	0.03	0.01	0.00
mlVAR_EB	N depressed	E quiet	0.00	0.00	0.00	0.00
mlVAR_EB	N depressed	N depressed	0.09	-0.01	0.20	0.37
mlVAR_EB	N depressed	N relaxed	-0.02	0.00	-0.02	0.00
mlVAR_EB	N depressed	N worried	0.08	0.01	0.16	0.06
mlVAR_EB	N relaxed	A kind	-0.01	0.07	-0.01	0.04
mlVAR_EB	N relaxed	A rude	-0.05	-0.02	-0.06	-0.02
mlVAR_EB	N relaxed	C lazy	-0.01	0.03	0.02	0.03
mlVAR_EB	N relaxed	C reliable	-0.04	0.02	-0.03	0.02
mlVAR_EB	N relaxed	E outgoing	-0.02	-0.02	-0.02	-0.02
mlVAR_EB	N relaxed	E quiet	0.00	-0.06	-0.01	-0.05
mlVAR_EB	N relaxed	N depressed	-0.01	-0.08	-0.01	-0.06
mlVAR_EB	N relaxed	N relaxed	0.02	0.03	0.03	0.05
mlVAR_EB	N relaxed	N worried	-0.09	-0.07	-0.25	-0.09
mlVAR_EB	N worried	A kind	0.02	-0.02	0.02	0.01
mlVAR_EB	N worried	A rude	0.04	0.05	0.05	0.05
mlVAR_EB	N worried	C lazy	0.01	0.01	0.01	0.00
mlVAR_EB	N worried	C reliable	0.08	-0.03	0.00	-0.01
mlVAR_EB	N worried	E outgoing	0.01	0.04	0.01	0.04
mlVAR_EB	N worried	E quiet	0.01	0.02	0.01	0.01
mlVAR_EB	N worried	N depressed	0.01	-0.04	0.04	0.07
mlVAR_EB	N worried	N relaxed	-0.05	0.02	-0.05	-0.05
mlVAR_EB	N worried	N worried	0.08	0.12	0.18	0.20
scaled	A kind	A kind	0.00	0.00	0.00	0.17
scaled	A kind	A rude	0.14	0.00	0.00	0.08
scaled	A kind	C lazy	0.00	0.00	0.00	0.00
scaled	A kind	C reliable	0.00	0.00	0.00	0.00
scaled	A kind	E outgoing	0.00	0.00	0.00	0.06
scaled	A kind	E quiet	0.00	0.00	0.00	0.00
scaled	A kind	N depressed	0.00	0.00	0.12	0.00
scaled	A kind	N relaxed	0.00	0.00	0.00	0.00
scaled	A kind	N worried	0.00	0.00	0.01	0.00
scaled	A rude	A kind	0.00	0.00	0.00	0.00
scaled	A rude	A rude	0.00	0.00	-0.05	0.00
scaled	A rude	C lazy	0.00	0.00	0.00	0.00
scaled	A rude	C reliable	0.00	0.00	0.00	0.00
scaled	A rude	E outgoing	0.00	0.00	0.00	0.00
scaled	A rude	E quiet	0.00	0.00	0.00	0.00
scaled	A rude	N depressed	0.00	0.00	0.00	0.00
scaled	A rude	N relaxed	0.00	0.00	0.00	0.00
scaled	A rude	N worried	0.00	0.00	0.00	0.00
scaled	C lazy	A kind	0.00	0.10	0.21	0.00
scaled	C lazy	A rude	0.30	0.00	0.40	0.00
scaled	C lazy	C lazy	0.00	0.00	0.20	-0.13
scaled	C lazy	C reliable	0.00	0.00	-0.44	-0.15
scaled	C lazy	E outgoing	0.00	0.00	0.00	0.04
scaled	C lazy	E quiet	0.00	0.00	-0.17	0.00
scaled	C lazy	N depressed	0.00	0.00	0.05	0.00
scaled	C lazy	N relaxed	0.00	0.00	0.00	0.00
scaled	C lazy	N worried	0.00	0.00	0.00	0.00
scaled	C reliable	A kind	0.00	0.00	0.00	0.00
scaled	C reliable	A rude	0.58	0.00	0.22	0.00
scaled	C reliable	C lazy	0.00	0.00	0.00	0.19
scaled	C reliable	C reliable	0.00	0.00	-0.36	0.00
scaled	C reliable	E outgoing	0.00	0.00	0.00	0.00
scaled	C reliable	E quiet	0.00	0.00	0.00	0.00
scaled	C reliable	N depressed	0.00	0.00	0.00	0.00
scaled	C reliable	N relaxed	0.00	0.00	0.00	0.00
scaled	C reliable	N worried	0.00	0.00	0.00	0.00
scaled	E outgoing	A kind	0.00	-0.08	-0.21	-0.13
scaled	E outgoing	A rude	-0.31	0.00	-0.02	0.00
scaled	E outgoing	C lazy	0.00	0.00	0.00	0.00
scaled	E outgoing	C reliable	0.00	0.00	-0.04	0.00
scaled	E outgoing	E outgoing	0.00	0.00	0.00	-0.02
scaled	E outgoing	E quiet	0.00	0.00	0.00	0.00
scaled	E outgoing	N depressed	0.00	0.00	-0.18	0.00
scaled	E outgoing	N relaxed	0.00	0.00	0.00	0.00
scaled	E outgoing	N worried	0.00	0.00	0.00	0.00
scaled	E quiet	A kind	0.00	0.00	0.21	0.00
scaled	E quiet	A rude	0.20	-0.02	0.15	0.00
scaled	E quiet	C lazy	0.00	0.00	0.00	0.00
scaled	E quiet	C reliable	0.00	0.00	0.48	0.00
scaled	E quiet	E outgoing	0.00	0.00	0.00	0.00
scaled	E quiet	E quiet	0.00	0.00	0.00	0.00
scaled	E quiet	N depressed	0.00	0.00	0.04	0.00
scaled	E quiet	N relaxed	0.00	0.00	-0.19	-0.13
scaled	E quiet	N worried	0.00	0.00	0.00	-0.05
scaled	N depressed	A kind	0.00	0.00	-0.04	0.00
scaled	N depressed	A rude	0.30	0.18	0.00	0.00
scaled	N depressed	C lazy	0.00	0.00	0.00	0.00
scaled	N depressed	C reliable	0.00	0.00	0.00	0.00
scaled	N depressed	E outgoing	0.00	0.00	-0.01	0.10
scaled	N depressed	E quiet	0.00	0.00	0.00	0.00
scaled	N depressed	N depressed	0.00	0.00	0.02	0.28
scaled	N depressed	N relaxed	0.00	0.00	0.00	0.00
scaled	N depressed	N worried	0.00	0.00	0.10	0.20
scaled	N relaxed	A kind	0.00	0.00	0.00	0.00
scaled	N relaxed	A rude	-0.62	0.00	0.00	0.00
scaled	N relaxed	C lazy	0.00	0.00	0.07	0.00
scaled	N relaxed	C reliable	0.00	0.00	0.15	0.00
scaled	N relaxed	E outgoing	0.00	0.00	0.00	0.00
scaled	N relaxed	E quiet	0.00	0.00	0.00	0.00
scaled	N relaxed	N depressed	0.00	0.00	0.00	0.00
scaled	N relaxed	N relaxed	0.00	0.00	0.00	0.00
scaled	N relaxed	N worried	0.00	0.00	-0.05	0.00
scaled	N worried	A kind	0.00	-0.31	0.17	-0.07
scaled	N worried	A rude	-0.15	0.00	0.35	0.00
scaled	N worried	C lazy	0.00	0.00	0.00	0.00
scaled	N worried	C reliable	0.00	0.00	0.00	0.00
scaled	N worried	E outgoing	0.00	0.00	0.00	0.00
scaled	N worried	E quiet	0.00	0.00	0.00	0.00
scaled	N worried	N depressed	0.00	0.00	0.41	0.04
scaled	N worried	N relaxed	0.00	0.00	-0.34	-0.06
scaled	N worried	N worried	0.00	0.00	0.06	0.00
unregularized	A kind	A kind	-0.01	-0.08	0.05	0.33
unregularized	A kind	A rude	0.11	0.22	-0.04	0.23
unregularized	A kind	C lazy	0.01	0.16	0.00	0.27
unregularized	A kind	C reliable	-0.05	-0.27	-0.02	0.06
unregularized	A kind	E outgoing	-0.05	-0.09	-0.13	0.23
unregularized	A kind	E quiet	0.25	0.13	0.14	-0.10
unregularized	A kind	N depressed	-0.12	0.02	0.16	0.03
unregularized	A kind	N relaxed	0.00	0.22	-0.13	0.29
unregularized	A kind	N worried	-0.02	-0.13	0.16	-0.16
unregularized	A rude	A kind	0.14	-0.19	-0.04	-0.26
unregularized	A rude	A rude	0.07	0.32	-0.12	0.06
unregularized	A rude	C lazy	-0.15	0.06	-0.13	0.28
unregularized	A rude	C reliable	0.18	-0.06	-0.10	0.18
unregularized	A rude	E outgoing	-0.01	-0.15	0.04	-0.05
unregularized	A rude	E quiet	-0.15	0.05	-0.20	0.30
unregularized	A rude	N depressed	-0.03	0.18	-0.06	-0.07
unregularized	A rude	N relaxed	0.04	-0.05	-0.09	0.46
unregularized	A rude	N worried	-0.14	0.23	0.05	-0.05
unregularized	C lazy	A kind	0.04	0.29	0.25	-0.15
unregularized	C lazy	A rude	0.22	0.30	0.27	-0.01
unregularized	C lazy	C lazy	0.17	0.00	0.27	-0.19
unregularized	C lazy	C reliable	-0.04	-0.02	-0.42	-0.14
unregularized	C lazy	E outgoing	-0.19	0.03	0.08	0.11
unregularized	C lazy	E quiet	0.09	-0.12	-0.32	0.13
unregularized	C lazy	N depressed	0.06	-0.10	0.20	0.00
unregularized	C lazy	N relaxed	-0.06	-0.10	0.19	-0.03
unregularized	C lazy	N worried	0.09	0.04	-0.11	0.16
unregularized	C reliable	A kind	0.05	0.16	-0.02	-0.12
unregularized	C reliable	A rude	0.41	0.23	0.16	-0.11
unregularized	C reliable	C lazy	0.08	-0.10	-0.12	0.12
unregularized	C reliable	C reliable	-0.01	0.14	-0.19	-0.21
unregularized	C reliable	E outgoing	-0.05	0.09	-0.02	0.00
unregularized	C reliable	E quiet	0.05	-0.10	-0.23	-0.15
unregularized	C reliable	N depressed	0.11	-0.10	-0.12	0.12
unregularized	C reliable	N relaxed	-0.11	-0.03	-0.12	-0.25
unregularized	C reliable	N worried	0.17	-0.02	0.04	0.13
unregularized	E outgoing	A kind	0.01	-0.29	-0.24	-0.25
unregularized	E outgoing	A rude	-0.19	0.11	-0.05	0.06
unregularized	E outgoing	C lazy	-0.16	0.28	-0.11	0.04
unregularized	E outgoing	C reliable	0.11	-0.28	-0.08	-0.17
unregularized	E outgoing	E outgoing	-0.04	-0.13	0.08	-0.11
unregularized	E outgoing	E quiet	-0.11	-0.01	-0.23	-0.07
unregularized	E outgoing	N depressed	-0.09	0.21	-0.11	0.04
unregularized	E outgoing	N relaxed	0.01	-0.14	-0.22	0.06
unregularized	E outgoing	N worried	-0.05	0.35	0.16	-0.15
unregularized	E quiet	A kind	-0.05	-0.10	0.11	0.13
unregularized	E quiet	A rude	0.13	0.03	-0.03	0.07
unregularized	E quiet	C lazy	-0.16	0.21	-0.03	0.27
unregularized	E quiet	C reliable	0.07	-0.34	0.15	0.00
unregularized	E quiet	E outgoing	-0.06	-0.13	0.10	0.07
unregularized	E quiet	E quiet	-0.03	0.10	0.10	-0.21
unregularized	E quiet	N depressed	-0.21	0.20	0.14	-0.08
unregularized	E quiet	N relaxed	0.01	-0.07	-0.06	-0.04
unregularized	E quiet	N worried	-0.07	0.29	0.01	-0.29
unregularized	N depressed	A kind	0.11	-0.19	-0.13	-0.12
unregularized	N depressed	A rude	0.26	0.48	0.08	-0.04
unregularized	N depressed	C lazy	0.09	-0.02	-0.09	-0.13
unregularized	N depressed	C reliable	-0.07	0.04	0.11	-0.09
unregularized	N depressed	E outgoing	0.05	-0.28	-0.04	0.17
unregularized	N depressed	E quiet	-0.19	0.00	-0.01	0.11
unregularized	N depressed	N depressed	0.28	0.03	-0.04	0.29
unregularized	N depressed	N relaxed	0.13	-0.23	-0.12	-0.05
unregularized	N depressed	N worried	0.11	0.03	0.08	0.25
unregularized	N relaxed	A kind	-0.03	-0.11	-0.11	-0.04
unregularized	N relaxed	A rude	-0.47	-0.13	0.07	0.03
unregularized	N relaxed	C lazy	-0.06	0.03	-0.07	-0.01
unregularized	N relaxed	C reliable	-0.01	0.04	0.17	-0.14
unregularized	N relaxed	E outgoing	0.35	-0.22	-0.05	0.00
unregularized	N relaxed	E quiet	-0.17	0.27	-0.09	0.18
unregularized	N relaxed	N depressed	0.08	0.00	-0.12	0.12
unregularized	N relaxed	N relaxed	-0.03	0.01	-0.21	0.03
unregularized	N relaxed	N worried	-0.07	-0.08	0.08	-0.15
unregularized	N worried	A kind	-0.20	-0.33	0.07	-0.09
unregularized	N worried	A rude	-0.13	-0.12	0.30	0.06
unregularized	N worried	C lazy	-0.29	0.03	-0.11	0.01
unregularized	N worried	C reliable	0.17	-0.02	0.06	0.02
unregularized	N worried	E outgoing	0.28	-0.19	-0.09	-0.01
unregularized	N worried	E quiet	-0.07	0.23	-0.30	0.17
unregularized	N worried	N depressed	-0.07	0.01	0.21	0.13
unregularized	N worried	N relaxed	-0.24	0.10	-0.47	-0.03
unregularized	N worried	N worried	0.10	-0.08	0.19	0.02

Contemporaneous

gVAR_fit %>% filter(SID %in% c("22652", "91339")) %>%
  unnest(contemp) %>% select(SID, wave, lambda_scale, Var1, Var2, weight) %>%
  unite(temp, SID, wave, sep = ".") %>%
  spread(key = temp, value = weight) %>%
  kable(., "html", booktabs = T, escape = F, digits = 2,
        col.names = c("Network Type", "Var1", "Var2", "W1", "W2", "W1", "W2"),
        align = c("l", "l", "l", rep("c", 4)),
        caption = "Contemporaneous Networks") %>%
  add_header_above(c(" " = 3, "S1" = 2,"S2" = 2)) %>%
  kable_styling(bootstrap_options = c("striped"),full_width = T) %>%
  scroll_box(width = "750px", height = "400px")

Contemporaneous Networks

			S1		S2
Network Type	Var1	Var2	W1	W2	W1	W2
.25	A_kind	A_rude	0.00	0.00	0.00	0.00
.25	A_kind	C_lazy	0.00	0.00	0.00	0.00
.25	A_kind	C_reliable	0.00	0.00	0.00	0.05
.25	A_kind	E_outgoing	0.00	0.00	0.00	0.00
.25	A_kind	E_quiet	-0.03	-0.07	0.01	-0.14
.25	A_kind	N_depressed	0.00	0.00	0.00	0.00
.25	A_kind	N_relaxed	0.00	0.00	0.00	0.00
.25	A_kind	N_worried	0.00	0.00	0.00	-0.14
.25	A_rude	C_lazy	0.00	0.00	0.00	0.00
.25	A_rude	C_reliable	0.00	0.00	0.00	0.18
.25	A_rude	E_outgoing	0.00	0.00	-0.04	0.00
.25	A_rude	E_quiet	0.00	0.00	0.00	0.00
.25	A_rude	N_depressed	0.00	0.00	0.00	0.00
.25	A_rude	N_relaxed	0.00	0.00	0.00	0.00
.25	A_rude	N_worried	0.00	0.00	0.00	0.00
.25	C_lazy	C_reliable	-0.59	-0.59	0.00	0.00
.25	C_lazy	E_outgoing	-0.11	0.00	-0.23	0.00
.25	C_lazy	E_quiet	0.06	0.00	0.00	0.00
.25	C_lazy	N_depressed	0.00	0.00	0.00	-0.24
.25	C_lazy	N_relaxed	0.23	0.00	0.49	0.35
.25	C_lazy	N_worried	0.00	0.00	0.00	-0.18
.25	C_reliable	E_outgoing	0.20	0.00	0.12	0.00
.25	C_reliable	E_quiet	0.00	0.00	0.32	0.00
.25	C_reliable	N_depressed	-0.35	0.00	-0.14	0.00
.25	C_reliable	N_relaxed	0.00	0.00	-0.01	0.00
.25	C_reliable	N_worried	0.00	0.00	0.00	0.00
.25	E_outgoing	E_quiet	-0.32	-0.52	-0.23	-0.12
.25	E_outgoing	N_depressed	0.00	0.00	-0.23	0.00
.25	E_outgoing	N_relaxed	0.03	0.00	0.00	0.00
.25	E_outgoing	N_worried	-0.10	0.00	-0.04	0.00
.25	E_quiet	N_depressed	0.00	0.00	0.20	0.00
.25	E_quiet	N_relaxed	0.00	0.00	0.01	0.00
.25	E_quiet	N_worried	0.00	0.00	0.12	0.00
.25	N_depressed	N_relaxed	0.00	0.00	-0.29	-0.08
.25	N_depressed	N_worried	0.19	0.08	0.06	0.01
.25	N_relaxed	N_worried	-0.22	-0.10	-0.42	-0.43
.5	A_kind	A_rude	0.00	0.00	0.00	0.00
.5	A_kind	C_lazy	0.00	0.00	0.00	0.00
.5	A_kind	C_reliable	0.00	0.00	0.00	0.02
.5	A_kind	E_outgoing	0.00	0.00	0.00	0.00
.5	A_kind	E_quiet	-0.10	-0.02	0.00	-0.13
.5	A_kind	N_depressed	0.00	0.00	0.00	0.00
.5	A_kind	N_relaxed	0.00	0.00	0.00	0.00
.5	A_kind	N_worried	-0.05	0.00	0.00	-0.12
.5	A_rude	C_lazy	0.00	0.00	0.00	0.00
.5	A_rude	C_reliable	0.00	0.00	0.00	0.05
.5	A_rude	E_outgoing	0.00	0.00	0.00	0.00
.5	A_rude	E_quiet	0.00	0.00	0.00	0.00
.5	A_rude	N_depressed	0.00	0.00	0.00	0.00
.5	A_rude	N_relaxed	0.00	0.00	0.00	0.00
.5	A_rude	N_worried	0.00	0.00	0.00	0.00
.5	C_lazy	C_reliable	-0.60	-0.54	0.00	0.00
.5	C_lazy	E_outgoing	-0.11	0.00	0.00	0.00
.5	C_lazy	E_quiet	0.06	0.00	0.00	0.00
.5	C_lazy	N_depressed	0.00	0.00	0.00	-0.04
.5	C_lazy	N_relaxed	0.29	0.00	0.37	0.24
.5	C_lazy	N_worried	0.00	0.00	0.00	-0.16
.5	C_reliable	E_outgoing	0.21	0.00	0.00	0.00
.5	C_reliable	E_quiet	0.00	0.00	0.13	0.00
.5	C_reliable	N_depressed	-0.37	0.00	0.00	0.00
.5	C_reliable	N_relaxed	0.00	0.00	0.00	0.00
.5	C_reliable	N_worried	0.00	0.00	0.00	0.00
.5	E_outgoing	E_quiet	-0.34	-0.49	-0.01	0.00
.5	E_outgoing	N_depressed	0.00	0.00	-0.03	0.00
.5	E_outgoing	N_relaxed	0.09	0.00	0.00	0.00
.5	E_outgoing	N_worried	-0.11	0.00	0.00	0.00
.5	E_quiet	N_depressed	-0.01	0.00	0.00	0.00
.5	E_quiet	N_relaxed	0.00	0.00	0.00	0.00
.5	E_quiet	N_worried	0.00	0.00	0.00	0.00
.5	N_depressed	N_relaxed	0.00	0.00	-0.31	-0.10
.5	N_depressed	N_worried	0.25	0.00	0.00	0.25
.5	N_relaxed	N_worried	-0.28	-0.03	-0.19	-0.34
association	A_kind	A_rude	-0.13	-0.24	0.12	0.19
association	A_kind	C_lazy	0.00	-0.04	0.04	0.13
association	A_kind	C_reliable	0.07	0.01	-0.13	0.38
association	A_kind	E_outgoing	0.11	0.46	-0.22	0.10
association	A_kind	E_quiet	-0.30	-0.42	0.06	-0.39
association	A_kind	N_depressed	-0.09	-0.25	0.38	-0.28
association	A_kind	N_relaxed	0.12	0.20	-0.30	0.21
association	A_kind	N_worried	-0.32	-0.12	0.24	-0.40
association	A_rude	C_lazy	0.08	-0.12	0.11	0.08
association	A_rude	C_reliable	-0.32	0.17	-0.20	0.52
association	A_rude	E_outgoing	-0.08	-0.18	-0.21	0.01
association	A_rude	E_quiet	0.05	0.06	-0.16	-0.06
association	A_rude	N_depressed	0.15	-0.09	0.38	-0.10
association	A_rude	N_relaxed	-0.13	-0.13	-0.29	0.29
association	A_rude	N_worried	0.14	0.30	0.19	-0.33
association	C_lazy	C_reliable	-0.79	-0.82	-0.18	-0.11
association	C_lazy	E_outgoing	-0.46	-0.19	-0.17	-0.03
association	C_lazy	E_quiet	0.29	0.10	-0.12	-0.09
association	C_lazy	N_depressed	0.39	-0.06	-0.28	-0.37
association	C_lazy	N_relaxed	0.42	0.23	0.62	0.52
association	C_lazy	N_worried	0.08	0.20	-0.37	-0.47
association	C_reliable	E_outgoing	0.54	0.16	0.16	0.02
association	C_reliable	E_quiet	-0.23	-0.12	0.49	-0.22
association	C_reliable	N_depressed	-0.63	0.01	0.00	-0.02
association	C_reliable	N_relaxed	-0.11	-0.11	-0.24	0.00
association	C_reliable	N_worried	-0.24	-0.20	-0.02	-0.02
association	E_outgoing	E_quiet	-0.49	-0.65	-0.33	-0.40
association	E_outgoing	N_depressed	-0.32	-0.21	-0.40	-0.17
association	E_outgoing	N_relaxed	0.11	0.20	0.05	-0.15
association	E_outgoing	N_worried	-0.31	-0.17	-0.29	0.22
association	E_quiet	N_depressed	0.03	0.01	0.25	0.21
association	E_quiet	N_relaxed	0.09	-0.02	-0.06	0.02
association	E_quiet	N_worried	0.08	0.08	0.27	0.20
association	N_depressed	N_relaxed	-0.09	-0.10	-0.67	-0.46
association	N_depressed	N_worried	0.42	0.40	0.49	0.53
association	N_relaxed	N_worried	-0.43	-0.40	-0.67	-0.67
gimme	A_kind	A_rude	0.00	0.00	0.05	-0.16
gimme	A_kind	C_lazy	0.00	0.00	0.00	0.00
gimme	A_kind	C_reliable	0.00	0.00	-0.20	0.24
gimme	A_kind	E_outgoing	0.00	0.00	0.00	0.00
gimme	A_kind	E_quiet	-0.18	-0.22	0.00	-0.22
gimme	A_kind	N_depressed	0.00	0.00	0.00	0.00
gimme	A_kind	N_relaxed	0.00	0.00	0.00	0.00
gimme	A_kind	N_worried	0.00	0.00	0.00	-0.25
gimme	A_rude	C_lazy	0.00	0.00	0.24	0.00
gimme	A_rude	C_reliable	0.00	0.00	-0.20	0.27
gimme	A_rude	E_outgoing	0.00	0.00	0.00	0.00
gimme	A_rude	E_quiet	0.00	0.00	0.38	0.00
gimme	A_rude	N_depressed	0.00	0.00	0.19	0.00
gimme	A_rude	N_relaxed	0.00	0.00	0.00	0.00
gimme	A_rude	N_worried	0.00	0.17	0.00	0.00
gimme	C_lazy	C_reliable	-0.44	-0.47	0.00	0.00
gimme	C_lazy	E_outgoing	0.00	0.00	0.00	0.00
gimme	C_lazy	E_quiet	0.00	0.00	0.00	0.00
gimme	C_lazy	N_depressed	0.00	0.00	0.00	-0.12
gimme	C_lazy	N_relaxed	0.24	0.36	0.27	0.24
gimme	C_lazy	N_worried	0.00	0.00	0.00	0.00
gimme	C_reliable	E_outgoing	0.27	0.00	0.00	0.00
gimme	C_reliable	E_quiet	0.00	0.00	0.51	0.00
gimme	C_reliable	N_depressed	-0.29	0.00	-0.34	0.00
gimme	C_reliable	N_relaxed	0.12	0.31	-0.41	0.00
gimme	C_reliable	N_worried	0.00	0.00	0.00	0.00
gimme	E_outgoing	E_quiet	-0.25	-0.32	-0.21	-0.27
gimme	E_outgoing	N_depressed	0.00	0.00	0.00	0.00
gimme	E_outgoing	N_relaxed	0.00	0.00	0.00	0.28
gimme	E_outgoing	N_worried	0.00	0.00	0.00	0.33
gimme	E_quiet	N_depressed	0.00	0.00	0.40	0.00
gimme	E_quiet	N_relaxed	0.00	0.00	0.00	0.00
gimme	E_quiet	N_worried	0.00	0.00	0.00	0.15
gimme	N_depressed	N_relaxed	0.00	0.00	-0.24	0.00
gimme	N_depressed	N_worried	0.18	0.23	-0.25	0.26
gimme	N_relaxed	N_worried	-0.23	-0.27	-0.31	-0.54
mlVAR_EB	A_kind	A_rude	-0.11	-0.18	-0.10	-0.05
mlVAR_EB	A_kind	C_lazy	-0.03	-0.03	0.00	0.01
mlVAR_EB	A_kind	C_reliable	0.13	0.10	0.10	0.21
mlVAR_EB	A_kind	E_outgoing	0.10	0.18	0.08	0.15
mlVAR_EB	A_kind	E_quiet	-0.14	-0.08	-0.08	-0.08
mlVAR_EB	A_kind	N_depressed	-0.04	-0.10	0.09	-0.08
mlVAR_EB	A_kind	N_relaxed	0.08	0.08	0.06	0.07
mlVAR_EB	A_kind	N_worried	-0.01	0.01	0.00	-0.09
mlVAR_EB	A_rude	C_lazy	0.03	0.05	0.06	0.05
mlVAR_EB	A_rude	C_reliable	-0.05	-0.02	-0.09	-0.02
mlVAR_EB	A_rude	E_outgoing	0.03	0.02	0.03	0.02
mlVAR_EB	A_rude	E_quiet	-0.09	-0.08	-0.09	-0.07
mlVAR_EB	A_rude	N_depressed	0.05	-0.04	0.17	0.05
mlVAR_EB	A_rude	N_relaxed	-0.04	-0.05	-0.08	-0.04
mlVAR_EB	A_rude	N_worried	0.03	0.05	0.03	-0.01
mlVAR_EB	C_lazy	C_reliable	-0.24	-0.52	-0.13	-0.21
mlVAR_EB	C_lazy	E_outgoing	-0.05	-0.03	-0.17	-0.04
mlVAR_EB	C_lazy	E_quiet	-0.06	0.00	-0.02	-0.01
mlVAR_EB	C_lazy	N_depressed	0.08	0.05	0.05	0.05
mlVAR_EB	C_lazy	N_relaxed	0.32	0.23	0.36	0.30
mlVAR_EB	C_lazy	N_worried	0.01	-0.01	-0.03	-0.05
mlVAR_EB	C_reliable	E_outgoing	0.05	0.04	0.09	0.05
mlVAR_EB	C_reliable	E_quiet	0.04	0.03	0.03	-0.06
mlVAR_EB	C_reliable	N_depressed	-0.28	-0.07	-0.12	-0.11
mlVAR_EB	C_reliable	N_relaxed	0.05	0.00	-0.02	0.03
mlVAR_EB	C_reliable	N_worried	-0.04	-0.05	-0.02	-0.06
mlVAR_EB	E_outgoing	E_quiet	-0.39	-0.59	-0.24	-0.35
mlVAR_EB	E_outgoing	N_depressed	-0.03	-0.09	-0.12	-0.09
mlVAR_EB	E_outgoing	N_relaxed	0.13	0.13	0.04	0.05
mlVAR_EB	E_outgoing	N_worried	-0.07	-0.04	-0.07	-0.01
mlVAR_EB	E_quiet	N_depressed	0.01	-0.01	0.03	0.03
mlVAR_EB	E_quiet	N_relaxed	-0.03	0.05	-0.01	0.03
mlVAR_EB	E_quiet	N_worried	0.00	0.02	0.03	0.03
mlVAR_EB	N_depressed	N_relaxed	-0.15	-0.08	-0.28	-0.19
mlVAR_EB	N_depressed	N_worried	0.17	0.31	0.17	0.20
mlVAR_EB	N_relaxed	N_worried	-0.37	-0.39	-0.38	-0.43
scaled	A_kind	A_rude	0.00	0.00	0.00	0.00
scaled	A_kind	C_lazy	0.00	0.00	0.00	0.00
scaled	A_kind	C_reliable	0.00	0.00	0.00	0.00
scaled	A_kind	E_outgoing	0.00	0.00	0.00	0.00
scaled	A_kind	E_quiet	0.00	0.00	0.00	-0.01
scaled	A_kind	N_depressed	0.00	0.00	0.00	0.00
scaled	A_kind	N_relaxed	0.00	0.00	0.00	0.00
scaled	A_kind	N_worried	0.00	0.00	0.00	-0.01
scaled	A_rude	C_lazy	0.00	0.00	0.00	0.00
scaled	A_rude	C_reliable	0.00	0.00	0.00	0.00
scaled	A_rude	E_outgoing	0.00	0.00	0.00	0.00
scaled	A_rude	E_quiet	0.00	0.00	0.00	0.00
scaled	A_rude	N_depressed	0.00	0.00	0.00	0.00
scaled	A_rude	N_relaxed	0.00	0.00	0.00	0.00
scaled	A_rude	N_worried	0.00	0.00	0.00	0.00
scaled	C_lazy	C_reliable	-0.48	-0.38	0.00	0.00
scaled	C_lazy	E_outgoing	-0.10	0.00	-0.09	0.00
scaled	C_lazy	E_quiet	0.00	0.00	0.00	0.00
scaled	C_lazy	N_depressed	0.00	0.00	0.00	-0.14
scaled	C_lazy	N_relaxed	0.00	0.00	0.42	0.27
scaled	C_lazy	N_worried	0.00	0.00	0.00	-0.18
scaled	C_reliable	E_outgoing	0.06	0.00	0.00	0.00
scaled	C_reliable	E_quiet	0.00	0.00	0.03	0.00
scaled	C_reliable	N_depressed	-0.11	0.00	0.00	0.00
scaled	C_reliable	N_relaxed	0.00	0.00	0.00	0.00
scaled	C_reliable	N_worried	0.00	0.00	0.00	0.00
scaled	E_outgoing	E_quiet	-0.08	-0.39	-0.15	0.00
scaled	E_outgoing	N_depressed	0.00	0.00	-0.07	0.00
scaled	E_outgoing	N_relaxed	0.00	0.00	0.00	0.00
scaled	E_outgoing	N_worried	0.00	0.00	0.00	0.00
scaled	E_quiet	N_depressed	0.00	0.00	0.11	0.00
scaled	E_quiet	N_relaxed	0.00	0.00	0.00	0.00
scaled	E_quiet	N_worried	0.00	0.00	0.05	0.00
scaled	N_depressed	N_relaxed	0.00	0.00	-0.13	-0.04
scaled	N_depressed	N_worried	0.00	0.00	0.00	0.02
scaled	N_relaxed	N_worried	0.00	0.00	-0.26	-0.34
unregularized	A_kind	A_rude	-0.22	-0.23	-0.80	-0.35
unregularized	A_kind	C_lazy	-0.03	-0.18	0.53	0.10
unregularized	A_kind	C_reliable	-0.03	-0.10	-0.77	0.49
unregularized	A_kind	E_outgoing	-0.07	-0.02	-0.53	-0.23
unregularized	A_kind	E_quiet	-0.34	-0.35	0.81	-0.19
unregularized	A_kind	N_depressed	-0.05	-0.28	-0.73	-0.17
unregularized	A_kind	N_relaxed	-0.05	0.39	-0.81	-0.16
unregularized	A_kind	N_worried	-0.32	0.30	-0.62	-0.28
unregularized	A_rude	C_lazy	-0.32	-0.21	0.61	0.05
unregularized	A_rude	C_reliable	-0.30	0.10	-0.64	0.75
unregularized	A_rude	E_outgoing	-0.14	-0.06	-0.42	0.10
unregularized	A_rude	E_quiet	-0.22	-0.20	0.72	0.22
unregularized	A_rude	N_depressed	-0.25	-0.48	-0.47	0.08
unregularized	A_rude	N_relaxed	0.03	0.30	-0.62	0.05
unregularized	A_rude	N_worried	-0.33	0.63	-0.45	-0.23
unregularized	C_lazy	C_reliable	-0.67	-0.74	0.41	-0.27
unregularized	C_lazy	E_outgoing	-0.28	-0.22	0.06	-0.05
unregularized	C_lazy	E_quiet	-0.10	-0.21	-0.59	0.26
unregularized	C_lazy	N_depressed	-0.25	-0.30	0.27	-0.39
unregularized	C_lazy	N_relaxed	0.56	0.47	0.54	0.30
unregularized	C_lazy	N_worried	0.01	0.33	0.39	0.00
unregularized	C_reliable	E_outgoing	0.18	-0.03	-0.33	-0.01
unregularized	C_reliable	E_quiet	0.07	-0.03	0.81	-0.11
unregularized	C_reliable	N_depressed	-0.58	0.01	-0.82	-0.11
unregularized	C_reliable	N_relaxed	0.18	0.24	-0.81	-0.03
unregularized	C_reliable	N_worried	0.05	-0.02	-0.67	0.12
unregularized	E_outgoing	E_quiet	-0.52	-0.60	0.28	-0.66
unregularized	E_outgoing	N_depressed	-0.03	-0.29	-0.55	-0.64
unregularized	E_outgoing	N_relaxed	0.37	0.20	-0.54	-0.13
unregularized	E_outgoing	N_worried	-0.24	0.11	-0.36	0.32
unregularized	E_quiet	N_depressed	0.00	-0.37	0.71	-0.28
unregularized	E_quiet	N_relaxed	0.24	0.25	0.77	-0.09
unregularized	E_quiet	N_worried	-0.20	0.27	0.74	0.34
unregularized	N_depressed	N_relaxed	-0.01	0.31	-0.87	-0.27
unregularized	N_depressed	N_worried	0.18	0.59	-0.68	0.10
unregularized	N_relaxed	N_worried	-0.28	-0.54	-0.79	-0.64

Centrality: Which Indicators Are Most Influential in Idiographic Personality Networks?

As with between-person effects, we can calculate centrality for individuals.

# create function to save both centrality measure and variable names to a data frame.
centrality_fun <- function(x, SID, wave, l) {
  data <- x %>%
    select(from, to, weight) %>%
    mutate(weight = as.numeric(weight))
  centrality <- centrality_auto(data.frame(data))
  df <- centrality$node.centrality %>% data.frame() %>%
    mutate(var = rownames(.))
  if(!"InStrength" %in% colnames(df)){
    df <- df %>% select(var, Betweenness, Closeness) %>% 
      bind_cols(data.frame(matrix(rep(0,9*4), nrow = 9, ncol = 4)) %>%
                  setNames(c("InStrength", "OutStrength", "OutExpectedInfluence", "InExpectedInfluence"))) 
  }
  return(df)
}

contemp_cen_fun <- function(x){
  centrality <- centrality_auto(x)$node.centrality %>% 
    data.frame() %>%
    mutate(var = rownames(.))
  return(centrality)
}

# 59057

####### PDC's #######
# calculate centrality for each subject for each wave and save them to a list #
gVAR_fit <- gVAR_fit %>%
  mutate(temp_centrality    = map(temp, centrality_fun),
         contemp_centrality = map(contemp_mat, contemp_cen_fun))

# save data for web app
gVAR_data <- gVAR_fit %>% 
  select(SID, wave, lambda_scale, gVAR_fit, temp_centrality, contemp_centrality) %>%
  mutate(PDC = map(gVAR_fit, ~.$PDC),
         PCC = map(gVAR_fit, ~.$PCC)) %>%
  select(-gVAR_fit)
# save(gVAR_data, temp_long, contemp_long, file = sprintf("%s/R/app_data.RData", res_path))

# save all lagged centrality to a long format df
(temp_centrality <- gVAR_fit %>%
  unnest(temp_centrality, .drop = T) %>%
  # select(-Degree) %>%
  mutate(type = "Lagged"))

# save all contemporaneous centrality to a long format df
(contemp_centrality <- gVAR_fit %>%
  unnest(contemp_centrality, .drop = T) %>%
  select(-Degree) %>%
  mutate(type = "Contemporaneous"))

Centrality Plots (Figure 6)

# function to create individual level centrality plots for each person
centrality_Plot_fun <- function(x, ls){
  centrality  %>%
    filter(SID %in% x &  grepl("trength", measure) & lambda_scale == ls) %>%
    arrange(measure, wave) %>%
    ggplot(aes(x = var, y = z, group = wave))+
      geom_line(aes(linetype = wave), color = "black", size = .3) + 
      geom_point(aes(shape = wave), size = 2) + 
      labs(x = NULL, y = "z-score", linetype = "Wave", shape = "Wave") +
      scale_y_continuous(limits = c(-3,3), breaks = seq(-3,3,1)) + 
      geom_hline(aes(yintercept = 0)) + 
      coord_flip() + 
      facet_grid(SID~type + measure) + 
      theme_classic()+ 
      theme(axis.text = element_text(face = "bold"),
            axis.title = element_text(face = "bold"),
            legend.title = element_text(face = "bold"),
            legend.position = "bottom")
}

centrality <- temp_centrality %>%
  gather(key = measure, value = centrality, 
         Betweenness:OutStrength) %>%
  group_by(SID, wave, lambda_scale, measure) %>%
  mutate(z = scale(centrality)) %>%
  ungroup() %>%
  full_join(contemp_centrality %>%
    gather(key = measure, value = centrality, 
           Betweenness:Strength) %>%
    group_by(SID, wave, lambda_scale, measure) %>%
    mutate(z = scale(centrality)) %>%
    ungroup())

# generate sample plot for subjects 1 & 2
centrality %>% 
    filter(SID %in% c("22652", "91339") &  grepl("trength", measure) & lambda_scale == ".25") %>%
    mutate(SID = recode(SID, `22652` = "1", `91339` = "2"),
           wave = as.character(wave)) %>%
    arrange(measure, wave) %>%
    ggplot(aes(x = var, y = z, group = wave))+
      geom_line(aes(linetype = wave), color = "black", size = .3) + 
      geom_point(aes(shape = wave), size = 2) + 
      labs(x = NULL, y = "z-score", linetype = "Wave", shape = "Wave") +
      scale_y_continuous(limits = c(-3,3), breaks = seq(-3,3,1)) + 
      geom_hline(aes(yintercept = 0)) + 
      coord_flip() + 
      facet_grid(SID~type + measure) + 
      theme_classic()+ 
      theme(axis.text = element_text(face = "bold"),
            axis.title = element_text(face = "bold"),
            legend.title = element_text(face = "bold"),
            legend.position = "bottom")

# ggsave(sprintf("%s/results/plots/centrality_plot_1+2.png", res_path), width = 5, height = 5)

Question 2: Is there longitudinal consistency in idiographic networks?

Edge Weight Consistency

Variable Centered: Rank-Order Consistency

cor_fun <- function(x){
    results <- cor(x$`1`, x$`2`, use = "pairwise", method = "spearman")
}

# short fun to convert contempaneous results to long format df
# different from previous in that we explicitly want to keep NAs
contemp_long_fun_na <- function(fit){
  PCC <- fit$PCC
  PCC <- PCC[,order(colnames(PCC))]
  PCC <- PCC[order(rownames(PCC)),]
  PCC[lower.tri(PCC, diag = T)] <- NA
  vars <- rownames(PCC)
  PCC.long <- tbl_df(PCC) %>%
    mutate(Var1 = vars,
           type = "Contemporaneous") %>%
    gather(key = Var2, value = weight, A_kind:N_worried) %>%
    unite(var, Var1, Var2, sep = ".", remove = F)
}

gVAR_fit <- gVAR_fit %>% 
  mutate(contemp_long_na = map(gVAR_fit, possibly(contemp_long_fun_na, NA_real_)))

# assign ranks to edges and caluclate rank-order correlations
contemp_cors_long <- contemp_long %>%
  filter(SID %in% w1w2_subs) %>%
  group_by(wave, lambda_scale, Var1, Var2) %>%
  mutate(rank = min_rank(desc(weight))) %>%
  ungroup() %>%
  select(SID, wave, lambda_scale, Var1, Var2, type, rank) %>%
  spread(key = wave, value = rank) %>%
  group_by(lambda_scale, Var1, Var2) %>%
  nest() %>%
  mutate(r = map(data, cor_fun))  %>%
  unnest(r, .drop = T) %>% 
  mutate(measure = "Rank-Order", Type = "Contemporaneous")

temp_cors_long <- temp_long %>%
  filter(SID %in% w1w2_subs) %>%
  group_by(wave, lambda_scale, from, to) %>%
  mutate(rank = dense_rank(desc(weight))) %>%
  ungroup() %>%
  select(SID, wave, lambda_scale, from, to , rank) %>%
  spread(key = wave, value = rank) %>%
  group_by(lambda_scale, from, to) %>%
  nest() %>%
  mutate(r = map(data, cor_fun)) %>%
  unnest(r, .drop = T) %>%
  mutate(measure = "Rank-Order", Type = "Lagged")

load(url(sprintf("%s/raw/master/results/fa_results.RData?raw=true", res_path)))
eigen_long <- fa_fit %>% 
  filter(!is.na(eigen)) %>% 
  unnest(eigen) %>%
  rename(Type = fa_type) %>%
  mutate(lambda_scale = "FA") 

eigen_cors_long <- eigen_long %>%
  mutate(Type = mapvalues(Type, c("dfa", "fa"), c("Lagged", "Contemporaneous"))) %>%
  group_by(SID, Type, var, lambda_scale) %>%
  mutate(n = n()) %>%
  filter(n == 2) %>%
  group_by(wave, Type, var) %>%
  mutate(rank = dense_rank(desc(eigen))) %>%
  select(-count, -nfact, -n, -eigen) %>%
  spread(wave, rank) %>%
  group_by(Type, var, lambda_scale) %>%
  summarize(r = cor(`1`, `2`, use = "pairwise"),
         measure = "Rank Order") %>%
  ungroup() %>%
  rename(Var1 = var)

Table

levs <- paste(rep(c("Contemporaneous", "Lagged"), each = 5), 
      rep(c("mean", "sd", "median", "min", "max"), times = 2), sep = ".")

temp_cors_long %>%
  rename(Var1 = from, Var2 = to) %>%
  full_join(contemp_cors_long) %>%
  full_join(eigen_cors_long) %>%
  group_by(Type, lambda_scale) %>%
  summarise_at(vars(r), funs(mean, sd, min, max, median), na.rm = T) %>%
  gather(key = est, value = value, mean:median) %>%
  unite(tmp, Type, est, sep = ".") %>%
  mutate(tmp = factor(tmp, levels = levs)) %>%
  spread(tmp, value) %>%
  kable(., booktabs = T, digits = 2, format = "html",
        col.names = c("sfs", rep(c("M", "SD", "Med", "Min", "Max"),2)),
        caption = "Descriptives of Contemporaneous Edge Weight Rank-Order Consistency") %>%
  kable_styling(bootstrap_options = c("striped","repeat_header"), full_width = F) %>%
  add_header_above(c(" " = 1, "Contemporaneous" = 5, "Lagged" = 5))

Descriptives of Contemporaneous Edge Weight Rank-Order Consistency

	Contemporaneous					Lagged
sfs	M	SD	Med	Min	Max	M	SD	Med	Min	Max
.25	0.09	0.10	0.06	-0.10	0.33	0.01	0.10	0.01	-0.23	0.31
.5	0.09	0.12	0.06	-0.14	0.37	-0.01	0.10	-0.01	-0.21	0.25
association	0.17	0.11	0.14	-0.06	0.44	-0.01	0.09	-0.01	-0.20	0.21
FA	NaN	NaN	NA	Inf	-Inf	0.16	0.52	0.06	-0.74	0.73
gimme	0.04	0.12	0.04	-0.20	0.33	-0.02	0.10	0.00	-0.36	0.23
mlVAR_EB	0.12	0.12	0.11	-0.11	0.44	0.03	0.09	0.03	-0.21	0.25
scaled	0.09	0.13	0.10	-0.21	0.40	0.01	0.10	0.00	-0.25	0.31
unregularized	0.03	0.11	0.05	-0.14	0.32	0.01	0.11	0.00	-0.22	0.25

Plot

And create heat maps showing the individual lagged and contemporaneous rank-order correlations (Figure 7)

# make heat maps of both contemporaneous and lagged rank order cors
## figure 6 in manuscript ##
contemp_cors_long %>%
  full_join(temp_cors_long %>%
  rename(Var1 = from, Var2 = to)) %>%
  mutate(Var1 = factor(Var1, levels = sort(unique(Var1))),
         Var2 = factor(Var2, levels = rev(sort(unique(Var2))))) %>%
  filter(lambda_scale == ".25") %>%
  ggplot(aes(x = Var1, y = Var2, fill = r)) + 
  geom_raster() + 
  scale_fill_gradient2(low = "blue", high = "red", mid = "white", 
   midpoint = 0, limit = c(-.5,.5), space = "Lab", 
   name="Edge Consistency\nCorrelations") +  
  geom_text(aes(label = round(r,2))) +
  facet_grid(.~Type) + 
  labs(x = "Node 1", y = "Node 2") + 
  theme_classic() +
  theme(strip.text = element_text(face = "bold"),
        axis.text = element_text(face = "bold"),
        axis.title = element_text(face = "bold"),
        axis.text.x = element_text(angle = 45, vjust = 1, hjust = 1),
        legend.title = element_text(face = "bold"),
        legend.text = element_text(face = "bold"),
        legend.position = "bottom")

# ggsave(file = sprintf("%s/results/plots/Rank_Edge_Cors_heatmap.png", res_path), width = 9, height = 6)

Person-Centered: Ipsative Consistency

Contemporaneous

(ip_contemp_cors <- contemp_long %>%
  select(SID, lambda_scale, wave, Var1:weight) %>%
  group_by(SID, wave, lambda_scale) %>%
  mutate(weight.c = as.numeric(scale(weight, scale = F))) %>%
  ungroup() %>%
  arrange(lambda_scale, wave, SID, Var1, Var2) %>%
  select(-weight) %>%
  spread(wave, weight.c) %>%
  group_by(SID, lambda_scale) %>%
  summarize(cors = cor(`1`, `2`, use = "pairwise.complete.obs"),
            spear = cor(`1`, `2`, use = "pairwise.complete.obs", method = "spearman")) %>%
  filter(!is.na(cors)) %>%
  mutate(type = "Contemporaneous"))

Lagged

(ip_temp_cors <- temp_long %>%
  select(SID, lambda_scale, wave, from:weight) %>%
  group_by(SID, lambda_scale, wave, lambda_scale) %>%
  mutate(weight.c = as.numeric(scale(weight, center = T, scale = F))) %>%
  ungroup() %>%
  select(-weight) %>%
  spread(wave, weight.c) %>%
  group_by(SID, lambda_scale) %>%
  summarize(cors = cor(`1`, `2`, use = "pairwise.complete.obs"),
            spear = cor(`1`, `2`, use = "pairwise.complete.obs", method = "spearman")) %>%
  filter(!is.na(cors)) %>%
  mutate(type = "Lagged"))

Centrality Consistency

Variable Centered: Rank-Order Consistency

Contemporaneous

# convert centrality to ranks for each measure and 
# calculate rank-order correlations
contemp_centrality_rank <- tbl_df(contemp_centrality) %>%
  filter(SID %in% w1w2_subs) %>%
  gather(key = measure, value = Centrality, Betweenness:Strength) %>%
  group_by(measure, lambda_scale, var, type, wave) %>%
  mutate(rank = min_rank(desc(Centrality))) %>%
  ungroup() %>%
  select(-Centrality, -ExpectedInfluence) %>%
  spread(key = wave, value = rank)  %>%
  group_by(measure, lambda_scale, var) %>%
  summarize(r = cor(`1`, `2`, use = "pairwise", method = "spearman"))

contemp_centrality_rank %>%
  group_by(measure, lambda_scale) %>%
  summarize(m = meanSD_r2z2r(r)[1], 
            sd = meanSD_r2z2r(r)[2], 
            min = min(r), 
            max = max(r),
            Type = "Contemporaneous") %>%
  kable(., booktabs = T, digits = 2, format = "html",
        caption = "Descriptives of Contemporaneous Centrality Rank-Order Consistency") %>%
  kable_styling(bootstrap_options = c("striped","repeat_header"), full_width = T)

Descriptives of Contemporaneous Centrality Rank-Order Consistency

measure	lambda_scale	m	sd	min	max	Type
Betweenness	.25	0.08	0.08	-0.03	0.24	Contemporaneous
Betweenness	.5	0.12	0.11	0.00	0.29	Contemporaneous
Betweenness	association	0.05	0.08	-0.11	0.14	Contemporaneous
Betweenness	gimme	0.05	0.06	-0.06	0.13	Contemporaneous
Betweenness	mlVAR_EB	0.06	0.06	-0.02	0.15	Contemporaneous
Betweenness	scaled	0.10	0.07	-0.01	0.19	Contemporaneous
Betweenness	unregularized	-0.01	0.12	-0.20	0.18	Contemporaneous
Closeness	.25	0.11	0.12	0.01	0.38	Contemporaneous
Closeness	.5	-0.02	0.12	-0.16	0.25	Contemporaneous
Closeness	association	0.18	0.08	0.02	0.29	Contemporaneous
Closeness	gimme	-0.05	0.10	-0.21	0.09	Contemporaneous
Closeness	mlVAR_EB	0.09	0.05	-0.01	0.16	Contemporaneous
Closeness	scaled	0.09	0.11	-0.06	0.28	Contemporaneous
Closeness	unregularized	0.19	0.06	0.13	0.28	Contemporaneous
Strength	.25	0.20	0.08	0.11	0.33	Contemporaneous
Strength	.5	0.19	0.06	0.07	0.24	Contemporaneous
Strength	association	0.20	0.10	0.02	0.35	Contemporaneous
Strength	gimme	0.09	0.09	-0.07	0.28	Contemporaneous
Strength	mlVAR_EB	0.21	0.10	0.05	0.40	Contemporaneous
Strength	scaled	0.26	0.05	0.18	0.35	Contemporaneous
Strength	unregularized	0.16	0.07	0.05	0.25	Contemporaneous

Lagged

# convert centrality to ranks for each measure and 
# calculate rank-order correlations
temp_centrality_rank <- tbl_df(temp_centrality) %>%
  filter(SID %in% w1w2_subs) %>%
  gather(key = measure, value = Centrality, Betweenness:OutStrength) %>%
  group_by(measure, lambda_scale, var, type, wave) %>%
  mutate(rank = min_rank(desc(Centrality))) %>%
  ungroup() %>%
  gather(key = measure2, value = value, Centrality, rank) %>%
  unite(measure3, measure, measure2, remove = F, sep = ".")  %>%
  select(-contains("Influence")) %>%
  spread(key = wave, value = value) %>%
  filter(measure2 == "rank" & grepl("trength", measure)) %>%
  group_by(var, lambda_scale, type, measure) %>%
  summarize(r = cor(`1`, `2`, use = "pairwise"))

temp_centrality_rank %>% 
  group_by(measure, lambda_scale) %>%
  summarize(mean = meanSD_r2z2r(r)[1],
            sd = meanSD_r2z2r(r)[2], 
            min = min(r, na.rm = T),
            max = max(r, na.rm = T),
            Type = "Lagged") %>%
  kable(., booktabs = T, digits = 2, format = "html",
        caption = "Descriptives of Lagged Centrality Rank-Order Consistency") %>%
  kable_styling(bootstrap_options = c("striped","repeat_header"), full_width = T)

Descriptives of Lagged Centrality Rank-Order Consistency

measure	lambda_scale	mean	sd	min	max	Type
InStrength	.25	0.20	0.06	0.12	0.25	Lagged
InStrength	.5	0.14	0.04	0.09	0.20	Lagged
InStrength	association	0.13	0.07	-0.01	0.20	Lagged
InStrength	gimme	0.08	0.07	-0.02	0.17	Lagged
InStrength	mlVAR_EB	0.04	0.10	-0.16	0.22	Lagged
InStrength	scaled	0.20	0.06	0.11	0.30	Lagged
InStrength	unregularized	-0.02	0.07	-0.14	0.09	Lagged
OutStrength	.25	0.21	0.07	0.14	0.32	Lagged
OutStrength	.5	0.15	0.06	0.08	0.24	Lagged
OutStrength	association	0.13	0.07	-0.03	0.21	Lagged
OutStrength	gimme	0.07	0.09	-0.06	0.19	Lagged
OutStrength	mlVAR_EB	0.15	0.09	0.01	0.27	Lagged
OutStrength	scaled	0.21	0.07	0.09	0.31	Lagged
OutStrength	unregularized	-0.09	0.06	-0.19	0.01	Lagged

Person-Centered: Profile Consistency

Contemporaneous

# ipsative consistency (profile correlations for each person)
(ip_contemp_cent_cors <- contemp_centrality %>%
  filter(SID %in% w1w2_subs) %>%
  # select(-ExpectedInfluence) %>%
  gather(key = measure, value = Centrality, Betweenness:Strength) %>%
  spread(key = wave, value = Centrality) %>%
  arrange(SID, type, measure, var) %>%
  filter(complete.cases(.)) %>%
  group_by(SID, lambda_scale, type, measure) %>%
  nest() %>%
  mutate(r = map(data, ~cor(.$`1`, .$`2`, use = "pairwise"))) %>%
  unnest(r, .drop = T))

Lagged

# ipsative consistency (profile correlations for each person)
(ip_temp_cent_cors <- tbl_df(temp_centrality) %>%
  select(-contains("ExpectedInfluence")) %>%
  filter(SID %in% w1w2_subs) %>%
  gather(key = measure, value = Centrality, Betweenness:OutStrength) %>%
  spread(key = wave, value = Centrality) %>%
  arrange(lambda_scale, SID, type, measure, var) %>%
  group_by(SID, lambda_scale, type, measure) %>%
  nest() %>%
  mutate(r = map(data, ~cor(.$`1`, .$`2`, use = "pairwise"))) %>%
  unnest(r, .drop = T))

Validity Check: What Are the Methodological Correlates of Consistency?

What is the relationship between ipsative consistency and the number of observations?

Do people who have more observations show more consistency across waves?
The answer to this appears to be yes, but it appears to be true regardless of the choice of how to model the data and mostly true for contemporaneous associations rather than lagged associations, with the exception of lagged (zero-order) association networks.

ip_temp_cors %>% full_join(ip_contemp_cors) %>%
  full_join(w1 %>% full_join(w2) %>% select(SID, count, wave) %>% distinct()) %>%
  filter(!is.na(type)) %>%
  group_by(wave, lambda_scale, type) %>%
  summarize(r = cor(cors, count, use = "pairwise")) %>%
  spread(key = type, value = r) %>%
  kable(., "html", booktabs = T, escape = F, digits = 2,
        caption = "Sample Size Correlations") %>%
  kable_styling(full_width = F)

Sample Size Correlations

wave	lambda_scale	Contemporaneous	Lagged
1	.25	0.46	0.20
1	.5	0.41	0.06
1	association	0.34	0.29
1	gimme	0.48	0.41
1	mlVAR_EB	-0.11	0.10
1	scaled	0.42	0.10
1	unregularized	0.49	0.24
2	.25	0.54	0.14
2	.5	0.52	0.08
2	association	0.41	0.42
2	gimme	0.53	0.43
2	mlVAR_EB	-0.11	0.11
2	scaled	0.41	0.21
2	unregularized	0.49	0.12

What is the realtionship between ipsative consistency and the imbalance in number of observations across waves?

Do people who had more observations in one wave than the other show more (or less) consistency than those who had roughly equal numbers?
There appear to be small correlations between imbalance and consistency for contemporaneous, but not lagged, associations, such that the larger the difference between waves, the higher the consistency estimates, which is not the direction I would have predicted if there were to be issues in the estimation.

ip_temp_cors %>% full_join(ip_contemp_cors) %>%
  full_join(w1 %>% full_join(w2) %>% select(SID, count, wave) %>% distinct()) %>%
  filter(!is.na(type)) %>%
  group_by(SID, lambda_scale, cors, type) %>%
  summarise(abs_diff = abs(count[wave == 1] - count[wave == 2])) %>%
  group_by(lambda_scale, type) %>%
  summarize(r = cor(cors, abs_diff, use = "pairwise")) %>%
  spread(key = type, value = r) %>%
  kable(., "html", booktabs = T, escape = F, digits = 2,
        caption = "Imbalance Correlations") %>%
  kable_styling(full_width = F)

Imbalance Correlations

lambda_scale	Contemporaneous	Lagged
.25	0.19	0.07
.5	0.17	0.04
association	0.21	0.19
gimme	0.32	0.23
mlVAR_EB	-0.06	0.04
scaled	0.06	0.07
unregularized	0.35	0.09

Disentangling Consistency from Reliability: Split-Half Networks

Person-Centered: Split-Half Profile Consistency

To test the reliability of the networks, we split each person’s responses in half and calculate a network of eah and then compare the two using profile correlations.

##Split Half Networks (Reliability Check)
# To test the reliability of the networks, we split each person's responses in half and
# calculate a network of eah and then compare the two using profile correlations.  

gVAR_split_fun <- function(sid, wave, spl_type, spl_num, row_num){
  df <- switch(as.numeric(wave), w1, w2) %>%
    filter(SID == sid)
  if(spl_type == "half"){
    # first havlf v. second half
    df <- df %>% 
      mutate(split = ifelse(count %% 2 == 0, count/2, count%/%2),
             split = ifelse(beep_seq <= split, 1, 2)) %>%
      filter(split == spl_num)
  } else {
    # odds v. evens
    df <- df %>%
      mutate(split = ifelse(beep_seq %in% seq(1,unique(count),2), 1, 2)) %>%
      filter(split == spl_num)
  }
  df <- df %>%
    group_by(SID, count, wave) %>% 
    nest() %>%
    mutate(data2 = map(data, jitter_fun)) %>%
    unnest(data2, .drop = T)
  print(row_num)
  gVAR_fun(df, sid, wave)
}

gVAR_fit_split <- unique(expand.grid(stringsAsFactors = F,
  SID = unique(w1$SID), wave = "1", 
  split_type = c("half", "every-other"), split_num = c(1,2)
  ) %>%
  full_join(expand.grid(stringsAsFactors = F,
    SID = unique(w2$SID), wave = "2",
    split_type = c("half", "every-other"), split_num = c(1,2)
  )) %>%
  full_join(unique(unique(w1 %>% select(SID, count, wave)) %>%
    full_join(unique(w2 %>% select(SID, count, wave))))) %>%
  filter(count >= 20) %>%
  tbl_df() %>%
  arrange(wave, SID)) %>%
  filter(!(wave == "1" & SID %in% c("88595", "7287", "30871", "42807", "53655",
            "70848", "6009", "29473", "75689", "28212", "25083", "34967"))) %>%
  filter(!(wave == "2" & SID %in% c("13674", "47525", "20054", "57290"))) %>%
  mutate(row = row_number(),
         gVAR_fit = pmap(list(SID, wave, split_type, split_num, row), 
                         possibly(gVAR_split_fun, NA_real_)))
save(gVAR_fit_split, file = sprintf("%s/results/split_half_gVAR_fit.RData", res_path))

# run the same extraction procedure as above for split-half networks
##split half networks
load(url(sprintf("%s/raw/master/results/split_half_gVAR_fit.RData?raw=true", res_path)))
gVAR_fit_split <- gVAR_fit_split %>%
  filter(!is.na(gVAR_fit)) %>%
  group_by(SID, wave, split_type) %>%
  mutate(n = n()) %>%
  # we can only assess individuals for whom we could estimate a network for each
  # half of their data within a wave
  filter(n == 2) %>%
  mutate(beta = map2(gVAR_fit, SID, possibly(temp_fun, NA_real_)),
         kappa_mat = map(gVAR_fit, possibly(contemp_mat_fun, NA_real_)),
         kappa = map(gVAR_fit, possibly(contemp_long_fun, NA_real_)))

split_beta_long <- gVAR_fit_split %>% filter(!is.na(beta)) %>% unnest(beta)
split_kappa_long <- gVAR_fit_split %>% filter(!is.na(kappa)) %>% unnest(kappa)

# extract PCC information
split_PCC_fit <- gVAR_fit_split$kappa_mat; names(split_PCC_fit) <- gVAR_fit_split$SID

# pull cross-wave lagged and contemporaneous effects
# and calculate correlations across waves
ip_split_cors <- split_beta_long %>% 
  ungroup() %>%
  mutate(type = "Lagged") %>%
  full_join(split_kappa_long %>%  ungroup() %>% select(-var) %>%
              mutate(type = "Contemporaneous") %>% 
              rename(from = Var1, to = Var2)) %>%
  group_by(SID, type, wave, split_type, split_num) %>%
  mutate(value.c = as.numeric(scale(weight, center = T, scale = F))) %>%
  ungroup() %>%
  arrange(wave, type, split_type, split_num, SID, from, to) %>%
  select(-weight, -n, -count, -row) %>%
  # unite(split, wave, split_num) %>%
  spread(split_num, value.c) %>%
  group_by(SID, wave, type, split_type) %>%
  summarize(r = cor(`1`, `2`, use = "pairwise.complete.obs"))

# descriptives
ip_split_cors %>%
  group_by(wave, type, split_type) %>%
  summarise(M = mean(r, na.rm = T),
            sd = sd(r, na.rm = T),
            min = min(r, na.rm = T),
            max = max(r, na.rm = T)) %>%
  kable(., booktabs = T, digits = 2, format = "html",
        caption = "Descriptives of Network Reliability") %>%
  kable_styling(bootstrap_options = c("striped","repeat_header"), full_width = T)

Descriptives of Network Reliability

wave	type	split_type	M	sd	min	max
1	Contemporaneous	every-other	0.43	0.29	-0.19	0.99
1	Contemporaneous	half	0.40	0.28	-0.32	0.92
1	Lagged	every-other	0.01	0.15	-0.45	0.55
1	Lagged	half	0.01	0.14	-0.61	0.53
2	Contemporaneous	every-other	0.50	0.31	-0.19	0.98
2	Contemporaneous	half	0.49	0.34	-0.34	0.96
2	Lagged	every-other	0.00	0.17	-0.47	0.60
2	Lagged	half	0.03	0.16	-0.59	0.58

Split-Half and Every-Other Reliability (Figures 4 & 5)

plot_fun <- function(df, type){
  plot <- df %>%
    mutate(split_type = dplyr::recode(split_type, `every-other` = "Odd-Even",
              `half` = "Split-Half")) %>%
  ggplot(aes(x = r, y = ..density.., fill = split_type)) +
  geom_histogram(color = "black", fill = "gray") +
  geom_density(alpha = .2) +
  scale_x_continuous(limits = c(-1,1), breaks = seq(-1,1,.5)) +
  labs(x = sprintf("%s Ipsative Correlations", type), y = "Density") +
  facet_grid(wave~split_type) +
  theme_classic() +
  theme(axis.text = element_text(face = "bold", size = rel(1.2)),
        axis.title = element_text(face = "bold", size = rel(1.2)),
        strip.text = element_text(face = "bold", size = rel(1.2)),
        plot.title = element_text(face = "bold", size = rel(1.2), hjust = .5),
        legend.position = "none")
  # ggsave(sprintf("%s/results/plots/%s_split_reliability.png", res_path, type), 
  #        width = 6, height = 4)
  print(plot)
  return(plot)
}

# plot histograms of ipsative lagged and contemporaneous consistency
ip_split_plot <- ip_split_cors %>%
  ungroup() %>%
  mutate(wave = sprintf("Wave %s", wave)) %>% 
  group_by(type) %>%
  nest() %>%
  mutate(plot = map2(data, type, plot_fun))

Question 3: Are there Individual Differences in Idiographic Network Consistency?

Congruence with the Reference Profile

load(url(sprintf("%s/blob/master/results/PAIRS_networks_pers_FINAL.RData?raw=true", res_path)))
profile_cor_fun <- function(id_list, wave, type){
  if(type == "Lagged"){
    id_list <- id_list %>% select(from, to, type, weight) %>% arrange(from, to)
    if(wave == "1"){pop_list <- lagged_effects_w1} else{pop_list <- lagged_effects_w2}
    pop_list <- pop_list %>% select(from, to, fixed) %>% arrange(from, to)
    id_list %>% full_join(pop_list) %>% summarize(r = cor(fixed, weight, use = "pairwise"))}
  else{
    id_list <- id_list %>% select(Var1, Var2, weight, type) %>% arrange(Var1, Var2)
    if(wave == "1"){pop_list <- contemp_eff_w1} else{pop_list <- contemp_eff_w2}
    pop_list <- pop_list %>% select(Var1, Var2, weight, type) %>% arrange(Var1, Var1) %>% rename(pcor = weight)
    id_list %>% full_join(pop_list) %>% group_by(type) %>% summarize(r = cor(pcor, weight, use = "pairwise"))}}

gVAR_fit <- gVAR_fit %>%
  mutate(temp_procor = map2(temp, wave, ~profile_cor_fun(.x, .y, "Lagged")),
         contemp_procor = map2(contemp, wave, ~profile_cor_fun(.x, .y, "Contemporaneous")))

(refCong_procors <- gVAR_fit %>% unnest(temp_procor, .drop = T) %>%
  mutate(type = "Lagged") %>%
  full_join(unnest(gVAR_fit, contemp_procor, .drop = T) %>%
              mutate(type = "Contemporaneous")))

Individual Differences in Network Edge Consistency (Figure 8)

To look at individual differences in Network consistency, we user a Fisher r to z transformation to compute the average and standard deviation. We also plot a histogram of individual ipsative profiles to show the full distribution of consistency.
### Plot

load(url(sprintf("%s/blob/master/results/fa_results.RData?raw=true", res_path)))

ip_cors <- ip_temp_cors %>% full_join(ip_contemp_cors) %>% 
  full_join(fa_cong %>% select(SID, fa_type, ip_eigen) %>% 
              mutate(fa_type = mapvalues(fa_type, c("dfa", "fa"), 
                        c("Lagged", "Contemporaneous"))) %>%
              rename(type = fa_type, cors = ip_eigen) %>%
              mutate(lambda_scale = "FA"))

ip_cors %>% filter(lambda_scale == ".25") %>%
  ggplot(aes(x = cors, y = ..density.., fill = type)) + 
    scale_fill_manual(values = c("blue", "springgreen4")) +
    geom_histogram(color = "black", fill = "gray") +
    geom_density(bw = .1, alpha = .3)+
    scale_x_continuous(limits = c(-1.1,1.1), breaks = seq(-1,1,1)) +
    labs(x = "Ipsative Correlations", y = "Density",
         title = "Ipsative Network Consistency") +
    facet_grid(~type) +
    theme_classic() +
    theme(axis.text = element_text(face = "bold", size = rel(1.2)),
          axis.title = element_text(face = "bold", size = rel(1.2)),
          strip.text = element_text(face = "bold", size = rel(1.2)),
          plot.title = element_text(face = "bold", size = rel(1.2), hjust = .5),
          legend.position = "none")

# ggsave(sprintf("%s/results/plots/ipsative_cor.png", res_path), 
       # width  = 6, height = 3)

new_names <- tribble(
  ~lambda_scale, ~new,
  ".25", "lambda==.025-.25~(Main)",
  ".5", "lambda==.025-.5~(Stricter)",
  "scaled", "lambda==.01%.%N~(Scaled)",
  "unregularized", "lambda==~0~(Unregularized)",
  "association", "Bivariate~Associations",
  "gimme", "GIMME",
  "mlVAR_EB", "Empirical~Bayes~Estimates",
  "FA", "P-Factor~Analysis"
) 

new <- new_names$new


p1 <-
  ip_cors %>% filter(type == "Lagged") %>%
  mutate(lambda_scale = factor(lambda_scale, labels = new, 
            levels = new_names$lambda_scale)) %>%
  ggplot(aes(x = cors, y = ..density.., fill = type)) + 
    scale_fill_manual(values = c( "springgreen4")) +
    geom_histogram(color = "black", fill = "gray") +
    geom_density(bw = .1, alpha = .3)+
    geom_vline(aes(xintercept = 0), linetype = "dashed") +
    scale_x_continuous(limits = c(-1.1,1.1), breaks = seq(-1,1,1)) +
    labs(x = "", y = NULL,
         title = "Lagged") +
    facet_wrap(~lambda_scale, nrow = 4, labeller = label_parsed) +
    theme_classic() +
    theme(axis.text = element_text(face = "bold", size = rel(1.2)),
          axis.text.y = element_blank(),
          axis.title = element_text(face = "bold", size = rel(1.2)),
          axis.title.x = element_text(hjust = 0),
          axis.ticks.y = element_blank(),
          axis.line.y = element_blank(),
          strip.text = element_text(face = "bold", size = rel(.8)),
          plot.title = element_text(face = "bold", size = rel(1.2), hjust = .5),
          legend.position = "none",
          strip.background = element_blank(),
          panel.border = element_rect(fill = NA, color = "black", size = 1))

p2 <-
  ip_cors %>% filter(type != "Lagged") %>%
  mutate(lambda_scale = factor(lambda_scale, labels = new, 
            levels = new_names$lambda_scale)) %>%
  ggplot(aes(x = cors, y = ..density.., fill = type)) + 
    scale_fill_manual(values = c("blue", "springgreen4")) +
    geom_histogram(color = "black", fill = "gray") +
    geom_density(bw = .1, alpha = .3)+
    geom_vline(aes(xintercept = 0), linetype = "dashed") +
    scale_x_continuous(limits = c(-1.1,1.1), breaks = seq(-1,1,1)) +
    labs(x = "", y = "Density",
         title = "Contemporaneous") +
    facet_wrap(~lambda_scale, nrow = 4, labeller = label_parsed) +
    theme_classic() +
    theme(axis.text = element_text(face = "bold", size = rel(1.2)),
          axis.title = element_text(face = "bold", size = rel(1.2)),
          axis.title.x = element_text(hjust = 1),
          axis.ticks.y = element_blank(),
          axis.line.y = element_blank(),
          strip.text = element_text(face = "bold", size = rel(.8)),
          plot.title = element_text(face = "bold", size = rel(1.2), hjust = .5),
          legend.position = "none",
          strip.background = element_blank(),
          panel.border = element_rect(fill = NA, color = "black", size = 1))

library(cowplot)
p <- cowplot::plot_grid(p2, p1, nrow = 1, rel_widths = c(53/100, 47/100))
p <- cowplot::ggdraw(add_sub(p, "Ipsative Consistency (Correlations)", vpadding=grid::unit(0,"lines"),y=6, hjust = .4, vjust=4.5, fontface = "bold"))
# ggsave(p, file = sprintf("%s/results/plots/ipsative_cor_all.png", data_path), width = 8, height = 8)

Table

ip_cors %>%
  group_by(type, lambda_scale) %>%
  summarize(mean_cor = meanSD_r2z2r(cors)[1],
            sd_cor = meanSD_r2z2r(cors)[2], 
            range = diff(range(cors, na.rm = T)),
            median = median(cors, na.rm = T)) %>%
  kable(., "html", escape = F, booktabs = T, digits = 2,
        col.names = c("Type", "Network Type", "Mean", "SD", "Range", "Median"))  %>%
  kable_styling(bootstrap_options = c("striped","repeat_header"), full_width = T)

Type	Network Type	Mean	SD	Range	Median
Contemporaneous	.25	0.64	0.41	1.12	0.65
Contemporaneous	.5	0.66	0.44	1.07	0.68
Contemporaneous	association	0.61	0.28	0.85	0.61
Contemporaneous	FA	0.97	0.37	0.15	0.97
Contemporaneous	gimme	0.41	0.31	1.20	0.43
Contemporaneous	mlVAR_EB	0.92	0.25	0.21	0.92
Contemporaneous	scaled	0.74	0.61	1.04	0.72
Contemporaneous	unregularized	0.38	0.25	0.89	0.35
Lagged	.25	0.03	0.16	0.89	0.01
Lagged	.5	0.01	0.16	1.05	0.00
Lagged	association	0.07	0.20	1.22	0.04
Lagged	FA	0.98	0.52	0.20	0.97
Lagged	gimme	0.07	0.20	0.85	0.04
Lagged	mlVAR_EB	0.43	0.19	0.71	0.44
Lagged	scaled	0.03	0.17	1.03	0.01
Lagged	unregularized	0.03	0.14	0.67	0.04

Individual Differences in Centrality Consistency (Figure 9)

Plot

As with ipsative edge weight consistency, we can look at the average consistency of profiles of centrality measures as well as the distribution of ipsative consistency across the population.

ip_cent_cors <- ip_temp_cent_cors %>% 
  full_join(ip_contemp_cent_cors)

ip_temp_cent_cors %>%
  full_join(ip_contemp_cent_cors ) %>%
  filter(grepl("trength", measure) & lambda_scale == ".25") %>%
  ggplot(aes(x = r, y = ..density.., fill = measure)) + 
  geom_histogram(binwidth = .1, color = "black", fill = "gray")+
  geom_density(bw = .1, alpha = .3)+
  labs(y = "Density", x = "Profile Correlation") + 
  facet_wrap(~type + measure, nrow = 1) + 
  theme_classic() + 
  theme(plot.title = element_text(hjust = .5),
        legend.position = "none",
        axis.text = element_text(face = "bold", size = rel(1.2)),
        axis.title = element_text(face = "bold", size = rel(1.2)),
        strip.text = element_text(face = "bold", size = rel(1.2)))

# ggsave(sprintf("%s/results/plots/centrality_ipsative_cor.png", res_path), 
#        width  = 6, height = 3)

Table

ip_temp_cent_cors %>%
  full_join(ip_contemp_cent_cors) %>%
  group_by(type, lambda_scale, measure) %>%
  summarize(mean = meanSD_r2z2r(r)[1],
            sd = meanSD_r2z2r(r)[2], 
            min = min(r,na.rm = T),
            max = max(r,na.rm = T),
            range = diff(range(r, na.rm = T))) %>%
  kable(., "html", booktabs = T, escape = F, digits = 2,
        caption = "Descriptives of Ipsative Centrality Correlations") %>%
  kable_styling(bootstrap_options = c("striped","repeat_header"), full_width = T)

Descriptives of Ipsative Centrality Correlations

type	lambda_scale	measure	mean	sd	min	max	range
Contemporaneous	.25	Betweenness	NaN	NA	Inf	-Inf	-Inf
Contemporaneous	.25	Strength	NaN	NA	Inf	-Inf	-Inf
Contemporaneous	.5	Betweenness	NaN	NA	Inf	-Inf	-Inf
Contemporaneous	.5	Strength	NaN	NA	Inf	-Inf	-Inf
Contemporaneous	gimme	Betweenness	NaN	NA	Inf	-Inf	-Inf
Contemporaneous	gimme	Strength	NaN	NA	Inf	-Inf	-Inf
Contemporaneous	scaled	Betweenness	NaN	NA	Inf	-Inf	-Inf
Contemporaneous	scaled	Closeness	NaN	NA	Inf	-Inf	-Inf
Contemporaneous	scaled	Strength	NaN	NA	Inf	-Inf	-Inf
Lagged	.25	Betweenness	0.06	0.40	-0.49	0.93	1.42
Lagged	.25	Closeness	-0.10	0.52	-1.00	1.00	2.00
Lagged	.25	InStrength	0.32	0.46	-0.47	0.95	1.42
Lagged	.25	OutStrength	0.22	0.40	-0.71	0.84	1.55
Lagged	.5	Betweenness	-0.06	0.40	-0.44	0.86	1.30
Lagged	.5	Closeness	-0.03	0.50	-0.67	1.00	1.67
Lagged	.5	InStrength	0.40	0.66	-0.39	1.00	1.39
Lagged	.5	OutStrength	0.13	0.37	-0.75	0.70	1.46
Lagged	association	Betweenness	0.04	0.39	-0.75	0.70	1.45
Lagged	association	Closeness	0.04	0.36	-0.75	0.75	1.50
Lagged	association	InStrength	0.00	0.39	-0.92	0.77	1.69
Lagged	association	OutStrength	0.03	0.37	-0.88	0.88	1.76
Lagged	gimme	Betweenness	0.02	0.49	-0.58	0.99	1.57
Lagged	gimme	Closeness	0.56	0.66	-1.00	1.00	2.00
Lagged	gimme	InStrength	-0.01	0.40	-0.59	0.84	1.43
Lagged	gimme	OutStrength	0.03	0.42	-0.69	1.00	1.69
Lagged	mlVAR_EB	Betweenness	0.58	0.46	-0.74	0.96	1.70
Lagged	mlVAR_EB	Closeness	0.46	0.32	-0.17	0.87	1.04
Lagged	mlVAR_EB	InStrength	0.65	0.38	-0.12	0.97	1.09
Lagged	mlVAR_EB	OutStrength	0.52	0.36	-0.31	0.91	1.23
Lagged	scaled	Betweenness	0.06	0.43	-0.52	0.86	1.38
Lagged	scaled	Closeness	0.02	0.50	-1.00	0.66	1.66
Lagged	scaled	InStrength	0.34	0.52	-0.33	0.98	1.31
Lagged	scaled	OutStrength	0.21	0.33	-0.52	0.80	1.31
Lagged	unregularized	Betweenness	0.06	0.43	-0.71	0.86	1.57
Lagged	unregularized	Closeness	0.00	0.38	-0.78	0.81	1.59
Lagged	unregularized	InStrength	0.01	0.44	-0.95	0.96	1.91
Lagged	unregularized	OutStrength	-0.04	0.40	-0.82	0.86	1.67

Footnotes

Footnote 1: Population Level Analyses

Bringmann et al. (2013) developed a technique for assessing ESM data using a network approach. This approach utilizes a series of univariate multilvel vector autoregressive models (mlVAR) in which all items are entered simultaneously as predictors and individually as outcomes. Below we run the models for the nine personality items at each wave. The mlVAR() function automatically within-centers data, so we will enter the raw data into the models. Because of the number of observations we have for each usbject we use a lag 1 factorization. Because we have more than 6 predictors, we use orthogonal estimation of lagged and contemporaneous effects.

Univariate Multilevel Autoregressive Models (footnote 1)

# raw data #
# not affected by using only complete cases of the data #
fit1_w1 <-   
  mlVAR(w1, 
        vars = colnames(w1)[6:14], #4:18
        idvar = "SID", 
        lags = 1,
        #dayvar = "day", 
        beepvar = "beep_seq",
        temporal = "orthogonal",
        contemporaneous = "orthogonal",
        verbose = TRUE, 
        scale = FALSE)

# raw data #
fit1_w2 <- 
  mlVAR(w2, 
        vars = colnames(w2)[5:13], #4:16
        idvar = "SID", 
        lags = 1,
        #dayvar = "day", 
        beepvar = "beep_seq",
        #covar = "Age",
        temporal = "orthogonal",
        contemporaneous = "orthogonal",
        verbose = TRUE,
        scale = FALSE)

#model summary
sum_fit1_w1        <- summary(fit1_w1)
sum_fit1_w2        <- summary(fit1_w2)

Network structure

The figure below presents the results of the four mlVAR models (1 lagged and 1 contemporaneous network for each wave). For the purposes of interpretation and comparison, the structure of the networks for the second wave have been constrained to match the first wave and only significant edges are displayed.

Plots

Fixed Effects

# The graphs below show the raw, directed network the estimated, directed network 
# using a univariate multilevel vector autoregressive models as (1) lagged and 
# (2) contemporaneous netowrks. Note that the code for the plot may appear complicated 
# at first glance but just largely deals with aesthetics of the graphs.  
b5_groups <- list(A = c(1,7), E = c(2, 6), C = c(3,8), N = c(4,5,9))
plot_fun <- function(mlVAR.obj, type, wave, sd = FALSE, Layout = "spring", 
                     noplot = TRUE, Nosig = 'hide'){
  plot <- plot(mlVAR.obj, type, layout = Layout, #labels = varnames2, 
               groups = b5_groups, nonsig = Nosig, curve = -1, legend = FALSE, 
               details = FALSE, mar = c(5,5,5,5), border.color = "black", 
               border.width = 2, title = sprintf("%s\nWave%s", str_to_title(type), wave),
               loop = .7, node.width = 1.6, edge.width = 1, asize = 5, label.font = 2,
               label.fill.vertical = 1, label.fill.horizontal = 1, edge.color = "blue", 
               color = t(brewer.pal(9, "Purples")[seq(1,7,2)]), DoNotPlot = noplot, SD = sd)
  # change negative edges to dashed
  plot$graphAttributes$Edges$lty[plot$Edgelist$weight < 0] <- 2
  # change edge colors based on weights
  if(sd == FALSE){
    if(type == "temporal"){
      plot$graphAttributes$Edges$color <-
      ifelse(plot$Edgelist$weight <.02, "thistle2",
      ifelse(plot$Edgelist$weight <.04, "mediumorchid", "midnightblue"))
      } else {
      plot$graphAttributes$Edges$color <- 
      ifelse(abs(plot$Edgelist$weight) < .1, "thistle2",
      ifelse(abs(plot$Edgelist$weight) < .3, "mediumorchid", "midnightblue"))}
  }
  #change variable names
  plot$graphAttributes$Nodes$labels <- str_replace(plot$graphAttributes$Nodes$labels, "_", "\n")
  # change labels of dark colored nodes to white
  dark_colors <- c("#9E9AC8", "#807DBA", "#6A51A3", "#54278F", "#3F007D")
  plot$graphAttributes$Nodes$label.color[plot$graphAttributes$Nodes$color %in% dark_colors] <- "white"
  return(plot)
}

# run but don't generate plots
plot_w1 <- plot_fun(fit1_w1, "temporal", "1")
plot_w1_contemp <- plot_fun(fit1_w1, "contemporaneous", "1")
plot_w2 <- plot_fun(fit1_w2, "temporal", "2", sd = FALSE, Layout = plot_w1$layout)
plot_w2_contemp <- plot_fun(fit1_w2, "contemporaneous", "2", Layout = plot_w1_contemp$layout)

# generate modified plots
par(mfrow = c(2,2))
plot(plot_w1)
plot(plot_w1_contemp)
plot(plot_w2)
plot(plot_w2_contemp)

Individual Differences

# generate sd plots
par(mfrow = c(2,2))
plot(plot_fun(fit1_w1, "temporal", "1", sd = TRUE, Layout = plot_w1$layout, noplot = TRUE))
plot(plot_fun(fit1_w1, "contemporaneous", "1", sd = TRUE, Layout = plot_w1_contemp$layout, noplot = TRUE, Nosig = 'show'))
plot(plot_fun(fit1_w2, "temporal", "2", sd = TRUE, Layout = plot_w1$layout, noplot = TRUE))
plot(plot_fun(fit1_w2, "contemporaneous", "2", sd = TRUE, Layout = plot_w1_contemp$layout, noplot = TRUE, Nosig = 'show'))

Centrality

In addition, we assessed the local network structure using centrality. Specifically, we calculated the strength centrality for lagged and contemporaneous networks at each wave separately. Because the lagged network is directed, we calculated both in strength and out strength for each node. We can use both centrality indices and edges to examine local properties of different nodes across both waves. In the population networks, central nodes represent consensus - behavioral patterns that were shared across people - while peripheral nodes represent idiosyncrasies - behavioral patterns that differed across people. For comparison across waves, we z-transformed all results for display in the figure below.
Centrality refers to the relative importance of a focal node to the structure and dynamics of a network. In other words, it provides information about a node’s role in the context of other nodes.

# save the lagged results to data frames
lagged_effects_w1 <- tbl_df(sum_fit1_w1$temporal) %>% mutate(wave = "1")
lagged_effects_w2 <- tbl_df(sum_fit1_w2$temporal) %>% mutate(wave = "2")

vars <- names(fit1_w1$output)

# save the contemporaneous results to matrices
contemp_effects_w1 <- fit1_w1$results$Theta$pcor$mean
contemp_effects_w2 <- fit1_w2$results$Theta$pcor$mean

colnames(contemp_effects_w1) <- vars; rownames(contemp_effects_w1) <- vars
colnames(contemp_effects_w2) <- vars; rownames(contemp_effects_w2) <- vars

# function to turn contemporaneous matrices to long format
contemp_long_fun <- function(fit, Wave){
  fit <- fit[,order(colnames(fit))]
  fit <- fit[order(rownames(fit)),]
  fit[lower.tri(fit, diag = T)] <- NA
  fit.long <- tbl_df(fit) %>%
    mutate(Var1 = colnames(.),
           type = "Contemporaneous", wave = Wave) %>%
    gather(key = Var2, value = weight, A_kind:N_worried) %>%
    filter(!is.na(weight)) %>%
    unite(var, Var1, Var2, sep = ".", remove = F)
}

# create long format contemporaneous results
contemp_eff_w1 <- contemp_long_fun(contemp_effects_w1, "1")
contemp_eff_w2 <- contemp_long_fun(contemp_effects_w2, "2")

### Run Centrality Analyses ###
#lagged
#raw
lagged_centrality_w1      <- centrality_auto(sum_fit1_w1$temporal[,c(1,2,4)])
lagged_centrality_w2      <- centrality_auto(sum_fit1_w2$temporal[,c(1,2,4)])

#contemporaneous
#raw
contemporaneous_centrality_w1      <- centrality_auto(contemp_effects_w1)
contemporaneous_centrality_w2      <- centrality_auto(contemp_effects_w2)

# save centrality results into data frame #
# lagged #
temp_cent <- lagged_centrality_w1$node.centrality %>% data.frame() %>%
  mutate(wave = "1", type = "Lagged", var = rownames(.)) %>%
  full_join(lagged_centrality_w2$node.centrality %>% data.frame() %>%
  mutate(wave = "2", type = "Lagged", var = rownames(.)))

# contemporaneous #
contemp_cent <- contemporaneous_centrality_w1$node.centrality %>%
  mutate(wave = "1", type = "Contemporaneous", var = rownames(.)) %>%
  full_join((contemporaneous_centrality_w2$node.centrality %>%
  mutate(wave = "2", type = "Contemporaneous", var = rownames(.))))

Centrality Plots

# Lagged #
# wrangle to long format and calculate standardized indices #
temp_cent_long <- temp_cent %>%
  gather(key = Measure, value = Centrality, Betweenness:OutStrength) %>%
  group_by(wave, Measure) %>%
  mutate(z = as.numeric(scale(Centrality)))

# Contemporaneous #
# wrangle to long format and calculate standardized indices #
contemp_cent_long <- contemp_cent %>%
  gather(key = Measure, value = Centrality, Betweenness:Strength) %>%
  group_by(wave, Measure) %>%
  mutate(z = as.numeric(scale(Centrality)))

# create combined data frame of lagged and contemporaneous effects #
combined_centrality <- temp_cent_long %>% full_join(contemp_cent_long) %>% ungroup()

combined_centrality %>%
  filter(grepl("trength", Measure)) %>%
  mutate(type = factor(type, levels = rev(sort(unique(type))))) %>%
  ggplot(aes(x = var, y = z, group = wave))+
  geom_line(aes(linetype = wave), color = "black", size = .3) + 
  geom_point(aes(shape = wave), size = 2) + 
  scale_y_continuous(limits = c(-3,3), breaks = seq(-3,3,1)) + 
  geom_hline(aes(yintercept = 0)) + 
  labs(x = NULL, y = "z-score") +
  coord_flip() + 
  facet_wrap(~type + Measure, nrow = 1) + 
  theme_classic()+ 
  theme(axis.text = element_text(face = "bold"),
        axis.title = element_text(face = "bold"),
        legend.title = element_text(face = "bold"),
        legend.position = "bottom")#c(.87, .25))

# ggsave(sprintf("%s/results/plots/centrality_plot.png", res_path), width = 5, height = 3.5)

Table Lagged

# wrangle to wide format for tabling #
temp_cent %>% select(-contains("fluence")) %>%
  gather(key = Measure, value = Centrality, Betweenness: OutStrength) %>%
  mutate(var = str_replace(var, "_", " ")) %>%
  unite(Measure, Measure, wave, sep = ".") %>%
  spread(key = Measure, value = Centrality) %>%
  mutate_at(vars(InStrength.1:OutStrength.2), funs(round(.,2))) %>%
  mutate_at(vars(Closeness.1:Closeness.2), funs(sprintf("%.2e", .))) %>%
  select(-type) %>%
  kable(., "html", booktabs = T, escape = F, digits = 2, 
        col.names = c("", rep(c("1","2"), times = 4))) %>%
  kable_styling(bootstrap_options = c("striped","repeat_header"), full_width = T) %>%
  #kable_styling(full_width = F) %>%
  column_spec(1, width = "4cm") %>%
  add_header_above(c(" " = 1,  "Betweenness" = 2, "Closeness" = 2,
                     "In Strength" = 2, "Out Strength" = 2))

	Betweenness		Closeness		In Strength		Out Strength
	1	2	1	2	1	2	1	2
A kind	7	5	2.14e-03	2.10e-03	0.11	0.14	0.13	0.13
A rude	3	0	3.01e-03	3.35e-03	0.07	0.07	0.22	0.16
C lazy	4	3	1.75e-03	2.43e-03	0.19	0.16	0.12	0.16
C reliable	2	8	1.67e-03	2.71e-03	0.09	0.16	0.09	0.18
E outgoing	1	3	1.35e-03	2.60e-03	0.21	0.17	0.07	0.13
E quiet	3	1	1.12e-03	1.69e-03	0.15	0.19	0.05	0.09
N depressed	4	0	2.32e-03	2.58e-03	0.12	0.11	0.14	0.16
N relaxed	4	1	2.15e-03	1.51e-03	0.20	0.26	0.16	0.10
N worried	13	16	3.04e-03	3.93e-03	0.16	0.16	0.32	0.31

Table Contemporaneous

# wrangle to wide format #
contemp_cent %>% select(-ExpectedInfluence) %>%
  gather(key = Measure, value = Centrality, Betweenness: Strength) %>%
  mutate(var = str_replace(var, "_", " ")) %>%
  unite(Measure, Measure, wave, sep = ".") %>%
  spread(key = Measure, value = Centrality) %>%
  mutate_at(vars(Strength.1:Strength.2), funs(round(.,2))) %>%
  mutate_at(vars(Closeness.1:Closeness.2), funs(sprintf("%.2e", .))) %>%
  select(-type) %>%
  kable(., "html", booktabs = T, escape = F, digits = 2, 
        col.names = c("", rep(c("1","2"), times = 3))) %>%
  kable_styling(bootstrap_options = c("striped","repeat_header"),full_width = T) %>%
  #kable_styling(full_width = F) %>%
  column_spec(1, width = "4cm") %>%
  add_header_above(c(" " = 1,  "Betweenness" = 2, "Closeness" = 2,
                     "Strength" = 2))

	Betweenness		Closeness		Strength
	1	2	1	2	1	2
A kind	3	6	1.23e-02	1.17e-02	0.68	0.70
A rude	0	1	9.38e-03	9.00e-03	0.57	0.50
C lazy	2	2	1.12e-02	1.13e-02	0.74	0.70
C reliable	2	2	1.14e-02	1.19e-02	0.67	0.68
E outgoing	6	6	1.13e-02	1.12e-02	1.07	1.08
E quiet	1	0	1.03e-02	9.98e-03	0.87	0.81
N depressed	1	1	1.10e-02	1.07e-02	0.77	0.74
N relaxed	8	8	1.33e-02	1.28e-02	1.05	1.02
N worried	0	0	1.16e-02	1.15e-02	0.75	0.80

Edge Weights

Contemporaneous networks are undirected and do not have feedback loops - that is, temporal precedence is unclear when behaviors are rated at the same time (see Figure 2). The strongest edge in the contemporaneous networks was between the two Extraversion items (quiet-outgoing: \(b_{W1}\) = -.61; \(b_{W2}\) = -.63). Indeed, many of the strongest relationships within the contemporaneous were within Big 5 domains (e.g. E1-E2, A1-A2) and were negative, which is not surprising given that one of the items in each domain was negatively keyed. Such relationships were coherent - participants reported feeling more lazy when they were less reliable (and vice versa; \(b_{W1}\) = -.28, \(b_{W2}\) = -.31) and more relaxed when they were less worried (and vice versa; \(b_{W1}\) = -.38, \(b_{W2}\) = -.38). There were notably strong relationships across domains as well. Reports of relaxation co-occurred with reports of feeling both lazy (\(b_{W1}\) = .26; \(b_{W2}\) = .24) and kind (\(b_{W1}\) = .11, \(b_{W2}\) = .08). Reports of kindness, in turn, were associated with feeling more outgoing ($b_{W1} = .13, \(b_{W2}\) = .12) and reliable (\(b_{W1}\) = .11, \(b_{W2}\) = .10).
The strongest edge in the lagged networks in both waves was the feedback loop of the Neuroticism item “worried” (worried, \(b_{W1}\) = .14; \(b_{W2}\) = .16). Feedback loops can be interpreted as partial autocorrelations, which has been termed inertia in the affect literature (Ong & Ram, 2016). In other words, worrying seemed to carry over across time points, strongly predicting itself four hours later. Worry also strongly predicted other nodes, including reports of relaxation (worried - relaxed, \(b_{W1}\) = -.11; \(b_{W2}\) = -.08), being quiet (worried - quiet, \(b_{W1}\) = 03; \(b_{W2}\) = .08), and feeling depressed (\(b_{W1}\) = .05; \(b_{W2}\) = .04). Notably, many of the strongest edges included one or more Neuroticism nodes - ratings of items from the Neuroticism domain strongly predicted each other across time - with the top 5 connections across both waves including a Neuroticism node. Such a pattern was not observed within other traits - Extraversion (b’s < |.005|), Agreeableness (b’s < |.02|), and Conscientiousness (b’s < |.03|) items weakly predicted each other over four hour intervals. However, several items exhibited strong inertia - reports of being lazy (\(b_{W1}\) = .06; \(b_{W2}\) = .08), reliable (\(b_{W1}\) = .04; \(b_{W2}\) = .06), and kind (\(b_{W1}\) = .03; \(b_{W2}\) = .06) were significant predictors of themselves. Feedback loops are critical features of lagged networks because although they do not explain shifts between behaviors over time, they do explain the likelihood of continuing to engage in a behavior once you begin.

Centrality

Next, we examined the centrality of different nodes in the network. Centrality indexes the relative importance of different nodes in the network - that is, nodes’ abilities to directly impact other nodes in the network. Central nodes represent patterns of influence shared across people, while peripheral nodes represent more unique patterns of influence. Centrality for all nodes in both lagged and contemporaneous networks across waves are displayed in Figure 3.
In the contemporaneous networks, how outgoing and relaxed participants reported had strong direct impacts on their other concurrent behaviors. The worried (N) and rudeness (A) nodes had notably little direct impact on other behaviors. In other words, in the moment, becoming more worried (N) or rude (A) would not strongly impact other concurrent behaviors, whereas feeling more outgoing and relaxed would. Nodes within traits tended to markedly differ in their centrality indices. For example, quiet (E) and depressed were at or below overall average centrality across all three measures, while outgoing (E) and relaxed (N) were well above average. In other words, behaviors, not traits, tended to drive concurrent behaviors.
In the lagged networks, worried (N) and relaxed (N) had the highest out-strength and in-strength, respectively, across both waves. That is, how worried you reported being strongly predicted many other behaviors four hours later but was little impacted by other behaviors. In contrast, how relaxed you reported was strongly predicted by what you were doing four hours ago but did not strongly predict your behavior later. Reports of how rude (A) and kind (A) were the lowest in in-strength. How rude or kind participants reported being was little impacted by their previous behaviors. Conversely, participants’ reports of how quiet (E), outgoing (E), and reliable (C) they were had little bearing on their later behaviors. Together this suggests that nodes related to emotions (e.g. worried, relaxed) notably impact behavior, and nodes related to social behaviors (e.g. kind, outgoing) are little impacted by previous behaviors and little impact future behaviors.

Network stability

Next, we assessed the stability of the population networks over time. To assess stability, we first calculated the profile correlation between the mlVAR fixed effects edge weights across the waves for both lagged and contemporaneous effects. Profile correlations index the stability of a profile of values over time - that is, how stable the positions of values are relative to all other values. Population networks were highly stable across waves. Contemporaneous network stability was almost at ceiling (\(r\) = .99). Lagged network stability was still quite strong (\(r\) = .68) but was significantly less stable than the contemporaneous networks (\(z\) = 9.28, \(p\) < .001).

# Profile (Ipsative) Edge Consistency
# The easiest and most straightforward way to assess the sconsistency in responses across
# time is simply to correlate the regression coefficients at each time point with those
# at the other time points. We do this below for models generated using raw and centered data.  
# For these correlations, we are correlating two vectors, each of which contains 81
# weights (9 x 9). We do so once for lagged and once for contemporaneous effects.  

cors <- data.frame(
  comparison = c("W1 v. W2"),
  type = c("lagged", "contemporaneous"),
  raw_cor = 
    c(cor(sum_fit1_w1$temporal$fixed,
          sum_fit1_w2$temporal$fixed),
      cor(sum_fit1_w1$contemporaneous$pcor,
          sum_fit1_w2$contemporaneous$pcor)))

cors %>%
  kable(., "html", booktabs = T, escape = F, digits = 2) %>%
  kable_styling(bootstrap_options = c("striped","repeat_header"), full_width = T)

comparison	type	raw_cor
W1 v. W2	lagged	0.63
W1 v. W2	contemporaneous	0.99

Next, we calculated the profile correlations between the centrality indices of lagged and contemporaneous effects across the waves separately for each measure of centrality. Across both waves, centrality was stable. Lagged in strength (\(r\) = .86) was slightly more stable than lagged out strength (\(r\) = .70). Contemporaneous strength centrality stability (\(r\) = .98) was significantly more stable than lagged in strength (\(z\) = 2.66, \(p\) < .01) but not out strength (\(z\) = 1.92, \(p\) = .054).

#### Profile (Ipsative) Centrality Consistency
combined_centrality %>%
  select(-z) %>%
  spread(key = wave, value = Centrality) %>%
  group_by(type, Measure) %>%
  summarize(r = cor(`1`, `2`, use = "pairwise")) %>%
  spread(key = type, value = r) %>% arrange(Contemporaneous) %>%
  kable(., "html", booktabs = T, escape = F, digits = 2,
  caption = "Profile Correlations of Population Level Centrality Consistency") %>%
  kable_styling(bootstrap_options = c("striped","repeat_header"), full_width = T)

Profile Correlations of Population Level Centrality Consistency

Measure	Contemporaneous	Lagged
Betweenness	NA	NA
Closeness	NA	NA
InStrength	NA	NA
OutStrength	NA	NA
Strength	NA	NA

Footnote 4: P-factor Analyses

Reference Factor Analyses

Typically, idiographic personality structure has been approached from a factor analytic standpoint using so-called P-technique factor analysis (e.g. Borkenau & Ostendorpf, 1998; Molenaar, 2004). Thus, to contrast with individual differences in network consistency, we also sought to replicate past research by conducting a series of individual level idiographic P-factor analyses of repeated-measures data. Concurrent examination of both factor analytic and network approaches allows us to better understand the relative advantages of each.
For the idiographic P-factor analyses, we calculated a time series varimax rotated factor analysis for each participant at each wave (\(N_{W1}\) = 348, \(N_{W2}\) = 146), using the psych package in R (Revelle, 2014). Using an eigenvalue criteria of 1, we found that the time-series factor solution ranged from 1 to 4 factors, with a median of 3 for both waves. 4.9% of the P-factor solutions in the first wave and 6.4% of the P-factor solutions in the second wave had the number of factors expected by the 9 ESM items putative factor structure (4). Despite this, the content of the 4 extracted factors of even this small minority of factor solutions differed from expectations. Across all subjects, the first factor explained between 11.4% and 37.7% (\(M_{W1}\) = 20.3%, \(SD_{W1}\) = 3.9%) of the variance in Wave 1 and between 9.6% and 37.7% (\(M_{W2}\) = 19.5%, \(SD_{W2}\) = 4.3%) in Wave 2. The cumulative variance explained by the extracted factor solution ranged from 12.4% to 79.6% (\(M_{W1}\) = 44.6%, \(SD_{W1}\) = 11.4%) in wave 1 and from 13.5% to 70.7% (\(M_{W2}\) = 42.1%, \(SD_{W2}\) = 10.8%) in wave 2.

# calculate individual for each variable
w1_pop <- w1 %>%
  mutate_at(vars(A_rude:N_relaxed), funs(mapvalues(., from = 1:5, to = 5:1))) %>%
  gather(key = item, value = value, A_rude:N_worried) %>%
  separate(item, c("trait", "item"), sep = "_") %>%
  group_by(SID, trait, wave) %>%
  summarize(esm = mean(value, na.rm = T)) %>%
  ungroup() %>% mutate(trait = paste(trait, "esm", sep = ".")) %>%
  spread(key = trait, value = esm) %>%
  full_join(w1 %>%
  group_by(SID) %>%
  summarize_at(vars(A_rude:N_worried), funs(mean)) %>%
  ungroup()) 
w2_pop <- w2 %>%
  mutate_at(vars(A_rude:N_relaxed), funs(mapvalues(., from = 1:5, to = 5:1))) %>%
  gather(key = item, value = value, A_rude:N_worried) %>%
  separate(item, c("trait", "item"), sep = "_") %>%
  group_by(SID, trait, wave) %>%
  summarize(esm = mean(value, na.rm = T)) %>%
  ungroup() %>% mutate(trait = paste(trait, "esm", sep = ".")) %>%
  spread(key = trait, value = esm) %>%
  full_join(w2 %>%
  group_by(SID) %>%
  summarize_at(vars(A_rude:N_worried), funs(mean)) %>%
  ungroup()) 

w1w2_subs <- w1_pop$SID[w1_pop$SID %in% w2_pop$SID]
w1_pop$wave <- "1"; w2_pop$wave <- "2"

composites.long <- w1_pop %>%
  full_join(w2_pop) %>%
  filter(SID %in% w1w2_subs) %>%
  gather(key = Variable, value = composite,-SID, -wave)

r_cors_w1 <- cor(w1_pop[,3:11])
r_cors_w2 <- cor(w2_pop[,3:11])

fa_w1  <- fa(r_cors_w1,  nfactors = 7, rotate = "varimax", 
              n.obs = length(w1_pop))
fa_w2 <- fa(r_cors_w2, nfactors = 7, rotate = "varimax", 
              n.obs = length(w2_pop))

factors_w1  <- sum(fa_w1$values  > 1)
factors_w2 <- sum(fa_w2$values > 1)

sink("/dev/null")
pop_var_w1 <- print(fa_w1)$Vaccounted[3, factors_w1]

## Factor Analysis using method =  minres
## Call: fa(r = r_cors_w1, nfactors = 7, n.obs = length(w1_pop), rotate = "varimax")
## Standardized loadings (pattern matrix) based upon correlation matrix
##               MR3   MR2   MR4   MR1   MR5   MR6   MR7   h2     u2 com
## A.esm       -0.09  0.11 -0.05 -0.02  0.46  0.00  0.00 0.23 0.7650 1.2
## C.esm       -0.92  0.18 -0.13 -0.07  0.26 -0.11  0.10 0.99 0.0088 1.4
## E.esm       -0.15  0.92 -0.14 -0.04  0.27  0.00 -0.04 0.96 0.0373 1.3
## N.esm        0.17 -0.11  0.80  0.55 -0.04  0.03 -0.04 0.99 0.0074 1.9
## A_rude       0.42  0.08  0.24  0.25  0.01  0.34  0.00 0.42 0.5844 3.4
## E_quiet      0.13 -0.95  0.03  0.07 -0.04 -0.03 -0.04 0.93 0.0695 1.1
## C_lazy       0.90 -0.20  0.02  0.16 -0.04 -0.04  0.09 0.89 0.1088 1.2
## N_relaxed   -0.06  0.09 -0.93 -0.15  0.12 -0.07 -0.02 0.92 0.0800 1.1
## N_depressed  0.24 -0.10  0.41  0.82 -0.05  0.08  0.01 0.92 0.0762 1.7
## 
##                        MR3  MR2  MR4  MR1  MR5  MR6  MR7
## SS loadings           1.98 1.87 1.77 1.11 0.37 0.14 0.02
## Proportion Var        0.22 0.21 0.20 0.12 0.04 0.02 0.00
## Cumulative Var        0.22 0.43 0.62 0.75 0.79 0.80 0.81
## Proportion Explained  0.27 0.26 0.24 0.15 0.05 0.02 0.00
## Cumulative Proportion 0.27 0.53 0.77 0.93 0.98 1.00 1.00
## 
## Mean item complexity =  1.6
## Test of the hypothesis that 7 factors are sufficient.
## 
## The degrees of freedom for the null model are  36  and the objective function was  7.42 with Chi Square of  75.43
## The degrees of freedom for the model are -6  and the objective function was  0 
## 
## The root mean square of the residuals (RMSR) is  0 
## The df corrected root mean square of the residuals is  NA 
## 
## The harmonic number of observations is  15 with the empirical chi square  0  with prob <  NA 
## The total number of observations was  15  with Likelihood Chi Square =  0  with prob <  NA 
## 
## Tucker Lewis Index of factoring reliability =  8.49
## Fit based upon off diagonal values = 1
## Measures of factor score adequacy             
##                                                    MR3  MR2  MR4  MR1  MR5
## Correlation of (regression) scores with factors   0.98 0.98 0.97 0.93 0.72
## Multiple R square of scores with factors          0.95 0.96 0.93 0.87 0.53
## Minimum correlation of possible factor scores     0.91 0.91 0.87 0.74 0.05
##                                                     MR6   MR7
## Correlation of (regression) scores with factors    0.50  0.53
## Multiple R square of scores with factors           0.25  0.28
## Minimum correlation of possible factor scores     -0.50 -0.44

pop_var_w2 <- print(fa_w2)$Vaccounted[3, factors_w2]

## Factor Analysis using method =  minres
## Call: fa(r = r_cors_w2, nfactors = 7, n.obs = length(w2_pop), rotate = "varimax")
## Standardized loadings (pattern matrix) based upon correlation matrix
##               MR3   MR2   MR5   MR1   MR4   MR6   MR7   h2     u2 com
## A.esm       -0.11  0.18 -0.07 -0.05  0.65 -0.05  0.00 0.48 0.5171 1.3
## C.esm       -0.91  0.12 -0.15 -0.12  0.22 -0.05  0.03 0.93 0.0679 1.3
## E.esm       -0.13  0.92 -0.15 -0.08  0.26  0.14  0.12 0.99 0.0128 1.4
## N.esm        0.18 -0.11  0.80  0.54 -0.05  0.09 -0.07 0.99 0.0061 2.0
## A_rude       0.44  0.12  0.17  0.23 -0.18  0.41  0.00 0.50 0.5021 3.4
## E_quiet      0.08 -0.96  0.04  0.08 -0.06  0.06  0.09 0.96 0.0390 1.1
## C_lazy       0.96 -0.15  0.05  0.13  0.02  0.05  0.02 0.96 0.0435 1.1
## N_relaxed   -0.08  0.08 -0.89 -0.12  0.09 -0.03 -0.03 0.83 0.1678 1.1
## N_depressed  0.24 -0.16  0.37  0.79 -0.08  0.09  0.01 0.86 0.1415 1.8
## 
##                        MR3  MR2  MR5  MR1  MR4  MR6  MR7
## SS loadings           2.07 1.89 1.65 1.04 0.60 0.22 0.03
## Proportion Var        0.23 0.21 0.18 0.12 0.07 0.02 0.00
## Cumulative Var        0.23 0.44 0.62 0.74 0.81 0.83 0.83
## Proportion Explained  0.28 0.25 0.22 0.14 0.08 0.03 0.00
## Cumulative Proportion 0.28 0.53 0.75 0.89 0.97 1.00 1.00
## 
## Mean item complexity =  1.6
## Test of the hypothesis that 7 factors are sufficient.
## 
## The degrees of freedom for the null model are  36  and the objective function was  7.14 with Chi Square of  72.55
## The degrees of freedom for the model are -6  and the objective function was  0 
## 
## The root mean square of the residuals (RMSR) is  0 
## The df corrected root mean square of the residuals is  NA 
## 
## The harmonic number of observations is  15 with the empirical chi square  0  with prob <  NA 
## The total number of observations was  15  with Likelihood Chi Square =  0  with prob <  NA 
## 
## Tucker Lewis Index of factoring reliability =  12.081
## Fit based upon off diagonal values = 1
## Measures of factor score adequacy             
##                                                    MR3  MR2  MR5  MR1  MR4
## Correlation of (regression) scores with factors   0.98 0.99 0.95 0.88 0.80
## Multiple R square of scores with factors          0.96 0.97 0.90 0.78 0.63
## Minimum correlation of possible factor scores     0.93 0.94 0.80 0.56 0.27
##                                                     MR6   MR7
## Correlation of (regression) scores with factors    0.65  0.56
## Multiple R square of scores with factors           0.42  0.32
## Minimum correlation of possible factor scores     -0.16 -0.37

sink()

Parallel Analysis

fa_par_w1 <- fa.parallel(w1_pop %>% select(A_rude:N_worried), fm = 'ml', 
      fa = 'fa', n.iter = 50, SMC = T, quant = .95)

## Parallel analysis suggests that the number of factors =  4  and the number of components =  NA

fa_par_w2 <- fa.parallel(w2_pop %>% select(A_rude:N_worried), fm = 'ml', 
      fa = 'fa', n.iter = 50, SMC = T, quant = .95)

## Parallel analysis suggests that the number of factors =  4  and the number of components =  NA

Idiographic P-Factor Analyses

cor_fun <- function(df, type){
  if(type == "fa"){r <- cor(df %>% select(A_rude:N_worried), use = "pairwise")}
  else{
    df <- df %>% select(A_rude:N_worried) %>% mutate_all(funs(lag = lag))
    r <- cor(df[,10:18], df[,1:9], use = "pairwise")
  }
  return(r)
}
par_fun <- function(df, n){
  fa.parallel(df, fm = 'ml', n.obs = n,
      fa = 'fa', n.iter = 50, SMC = T, quant = .95)
}
fa_fit <- w1 %>% full_join(w2) %>%
  group_by(SID, wave, count) %>%  
  arrange(SID, wave) %>%
  nest() %>%
  full_join(crossing(wave = c("1","2"), fa_type = c("fa", "dfa"))) %>%
  mutate(cor = map2(data, fa_type, possibly(cor_fun, NA_real_)),
         fa_par = map2(data, count, possibly(par_fun, NA_real_)),
         nfact = map_dbl(fa_par, possibly(~.$nfact, NA_real_)))

fa_fun <- function(df,nfact, SID, wave, fa_type){
  print(sprintf("%s Wave %s %s", SID, wave, fa_type))
  symm <- ifelse(fa_type == "dfa", FALSE, TRUE)
  if(!is.na(nfact) & nfact != 0){fa(df, nfactors = nfact, symmetric = symm)}
  }
fa_fit <- fa_fit %>%
  filter(!(SID == 13266 & wave == 1 & fa_type == "dfa")) %>%
  filter(!(SID == 28072 & wave == 2 & fa_type == "dfa")) %>%
  mutate(fa = pmap(list(cor, nfact, SID, wave, fa_type), possibly(fa_fun, NA_real_)))

Idiographic Factors

Vaccounted_fun <- function(fa, nfactor){
  y <- print(fa)$Vaccounted[3,]
  z <- y[nfactor]
  return(z)
}

sink("/dev/null")
#eigenvalue > 1 rule
fa_fit <- fa_fit %>% filter(!nfact == 0 & !is.na(nfact)) %>%
  mutate(Vacc_first = map_dbl(fa, possibly(~print(.)$Vaccounted[3,1], NA_real_)),
         Vacc_par = map2_dbl(fa, nfact, possibly(Vaccounted_fun, NA_real_)))
sink()

# descriptives of extracted factors from p-factors
fa_fit %>%
  select(fa_type, SID, wave, nfact, Vacc_first, Vacc_par) %>%
  gather(key = measure, value = value, nfact, Vacc_first, Vacc_par) %>%
  group_by(fa_type, wave, measure) %>%
  summarize_at(vars(value), funs(M = mean, SD = sd, Median = median, Min = min, Max = max), na.rm = T) %>%
  ungroup() %>%
  gather(key = ct, value = value, M:Max) %>%
  unite(measure, measure, wave, sep = ".") %>%
  spread(key = measure, value = value) %>%
  select(-fa_type) %>%
  kable(., "html", escape = F, booktabs = T, digits = 2,
        col.names = c("Measure", rep(c("W1", "W2"), times = 3)),
        caption = "Descriptives of P-factor Analyses") %>%
  add_header_above(header = c(" " = 1, "Parallel Analysis Factors" = 2, "Variance Accounted" = 2, 
                              "Variance First Factor" = 2)) %>%
  kable_styling(bootstrap_options = c("striped","repeat_header"), full_width = T) %>%
  group_rows("DFA", 1,5) %>%
  group_rows("FA", 6,10)

Descriptives of P-factor Analyses

	Parallel Analysis Factors		Variance Accounted		Variance First Factor
Measure	W1	W2	W1	W2	W1	W2
DFA
M	3.36	3.40	0.35	0.35	0.85	0.84
Max	7.00	9.00	0.68	0.77	1.00	1.00
Median	3.00	3.00	0.32	0.34	0.87	0.88
Min	1.00	1.00	0.18	0.18	0.47	0.50
SD	1.35	1.45	0.10	0.11	0.10	0.11
FA
M	3.32	3.44	0.23	0.23	0.61	0.60
Max	7.00	7.00	0.44	0.42	0.87	0.83
Median	3.00	3.00	0.22	0.21	0.62	0.61
Min	1.00	1.00	0.12	0.14	0.28	0.32
SD	1.35	1.37	0.05	0.05	0.11	0.11

Ipsative Consistency

cong_fun <- function(sid, type){
  fa1 <- (fa_fit %>% filter(fa_type == type & SID == sid & wave == 1))$fa[[1]]
  fa2 <- (fa_fit %>% filter(fa_type == type & SID == sid & wave == 2))$fa[[1]]
  factor.congruence(fa1, fa2)
}

ip_eigen_fun <- function(sid, type){
  fa1 <- (fa_fit %>% filter(fa_type == type & SID == sid & wave == 1))$fa[[1]]
  fa2 <- (fa_fit %>% filter(fa_type == type & SID == sid & wave == 2))$fa[[1]]
  cor(fa1$values, fa2$values, use = "pairwise")
}

fa_cong <- fa_fit %>% 
  filter(!is.na(fa)) %>% 
  group_by(SID, fa_type) %>% 
  summarize(n = n()) %>% 
  ungroup() %>%
  filter(n == 2) %>%
  mutate(congruence = map2(SID, fa_type, possibly(cong_fun, NA_real_)),
         ip_eigen = map2_dbl(SID, fa_type, possibly(ip_eigen_fun, NA_real_)))

fa_cong %>% 
  ggplot(aes(x = ip_eigen)) + 
  geom_histogram(aes(fill = fa_type), color = "black") + 
  facet_grid(~fa_type) + 
  theme_classic() + 
  theme(legend.position = "none")

Rank Order Consistency

fa_extract_fun <- function(fit){
  tibble(var = rownames(fit$weights), eigen = fit$values)
}

fa_fit <- fa_fit %>%
  mutate(eigen = map(fa, possibly(fa_extract_fun, NA_real_)))

fa_fit %>%
  filter(!is.na(eigen)) %>%
  group_by(SID, fa_type) %>%
  mutate(n = n()) %>% 
  filter(n == 2) %>%
  unnest(eigen) %>%
  group_by(wave, fa_type, var) %>%
  mutate(rank = dense_rank(desc(eigen))) %>%
  select(-count, -nfact, -eigen, -n, -Vacc_first, -Vacc_par) %>%
  spread(key = wave, value = rank) %>%
  group_by(fa_type, var) %>%
  summarize(r = cor(`1`, `2`, use = "pairwise")) %>%
  spread(key = fa_type, value = r)
save(fa_fit, fa_cong, file = "/Users/emoriebeck/Box/network/PAIRS/consistency/results/fa_results.RData")

Footnote 5: Pairwise Profile Congruence

Differences were not unique to the two example subjects. Globally, we can compare each participant’s profile of edge weights to all other participants’ profiles to assess similarity in network structure. On average, pairwise congruence was higher in the contemporaneous networks (\(M_{W1}\) = .44; \(M_{W2}\) = .45) than the lagged networks (\(M_{W1}\) = .01; \(M_{W2}\) = .01). The contemporaneous estimates suggest that there are similarities across people, with their profiles showing modest correlations, on average. This is to be expected, as it suggests that relationship between variables that are presumed in factor models is justified. In contrast, it appears that people have very little overlap in lagged edge weights, on average.

r_fun <- function(df){
  x <- data.frame((df %>% select(-SID)))
  rownames(x) <- df$SID
  y <- cor(t(x), use = "pairwise")
  y[upper.tri(y, diag = T)] <- NA
  z <- tbl_df(data.frame(y) %>%
    mutate(SID = colnames(y)) %>%
    gather(key = SID2, value = r, -SID) %>%
    mutate(SID2 = gsub("X", "", SID2))) %>%
    filter(!is.na(r)) 
}

pcong_cors <- temp_long %>%
  select(SID, wave, lambda_scale, type, from, to, weight) %>%
  unite(var, from, to, sep = ".") %>%
  full_join(contemp_long %>% select(SID, wave, lambda_scale, var, type, weight)) %>%
  spread(key = var, value = weight) %>%
  group_by(wave, lambda_scale, type) %>%
  nest() %>%
  mutate(pcr = map(data, r_fun))

pcong_cors %>% unnest(pcr, .drop = T) %>%
  filter(!is.na(r)) %>%
  group_by(wave, lambda_scale, type) %>%
  summarize(mean = meanSD_r2z2r(r)[1],
            sd = sd(r, na.rm = T)) %>%
  kable(., "html", digits = 2, booktabs = T,
        col.names = c("Wave","Model", "Type", "M", "SD"),
        caption = "Descriptives of Pairwise Congruence Correlations") %>%
  kable_styling(bootstrap_options = c("striped","repeat_header"), full_width = T)

Descriptives of Pairwise Congruence Correlations

Wave	Model	Type	M	SD
1	.25	Contemporaneous	0.49	0.26
1	.25	Lagged	0.01	0.14
1	.5	Contemporaneous	0.51	0.26
1	.5	Lagged	0.00	0.15
1	association	Contemporaneous	0.49	0.19
1	association	Lagged	0.04	0.17
1	gimme	Contemporaneous	0.32	0.21
1	gimme	Lagged	0.02	0.16
1	mlVAR_EB	Contemporaneous	0.90	0.05
1	mlVAR_EB	Lagged	0.64	0.18
1	scaled	Contemporaneous	0.61	0.30
1	scaled	Lagged	0.00	0.15
1	unregularized	Contemporaneous	0.26	0.19
1	unregularized	Lagged	0.01	0.14
2	.25	Contemporaneous	0.49	0.28
2	.25	Lagged	0.01	0.14
2	.5	Contemporaneous	0.50	0.29
2	.5	Lagged	0.00	0.15
2	association	Contemporaneous	0.49	0.20
2	association	Lagged	0.04	0.18
2	gimme	Contemporaneous	0.32	0.24
2	gimme	Lagged	0.04	0.19
2	mlVAR_EB	Contemporaneous	0.92	0.06
2	mlVAR_EB	Lagged	0.73	0.14
2	scaled	Contemporaneous	0.60	0.34
2	scaled	Lagged	0.00	0.14
2	unregularized	Contemporaneous	0.35	0.19
2	unregularized	Lagged	0.02	0.14

Footnote 7: ESM Composites

For comparison purposes, we also created a composite of ESM assessments. We first calculated individuals’ mean ratings on each of the personality variable for each wave. We also calculated composite scores for each Big 5 traits at each measurement point. Then, for each wave and item / composite combination, we assigned ranks to participants based on their ESM composites. We used the ranks to calculate rank-order correlations using Spearman correlations, resulting in 14 rank order correlations (1 for each of the 9 items and 4 composites). Overall, rank order stability was fairly high across the two years for both items (range .46 to .74) and composites (range .68 to .79). Finally, we calculated profile correlations for items and composites separately for each participant across the two waves. Overall, consistency of both Big 5 composites (M = .97) and items (M = .91) was very strong over two years. There were also interindividual differences in intraindividual consistency of both composites (SD = .66) and items (SD = .47). We use these to as a benchmark to which to compare rank order and ipsative consistency of idiographic networks.

#compute scale scores for BF domains
#first create the keys by location (the conventional way)
keys.list <- list(
  E.esm = c(-2, 6),
  A.esm = c(-1, 7),
  C.esm = c(-3, 8),
  N.esm = c(-4, 5, 9))
keys <- make.keys(9,keys.list,item.labels=colnames(w1)[5:13])

# score the items and save to columns
ncol_w1 <- dim(w1)[2]; ncol_w2 <- dim(w2)[2]
w1[(ncol_w1 + 1):(ncol_w1 + 4)] <- scoreItems(keys,w1[,5:13],min=1,max=5)$scores
w2[(ncol_w2 + 1):(ncol_w2 + 4)] <- scoreItems(keys,w2[,5:13],min=1,max=5)$scores
# rename the new columns
colnames(w1)[(ncol_w1 + 1):(ncol_w1 + 4)] <- names(keys.list)
colnames(w2)[(ncol_w2 + 1):(ncol_w2 + 4)] <- names(keys.list)

# create a data frame merging the responses
esm.composites <- w1 %>% mutate(wave = "1") %>%
  select(SID, wave, A_rude:N_worried, E.esm:N.esm) %>%
  full_join(w2 %>% mutate(wave = "2") %>%
    select(SID, wave, A_rude:N_worried, E.esm:N.esm)) %>%
  mutate(SID = as.character(SID)) %>%
  gather(key = item, value = value, A_rude:N.esm) %>%
  group_by(SID, wave, item) %>%
  summarize(mean = mean(value, na.rm = T)) %>%
  spread(key = item, value = mean)

Variable Centered: Rank-Order Change

cor_fun <- function(x){
  cor(x$`1`, x$`2`, use = "pairwise", method = "spearman")
}

mean_cors <- composites.long %>%
  mutate(type = ifelse(grepl("_", Variable) == T, "Item", "Composite")) %>%
  filter(SID %in% w1w2_subs) %>%
  group_by(Variable, wave) %>%
  mutate(rank = min_rank(desc(composite))) %>%
  select(-composite) %>%
  spread(key = wave, value = rank) %>%
  group_by(Variable, type) %>%
  nest() %>%
  mutate(r = map_dbl(data, cor_fun))

mean_cors %>%
  group_by(type) %>%
  summarise(meanr = meanSD_r2z2r(r)[1],
            sd = meanSD_r2z2r(r)[2], 
            min = min(r),
            max = max(r)) %>%
  kable(., booktabs = T, digits = 2, format = "html",
        caption = "Descriptives of ESM Composites Rank-Order Consistency") %>%
  kable_styling(bootstrap_options = c("striped","repeat_header"), full_width = T)

Descriptives of ESM Composites Rank-Order Consistency

type	meanr	sd	min	max
Composite	0.63	0.24	0.46	0.79
Item	0.66	0.13	0.47	0.73

Person-Centered: Ipsative Change

w1$wave <- "1"; w2$wave <- "2" 
mean_profile_cors <- w1 %>%
  select(SID, wave, A_rude:N_worried, E.esm:N.esm) %>%
  full_join(select(w2, wave, SID, A_rude:N_worried, E.esm:N.esm)) %>%
  filter(SID %in% w1w2_subs) %>%
  group_by(wave, SID) %>%
  summarize_at(vars(A_rude:N.esm), funs(mean(., na.rm = T))) %>%
  ungroup() %>%
  gather(key = variable, value = rating, A_rude:N.esm) %>%
  mutate(type = ifelse(grepl("_", variable) == T, "Item", "Composite")) %>%
  group_by(SID, type) %>%
  summarize(cor = cor(rating[wave == "1"], rating[wave == "2"], use = "pairwise.complete.obs")) %>%
  ungroup()

mean_profile_cors %>%
  ggplot(aes(x = cor)) + 
  geom_histogram(color = "black", fill = "gray") + 
  facet_grid(.~type) + 
  labs(x = "Profile Correlation", y = "Frequency", title = "Profile Correlations of ESM Composites") + 
  scale_x_continuous(lim = c(0,1), breaks = seq(0,1,.25)) + 
  theme_bw()

mean_profile_cors %>%
  group_by(type) %>%
  summarise(meanr = meanSD_r2z2r(cor)[1],
            sd = meanSD_r2z2r(cor)[2], 
            min = min(cor),
            max = max(cor)) %>%
  kable(., booktabs = T, digits = 2, format = "html",
        caption = "Descriptives of ESM Composites Ipsative Consistency") %>%
  kable_styling(bootstrap_options = c("striped","repeat_header"), full_width = T)

Descriptives of ESM Composites Ipsative Consistency

type	meanr	sd	min	max
Composite	0.97	0.62	0.14	1
Item	0.93	0.45	0.56	1

Footnote 8: Idiographic Individual Differences

An alternative way to index individual differences in network structure. To index whether people in general had similar edge weights (e.g., the relationship between outgoing and quiet, both items from extraversion) we is to examine the standard deviation of edge weights for both contemporaneous and lagged idiographic networks see Figure Sxx in the Supplementary Materials). Significant variation existed across edges and types of networks. In the contemporaneous networks, the outgoing (E) - quiet (E) edge in the contemporaneous networks was most variable in both waves (\(SD_{W1}\) = .28; \(SD_{W2}\) = .32), indicating that not everyone had the typical strong, negative correlation between feeling outgoing and quiet. The most variable edges for lagged were lazy (C) - rude (A) in wave 1 (\(SD\) = .87) and reliable (C) - worried (N) in wave 2 (\(SD\) = .92). Overall, it appears that most people manifest some similar behaviors when they feel they are rude (e.g., not-quiet, not-relaxed), but this is not always the case as many people were rude while being lazy time (Subject 10493) while others were rude while not being lazy (Subject 10549).

Edge Weights

Table

temp_long %>%
  rename(Var1 = from, Var2 = to) %>%
  select(wave, SID, type, lambda_scale, Var1, Var2, weight) %>%
  full_join(select(contemp_long, SID, wave, type, lambda_scale, Var1, Var2, weight)) %>%
  filter(Var1 == "E_quiet" & Var2 == "N_depressed" & type == "Contemporaneous") %>%
  mutate(sign = sign(weight)) %>%
  arrange(desc(weight))

  # group_by(wave, sign) %>%
  # summarize(n = n())

ew_ind_diff <- temp_long %>%
  rename(Var1 = from, Var2 = to) %>%
  select(wave, type, lambda_scale, Var1, Var2, weight) %>%
  full_join(select(contemp_long, wave, type, lambda_scale, Var1, Var2, weight)) %>%
  group_by(type, wave, lambda_scale, Var1, Var2) %>%
  summarize(sd = fisherz2r(sd(fisherz(weight), na.rm = T)))

Plot

graph_fun <- function(dat, type, wave){
  if(type == "Lagged"){
    qgraph(dat, directed = T, node.width = 2,  arrows = T, layout = "circle",
           title=sprintf("%s Wave %s", type, wave), edge.color = "blue", 
           mar = rep(6,6), label.font = 2, label.fill.vertical = 1, 
           label.fill.horizontal = 1, color = "white")
  } else if(type == "Contemporaneous"){
    qgraph(dat, directed = F, node.width = 2, arrows = F, layout = "circle",
           title=sprintf("%s Wave %s", type, wave), edge.color = "blue", 
           mar = rep(6,6), label.font = 2, label.fill.vertical = 1, 
           label.fill.horizontal = 1, color = "white")
  }
}

par(mfrow = c(2,2))
ew_ind_diff_nested <- ew_ind_diff %>% ungroup() %>%
  mutate(Var1 = str_replace(Var1, "_", "\n"),
         Var2 = str_replace(Var2, "_", "\n")) %>%
  group_by(type, lambda_scale, wave) %>%
  nest() %>%
  arrange(wave, lambda_scale, rev(type)) %>%
  mutate(graph = pmap(list(data, type, wave), graph_fun))

Individual Differences in Centrality

To look at individual differences in centrality, we can compute the standard deviation of the each edge for each measure. We do so below and display in a plot.

temp_centrality %>%
  gather(key = measure, value = value, Betweenness:OutStrength) %>%
  full_join(contemp_centrality %>%
              gather(key = measure, value = value, Betweenness:Strength)) %>%
  group_by(SID, wave, type, lambda_scale, measure) %>%
  mutate(z = as.numeric(scale(value))) %>%
  ungroup() %>%
  group_by(wave, lambda_scale, var, measure, type) %>%
  mutate(sd = sd(value, na.rm = T)) %>%
  filter(grepl("trengt", measure) & lambda_scale == ".25") %>%
  ungroup() %>%
  mutate(wave = as.character(wave)) %>%
  ggplot(aes(x = var, y = sd, group = wave))+
  geom_line(aes(linetype = wave), color = "black", size = .3) + 
  geom_point(aes(shape = wave), size = 2) + 
  labs(x = NULL, y = "z-score", linetype = "Wave", shape = "Wave") +
  scale_y_continuous(limits = c(0,3), breaks = seq(0,3,1)) + 
  geom_hline(aes(yintercept = 0)) + 
  coord_flip() + 
  facet_wrap(~type + measure, nrow = 1) + 
  theme_classic()+ 
  theme(axis.text = element_text(face = "bold"),
        axis.title = element_text(face = "bold"),
        legend.title = element_text(face = "bold"),
        legend.position = "bottom")#c(.75, .25))

Supplemental Analyses

Relationships Among Consistency Estimates: Do Networks Capture Unique Aspects of Development?

consis_rs <- ip_cors %>% 
  mutate(type2 = "Edge Weights") %>% rename(r = cors) %>%
  full_join(ip_cent_cors %>% filter(grepl("trength", measure)) %>%
              mutate(type2 = "Centrality") %>% unite(type, type, measure, sep = ".")) %>% 
  full_join(mean_profile_cors %>% mutate(type2 = type, type = "Composites") %>%
              rename(r = cor)) %>%
  full_join(ip_split_cors %>% mutate(type2 = sprintf("Edge Weights: %s", split_type)) %>%
              select(-split_type) %>% unite(type, type, wave, sep = ".")) %>%
  ungroup() %>%
  unite(type, type, type2, lambda_scale, sep = "_") %>%
  spread(key = type, value = r) %>%
  select(-SID) %>%
  cor(., use = "pairwise")


# new.names <- c("Big 5 Composites", "Big 5 Item Composites", "Contemporaneous Edge Weights",
#                "Contemporaneous Odd-Even Edge Weights (W1)","Contemporaneous Split-Half Edge Weights (W1)",
#                "Contemporaneous Odd-Even Edge Weights (W2)","Contemporaneous Split-Half Edge Weights (W2)",
#                "Contemporaneous Strength Centrality", "Lagged Edge Weights",
#                "Lagged Odd-Even Edge Weights (W1)","Lagged Split-Half Edge Weights (W1)",
#                "Lagged Odd-Even Edge Weights (W2)","Lagged Split-Half Edge Weights (W2)",
#                "Lagged In Strength Centrality", "Lagged Out Strength Centrality")
# 
# consis_rs[lower.tri(consis_rs)] <- sprintf("%.2f", consis_rs[lower.tri(consis_rs)])
# consis_rs[upper.tri(consis_rs)] <- ""
# diag(consis_rs) <- "--"
# data.frame(consis_rs) %>%
#   mutate(var = rownames(.),
#          var = mapvalues(var, unique(var), new.names),
#     var = paste(seq(1, ncol(consis_rs),1), var, sep = ". ")) %>%
#   select(var, everything()) %>%
#   setNames(c("Measure", seq(1,ncol(consis_rs),1))) %>%
#   kable(., "html", booktabs = T, escape = F, digits = 2,
#         caption = "Correlations Between Ipsative Consistency Measures") %>%
#   kable_styling(bootstrap_options = c("striped","repeat_header"), full_width = T)

Profile Consistency Correlates: Profile Consistency Correlates: What May Explain Individual Differences in Network Consistency?

# read outcome variables from wave 1 #
target.ratings.initial.w1 <- tbl_df(read.csv(url(sprintf("%s/raw/master/data/Wave 1/target_w1_RENAMED.csv?raw=T", res_path)))) %>%
  mutate(wave = "1", ts.IDnum.w1 = ifelse(ts.IDnum.w1 < 10000, ts.IDnum.w1 + 10000, ts.IDnum.w1))
target.ratings.initial.w2 <- tbl_df(read.csv(sprintf("%s/data/Wave 4/home_w4_RENAMED.csv", res_path))) %>%
  mutate(wave = "2", ts.IDnum.w4 = ifelse(ts.IDnum.w4 < 10000, ts.IDnum.w4 + 10000, ts.IDnum.w4))

names.w1 <- colnames(target.ratings.initial.w1)
names.w2 <- colnames(target.ratings.initial.w2)
names.w1 <- gsub(".w1", "", names.w1)
names.w2 <- gsub(".w4", "", names.w2)
colnames(target.ratings.initial.w1) <- names.w1
colnames(target.ratings.initial.w2) <- names.w2

target.ratings <- target.ratings.initial.w1 %>%
  full_join(target.ratings.initial.w2 %>%
              mutate(ts.AGQ03 = as.numeric(stringr::str_replace(ts.AGQ03, ",", ".")))) %>% 
  select(wave,contains("ts"), -X, -(ts.startDateTime:ts.PRO17), -ts.DEM04, 
         -(ts.DEM06:ts.DEM09), -(ts.DEM15:ts.DEM16), contains(".con")) %>%
  gather(key = item, value = value, -ts.IDnum, -wave) %>%
  group_by(item) %>% mutate(n = sum(!is.na(value))) %>%
  filter((wave == "1" & n > 200) | (wave == "2" & n > 175)) %>%
  select(-n) %>% rename(SID = ts.IDnum) %>%
  mutate(SID = ifelse(as.numeric(SID) < 10000, as.numeric(SID) + 10000, SID)) %>%
  spread(key = item, value = value)

target.depression <- target.ratings %>%
  select(SID, wave, contains("CESD")) %>%
  mutate_at(vars(ts.CESD01:ts.CESD10), funs(recode(., `1` = 0, `2` = 1, `3` = 2, `4` = 3))) %>%
  mutate_at(vars(ts.CESD06, ts.CESD03), funs(recode(., `0` = 3, `1` = 2, `2` = 1, `3` = 0))) %>%
  gather(key = item, value = rating, ts.CESD01:ts.CESD10) %>%
  group_by(SID, wave) %>%
  summarize(depression = sum(rating, na.rm = T)) %>%
  arrange(desc(depression)) %>%
  # mutate(wave = recode(wave, `4` = 2, `1` = 1)) %>%
  spread(key = wave, value = depression) %>%
  filter(!is.na(`1`) & !is.na(`2`)) %>%
  gather(key = wave, value = depression, -SID) %>%
  group_by(SID) %>%
  mutate(change = sign(depression[wave == 2] - depression[wave == 1])) %>% 
  ungroup() 

BFI_items <- paste("ts.", "BFI", c(paste("0", 1:9, sep = ""), 10:44), sep = "")
BFI_key <- c(paste(rep(c("E", "A", "C", "N", "O"), times = 8), 
             paste("0", rep(1:8, each = 5), sep = ""), sep = "_"), 
             "O_09", "A_09", "C_09", "O_10")

keys <- list(Extraversion = c(1, -6, 11, 16, -21, 26, -31, 36),
             Agreeableness = c(-2, 7, -12, 17, 22, -27, 32, -37, 42),
             Conscientiousness = c(3, -8, 13, -18, -23, 28, 33, 38, -43),
             Neuroticism = c(4, -9, 14, 19, -24, 29, -34, 39),
             Openness = c(5, 10, 15, 20, 25, 30, -35, 40, -41, 44))

BFI <- target.ratings %>% select(SID, wave, one_of(BFI_items)) %>%
  mutate_at(vars(-wave), funs(as.numeric)) %>%
  setNames(c("SID", "wave", BFI_key))
BFI[,47:51] <- scoreItems(keys, BFI[,3:46], min = 1, max = 15)$scores
colnames(BFI)[47:51] <- names(keys)

stab_cors <- ip_cors %>%
  left_join(target.ratings) %>%
  left_join(BFI %>% select(SID, wave, Extraversion:Openness)) %>%
  left_join(esm.composites %>% ungroup() %>% mutate(SID = as.numeric(SID))) %>%
  left_join(target.depression) %>%
  mutate_at(vars(ts.ACT01:ts.WT02), funs(as.numeric)) %>%
  gather(key = outcome, value = value, ts.ACT01:change) %>%
  group_by(wave, lambda_scale, outcome, type) %>%
  summarize(r = cor(cors, value, use = "pairwise")) %>%
  filter(!is.na(r)) %>%
  unite(comb, type, wave) %>%
  spread(key = comb, value = r) %>%
  filter(!is.na(Contemporaneous_2))

outcomes <- 
tribble(
  ~oldName, ~newName, ~prettyName,
  "ts.NQ02",   "intelligent", "Intelligence",
  "ts.NQ15",   "compassionate", "Compassion",
  "ts.NQ24",   "dominant", "Dominance",
  "ts.NQ33",   "lonely", "Loneliness",
  "ts.VQ16",   "giveBack", "Enjoys Giving Back",
  "ts.AGQ09",  "probWPartying", "Partying Causes Life Problems",
  "ts.ACT22",  "talkRelat", "Talks About Relationships",
  "ts.LE09",   "winAward", "Won An Award",
  "A.esm", "A.esm", "State Agreeableness",
  "E.esm", "E.esm", "State Extraversion",
  "C.esm", "C.esm", "State Conscientiousness",
  "N.esm", "N.esm", "State Neuroticism",
  "Agreeableness", "Agreeableness", "Trait Agreeableness",
  "Extraversion", "Extraversion", "Trait Extraversion",
  "Conscientiousness", "Conscientiousness", "Trait Conscientiousness",
  "Neuroticism", "Neuroticism", "Trait Neuroticism",
  "Openness", "Openness", "Trait Openness"
)

stab_cors %>% ungroup() %>%
  filter(outcome %in% outcomes$oldName) %>%
  mutate(#type = ifelse(grepl(".esm", outcome) == T, "ESM", "Trait"), 
         outcome = mapvalues(outcome, outcomes$oldName, outcomes$prettyName)) %>%
  arrange(outcome) %>%
#  select(-type) %>%
  mutate_at(vars(Contemporaneous_1:Lagged_2), funs(round(.,2))) %>%
  kable(., "html", booktabs = T, digits = 2, 
            caption = "Correlates of Measures of Ipsative Consistency",
            col.names = c("Outcome", "W1", "W2", "W1", "W2"),
            align = c("l", "r", "r", "r", "r")) %>%
  add_header_above(c("Outcome" = 1, "Contemporaneous" = 2, "Lagged" = 2))