Analisis de red
Nicolas Martínez Ramos
2024-05-25
Análisis de red - COVID 19
Se presetan los siguientes resultados acorde al análisis de red
Cargue base de datos
Se carga la base de datos, recodificando los respectivos datos perdidos, se seleccionan las variables de trabajo
revisando la informatividad de los ítems mediante estadísticas descriptivas
## Source of stress, mean= 1.96 SD= 1.17 Skew= 0.14 Kurtosis= 2.3
## Perceived stress, mean= 2.18 SD= 0.86 Skew= 0 Kurtosis= 3.35
## Resilience, mean= 3.11 SD= 0.9 Skew= -1.19 Kurtosis= 4.98
## PHQ, mean= 0.74 SD= 0.86 Skew= 1.05 Kurtosis= 3.42## vars n mean sd median trimmed mad min max range skew kurtosis se
## CE 1 148 2.60 0.99 3 2.62 1.48 1 4 3 -0.07 -1.05 0.08
## R1 2 148 3.34 0.81 4 3.48 0.00 0 4 4 -1.31 1.86 0.07
## R2 3 148 3.31 0.78 3 3.43 1.48 0 4 4 -1.45 3.27 0.06
## R3 4 148 2.74 0.97 3 2.81 1.48 0 4 4 -0.41 -0.25 0.08
## R4 5 148 2.84 1.05 3 2.96 1.48 0 4 4 -0.75 0.17 0.09
## R5 6 148 3.50 0.81 4 3.67 0.00 0 4 4 -2.12 5.43 0.07
## R6 7 148 3.42 0.84 4 3.58 0.00 0 4 4 -1.86 4.15 0.07
## R7 8 148 3.01 0.92 3 3.11 1.48 0 4 4 -1.05 1.34 0.08
## R8 9 148 2.74 1.03 3 2.82 1.48 0 4 4 -0.40 -0.53 0.08
## R9 10 148 3.23 0.90 3 3.37 1.48 0 4 4 -1.47 2.69 0.07
## R10 11 148 2.93 0.93 3 3.02 1.48 0 4 4 -0.92 0.99 0.08
## PS1 12 148 1.93 0.82 2 1.87 1.48 0 4 4 0.51 -0.09 0.07
## PS2 13 148 1.38 0.91 1 1.31 0.00 0 4 4 0.88 0.81 0.07
## PS3 14 148 2.13 0.91 2 2.12 1.48 0 4 4 0.01 -0.16 0.08
## PS4 15 148 3.04 0.77 3 3.11 0.00 0 4 4 -0.86 1.33 0.06
## PS5 16 148 2.88 0.84 3 2.93 0.74 0 4 4 -0.59 0.25 0.07
## PS6 17 148 1.57 0.93 1 1.55 1.48 0 4 4 0.48 -0.21 0.08
## PS7 18 148 2.84 0.79 3 2.87 0.00 0 4 4 -0.63 1.07 0.06
## PS8 19 148 2.67 0.80 3 2.71 0.00 0 4 4 -0.59 0.29 0.07
## PS9 20 148 1.99 0.89 2 1.96 1.48 0 4 4 0.20 -0.30 0.07
## PS10 21 148 1.37 0.92 1 1.32 1.48 0 4 4 0.56 0.08 0.08
## SS1 22 147 2.29 1.10 2 2.28 1.48 0 4 4 -0.02 -0.93 0.09
## SS2 23 146 1.90 1.06 2 1.92 1.48 0 4 4 -0.01 -0.83 0.09
## SS3 24 148 2.42 1.18 3 2.46 1.48 0 4 4 -0.23 -1.00 0.10
## SS4 25 148 2.69 1.17 3 2.78 1.48 0 4 4 -0.51 -0.73 0.10
## SS5 26 148 1.71 1.17 2 1.66 1.48 0 4 4 0.30 -0.81 0.10
## SS6 27 147 1.86 1.26 2 1.82 1.48 0 4 4 0.14 -1.06 0.10
## SS7 28 148 0.85 1.28 0 0.62 0.00 0 4 4 1.26 0.27 0.11
## PHQ1 29 148 0.91 0.89 1 0.80 1.48 0 3 3 0.81 -0.03 0.07
## PHQ2 30 148 0.67 0.90 0 0.51 0.00 0 3 3 1.25 0.65 0.07
## PHQ3 31 148 0.68 0.84 0 0.55 0.00 0 3 3 1.12 0.56 0.07
## PHQ4 32 148 0.72 0.83 1 0.60 1.48 0 3 3 0.99 0.33 0.07
## MD 33 148 3.67 2.56 3 3.54 2.97 0 10 10 0.36 -0.89 0.21
## Dep 34 147 1.94 0.24 2 2.00 0.00 1 2 1 -3.62 11.20 0.02
## Covid 35 148 1.78 0.42 2 1.84 0.00 1 2 1 -1.32 -0.27 0.03Todos los ítems apotan información siguiendo los mínimos propuestos en la literatura.
Graficación de la red
library(ggplot2)
library(bootnet)
library(networktools)
RED3 <- RED3 %>%
mutate(
# Recodificación de ítems invertidos (escala 0-4)
PS4_r = 4 - PS4,
PS5_r = 4 - PS5,
PS7_r = 4 - PS7,
PS8_r = 4 - PS8,
# Sumas
R = rowSums(select(., R1:R10), na.rm = TRUE),
PHQA = rowSums(cbind(PHQ1, PHQ2), na.rm = TRUE),
PHQD = rowSums(cbind(PHQ3, PHQ4), na.rm = TRUE),
PS = rowSums(cbind(PS1, PS2, PS3, PS4_r, PS5_r, PS6, PS7_r, PS8_r, PS9, PS10), na.rm = TRUE),
PHQT = rowSums(cbind(PHQA, PHQD), na.rm = TRUE),
)
RED3 <- RED3 %>%
mutate(
Dd = ifelse(Dep== 2, 0, Dep)
)
RED4 <- RED3 %>% select(R, PS, PHQA, PHQD, Dd)
library(bootnet)
library(qgraph)
library(networktools)
Network2 <- estimateNetwork(
RED4,
default = "EBICglasso",
weighted = TRUE,
corMethod = "spearman" #threshold = TRUE
)
plot(Network2, label.cex=.60, label.scale=F,vsize = 6, esize = 15, details = F, palette="pastel", vTrans = 230, usePCH=FALSE, edge.width = 1, loop = 1, node.width = 0.8, border.width = 2, legend.mode = "groups", legend.cex = 0.5,
GLratio = 5)
describe(RED3)## vars n mean sd median trimmed mad min max range skew kurtosis se
## CE 1 148 2.60 0.99 3 2.62 1.48 1 4 3 -0.07 -1.05 0.08
## R1 2 148 3.34 0.81 4 3.48 0.00 0 4 4 -1.31 1.86 0.07
## R2 3 148 3.31 0.78 3 3.43 1.48 0 4 4 -1.45 3.27 0.06
## R3 4 148 2.74 0.97 3 2.81 1.48 0 4 4 -0.41 -0.25 0.08
## R4 5 148 2.84 1.05 3 2.96 1.48 0 4 4 -0.75 0.17 0.09
## R5 6 148 3.50 0.81 4 3.67 0.00 0 4 4 -2.12 5.43 0.07
## R6 7 148 3.42 0.84 4 3.58 0.00 0 4 4 -1.86 4.15 0.07
## R7 8 148 3.01 0.92 3 3.11 1.48 0 4 4 -1.05 1.34 0.08
## R8 9 148 2.74 1.03 3 2.82 1.48 0 4 4 -0.40 -0.53 0.08
## R9 10 148 3.23 0.90 3 3.37 1.48 0 4 4 -1.47 2.69 0.07
## R10 11 148 2.93 0.93 3 3.02 1.48 0 4 4 -0.92 0.99 0.08
## PS1 12 148 1.93 0.82 2 1.87 1.48 0 4 4 0.51 -0.09 0.07
## PS2 13 148 1.38 0.91 1 1.31 0.00 0 4 4 0.88 0.81 0.07
## PS3 14 148 2.13 0.91 2 2.12 1.48 0 4 4 0.01 -0.16 0.08
## PS4 15 148 3.04 0.77 3 3.11 0.00 0 4 4 -0.86 1.33 0.06
## PS5 16 148 2.88 0.84 3 2.93 0.74 0 4 4 -0.59 0.25 0.07
## PS6 17 148 1.57 0.93 1 1.55 1.48 0 4 4 0.48 -0.21 0.08
## PS7 18 148 2.84 0.79 3 2.87 0.00 0 4 4 -0.63 1.07 0.06
## PS8 19 148 2.67 0.80 3 2.71 0.00 0 4 4 -0.59 0.29 0.07
## PS9 20 148 1.99 0.89 2 1.96 1.48 0 4 4 0.20 -0.30 0.07
## PS10 21 148 1.37 0.92 1 1.32 1.48 0 4 4 0.56 0.08 0.08
## SS1 22 147 2.29 1.10 2 2.28 1.48 0 4 4 -0.02 -0.93 0.09
## SS2 23 146 1.90 1.06 2 1.92 1.48 0 4 4 -0.01 -0.83 0.09
## SS3 24 148 2.42 1.18 3 2.46 1.48 0 4 4 -0.23 -1.00 0.10
## SS4 25 148 2.69 1.17 3 2.78 1.48 0 4 4 -0.51 -0.73 0.10
## SS5 26 148 1.71 1.17 2 1.66 1.48 0 4 4 0.30 -0.81 0.10
## SS6 27 147 1.86 1.26 2 1.82 1.48 0 4 4 0.14 -1.06 0.10
## SS7 28 148 0.85 1.28 0 0.62 0.00 0 4 4 1.26 0.27 0.11
## PHQ1 29 148 0.91 0.89 1 0.80 1.48 0 3 3 0.81 -0.03 0.07
## PHQ2 30 148 0.67 0.90 0 0.51 0.00 0 3 3 1.25 0.65 0.07
## PHQ3 31 148 0.68 0.84 0 0.55 0.00 0 3 3 1.12 0.56 0.07
## PHQ4 32 148 0.72 0.83 1 0.60 1.48 0 3 3 0.99 0.33 0.07
## MD 33 148 3.67 2.56 3 3.54 2.97 0 10 10 0.36 -0.89 0.21
## Dep 34 147 1.94 0.24 2 2.00 0.00 1 2 1 -3.62 11.20 0.02
## Covid 35 148 1.78 0.42 2 1.84 0.00 1 2 1 -1.32 -0.27 0.03
## PS4_r 36 148 0.96 0.77 1 0.89 0.00 0 4 4 0.86 1.33 0.06
## PS5_r 37 148 1.12 0.84 1 1.07 0.74 0 4 4 0.59 0.25 0.07
## PS7_r 38 148 1.16 0.79 1 1.13 0.00 0 4 4 0.63 1.07 0.06
## PS8_r 39 148 1.33 0.80 1 1.29 0.00 0 4 4 0.59 0.29 0.07
## R 40 148 31.07 6.75 32 31.78 5.93 0 40 40 -1.51 3.96 0.55
## PHQA 41 148 1.58 1.65 1 1.34 1.48 0 6 6 1.10 0.53 0.14
## PHQD 42 148 1.40 1.56 1 1.17 1.48 0 6 6 1.09 0.61 0.13
## PS 43 148 14.93 5.84 14 14.67 5.93 3 34 31 0.49 0.16 0.48
## PHQT 44 148 2.98 2.99 2 2.54 2.97 0 12 12 1.15 0.69 0.25
## Dd 45 147 0.06 0.24 0 0.00 0.00 0 1 1 3.62 11.20 0.02lapply(RED3[,40:45], shapiro.test)## $R
##
## Shapiro-Wilk normality test
##
## data: X[[i]]
## W = 0.88843, p-value = 3.702e-09
##
##
## $PHQA
##
## Shapiro-Wilk normality test
##
## data: X[[i]]
## W = 0.83832, p-value = 1.773e-11
##
##
## $PHQD
##
## Shapiro-Wilk normality test
##
## data: X[[i]]
## W = 0.82042, p-value = 3.446e-12
##
##
## $PS
##
## Shapiro-Wilk normality test
##
## data: X[[i]]
## W = 0.98059, p-value = 0.03445
##
##
## $PHQT
##
## Shapiro-Wilk normality test
##
## data: X[[i]]
## W = 0.862, p-value = 1.886e-10
##
##
## $Dd
##
## Shapiro-Wilk normality test
##
## data: X[[i]]
## W = 0.2551, p-value < 2.2e-16Red clusterizada
Al codificar los datos pérdidos correctamente, la distribución de la red cambió drásticamente.
Groups <- c("R","PS", rep("PHQ",2), "Dd")
plot(Network2,
groups = Groups, palette = "pastel",
filetype = "png",
filename = "network_plot",
width = 10,
height = 10,
res = 300,
label.cex = 1.5,
label.scale = FALSE,
vsize = 8,
esize = 17,
edge.width = 1.5,
node.width = 1,
posCol = "#228B22",
negCol = "#FF0000",
vTrans = 200,
border.width = 2.5,
usePCH = FALSE,
legend.mode = "groups",
legend.cex = 1.5,
GLratio = 6,
loop = 1,
details = FALSE,
layout = "spring")plot(Network2,
groups = Groups, palette = "pastel",
label.cex = 1.5,
label.scale = FALSE,
vsize = 8,
esize = 17,
edge.width = 1.5,
node.width = 1,
posCol = "#228B22",
negCol = "#FF0000",
vTrans = 200,
border.width = 2.5,
usePCH = FALSE,
legend.mode = "groups",
legend.cex = 1.5,
GLratio = 6,
loop = 1,
details = FALSE,
layout = "spring")
Estimando medidas de precisión - centralidad
library("qgraph")
p<-centralityPlot(Network2, scale = c("z-scores"),
include = c("Strength", "Closeness", "Betweenness", "ExpectedInfluence"), theme_bw = TRUE, print = TRUE, verbose = TRUE, weighted = TRUE, decreasing = T)
ggplot2::ggsave(
"centrality_plot.png",
plot = p,
width = 10,
height = 8
)
p
Centralidad
Estabilidad de la red - Medidas de precisión
- Estabilidad de la red
- ICs de las ponderaciones
- Diferencias entre los pesos de los nodos
Estas mediciones pemiten saber si las diferencias encontradas en el análisis de las medidas de centralidad realmente son comparables.
Estabilidad de la red
set.seed(2026)
CentralStability <- bootnet(Network2, nCores = 4, nBoots = 1000, type = "case", statistics = c("strength", "closeness", "betweenness", "expectedInfluence"))
corStability(CentralStability)## === Correlation Stability Analysis ===
##
## Sampling levels tested:
## nPerson Drop% n
## 1 37 75.0 100
## 2 49 66.9 84
## 3 60 59.5 95
## 4 72 51.4 110
## 5 83 43.9 100
## 6 95 35.8 108
## 7 106 28.4 104
## 8 118 20.3 109
## 9 129 12.8 91
## 10 141 4.7 99
##
## Maximum drop proportions to retain correlation of 0.7 in at least 95% of the samples:
##
## betweenness: 0.358
## - For more accuracy, run bootnet(..., caseMin = 0.284, caseMax = 0.439)
##
## closeness: 0.669
## - For more accuracy, run bootnet(..., caseMin = 0.595, caseMax = 0.75)
##
## expectedInfluence: 0.669
## - For more accuracy, run bootnet(..., caseMin = 0.595, caseMax = 0.75)
##
## strength: 0.669
## - For more accuracy, run bootnet(..., caseMin = 0.595, caseMax = 0.75)
##
## Accuracy can also be increased by increasing both 'nBoots' and 'caseN'.plot(CentralStability, statistics= c("strength","closeness","betweenness", "expectedInfluence"))
dev.new()
plot(CentralStability,
statistics = c("strength", "closeness", "betweenness", "expectedInfluence"))
dev.copy(png, "central_stability.png",width = 3000, height = 2400, res = 500)## png
## 4dev.off()## png
## 2La estabilidad de la red es muy baja, el punto de corte de la correlación debe ser 0.70, después del remuestreo del 85% ninguna medida consigue ese umbral. Se puede deber al pequeño tamaño muestral.
CI pesos o ponderaciones entre nodos
En este caso como se aplicó un método de penalización se presentan las ponderaciones o pesos de las relaciones entre los nodos
EdgeWgt<- bootnet(Network2, nBoots = 1000, nCores = 4)
B <- summary(EdgeWgt)
library(openxlsx)
write.xlsx(B, "EdgeWeights.xlsx")
library(kableExtra)
kable(B, "html") %>%
kable_styling(full_width = FALSE, bootstrap_options = c("striped", "hover", "condensed"))| type | id | node1 | node2 | sample | mean | sd | CIlower | CIupper | q2.5 | q97.5 | q2.5_non0 | mean_non0 | q97.5_non0 | var_non0 | sd_non0 | prop0 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| edge | PHQA–Dd | PHQA | Dd | 0.0000000 | 0.0083929 | 0.0262756 | -0.0525511 | 0.0525511 | 0.0000000 | 0.0721815 | -0.1214327 | 0.0283543 | 0.1003014 | 0.0017701 | 0.0420726 | 0.704 |
| edge | PHQA–PHQD | PHQA | PHQD | 0.3586551 | 0.3544301 | 0.0796176 | 0.1994200 | 0.5178903 | 0.2055575 | 0.5074810 | 0.2055575 | 0.3544301 | 0.5074810 | 0.0063390 | 0.0796176 | 0.000 |
| edge | PHQD–Dd | PHQD | Dd | 0.1871931 | 0.1797134 | 0.0509468 | 0.0852995 | 0.2890868 | 0.0810349 | 0.2782474 | 0.0811724 | 0.1798933 | 0.2783064 | 0.0025658 | 0.0506536 | 0.001 |
| edge | PS–Dd | PS | Dd | 0.0572379 | 0.0670055 | 0.0543432 | -0.0514484 | 0.1659242 | 0.0000000 | 0.1954095 | 0.0073796 | 0.0793903 | 0.2020731 | 0.0025153 | 0.0501524 | 0.156 |
| edge | PS–PHQA | PS | PHQA | 0.3078669 | 0.3089989 | 0.0634432 | 0.1809806 | 0.4347532 | 0.1776779 | 0.4325152 | 0.1776779 | 0.3089989 | 0.4325152 | 0.0040250 | 0.0634432 | 0.000 |
| edge | PS–PHQD | PS | PHQD | 0.3662954 | 0.3609275 | 0.0744050 | 0.2174854 | 0.5151054 | 0.2128951 | 0.4989610 | 0.2128951 | 0.3609275 | 0.4989610 | 0.0055361 | 0.0744050 | 0.000 |
| edge | R–Dd | R | Dd | 0.0000000 | 0.0222557 | 0.0671108 | -0.1342216 | 0.1342216 | -0.0712516 | 0.1983828 | -0.1039240 | 0.0656511 | 0.2586502 | 0.0104543 | 0.1022464 | 0.661 |
| edge | R–PHQA | R | PHQA | -0.1243392 | -0.1172044 | 0.0737969 | -0.2719330 | 0.0232545 | -0.2618743 | 0.0000000 | -0.2633905 | -0.1292220 | -0.0176121 | 0.0044503 | 0.0667109 | 0.093 |
| edge | R–PHQD | R | PHQD | -0.0297674 | -0.0466931 | 0.0511711 | -0.1321095 | 0.0725747 | -0.1664137 | 0.0000000 | -0.1773554 | -0.0692776 | -0.0032687 | 0.0023199 | 0.0481657 | 0.326 |
| edge | R–PS | R | PS | -0.3769716 | -0.3746776 | 0.0646336 | -0.5062388 | -0.2477045 | -0.4929840 | -0.2433474 | -0.4929840 | -0.3746776 | -0.2433474 | 0.0041775 | 0.0646336 | 0.000 |
| strength | Dd | Dd | 0.2444310 | 0.2940443 | 0.1152654 | 0.0139003 | 0.4749618 | 0.1403310 | 0.6127137 | 0.1413055 | 0.2943387 | 0.6127472 | 0.0132127 | 0.1149465 | 0.001 | |
| strength | PHQA | PHQA | 0.7908612 | 0.7929025 | 0.0814560 | 0.6279492 | 0.9537732 | 0.6485684 | 0.9721893 | 0.6485684 | 0.7929025 | 0.9721893 | 0.0066351 | 0.0814560 | 0.000 | |
| strength | PHQD | PHQD | 0.9419110 | 0.9424642 | 0.0835964 | 0.7747182 | 1.1091039 | 0.7768520 | 1.1030018 | 0.7768520 | 0.9424642 | 1.1030018 | 0.0069884 | 0.0835964 | 0.000 | |
| strength | PS | PS | 1.1083718 | 1.1117597 | 0.0911248 | 0.9261221 | 1.2906215 | 0.9432889 | 1.2942867 | 0.9432889 | 1.1117597 | 1.2942867 | 0.0083037 | 0.0911248 | 0.000 | |
| strength | R | R | 0.5310783 | 0.5741813 | 0.1119661 | 0.3071460 | 0.7550105 | 0.3901286 | 0.8466610 | 0.3901286 | 0.5741813 | 0.8466610 | 0.0125364 | 0.1119661 | 0.000 |
EdgeWgt2 <- bootnet(Network2,
nBoots = 1000,
nCores = 4,
statistics = c("strength", "expectedInfluence",
"closeness", "betweenness"))
C <- summary(EdgeWgt)
kable(C, "html") %>%
kable_styling(full_width = FALSE, bootstrap_options = c("striped", "hover", "condensed"))| type | id | node1 | node2 | sample | mean | sd | CIlower | CIupper | q2.5 | q97.5 | q2.5_non0 | mean_non0 | q97.5_non0 | var_non0 | sd_non0 | prop0 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| edge | PHQA–Dd | PHQA | Dd | 0.0000000 | 0.0083929 | 0.0262756 | -0.0525511 | 0.0525511 | 0.0000000 | 0.0721815 | -0.1214327 | 0.0283543 | 0.1003014 | 0.0017701 | 0.0420726 | 0.704 |
| edge | PHQA–PHQD | PHQA | PHQD | 0.3586551 | 0.3544301 | 0.0796176 | 0.1994200 | 0.5178903 | 0.2055575 | 0.5074810 | 0.2055575 | 0.3544301 | 0.5074810 | 0.0063390 | 0.0796176 | 0.000 |
| edge | PHQD–Dd | PHQD | Dd | 0.1871931 | 0.1797134 | 0.0509468 | 0.0852995 | 0.2890868 | 0.0810349 | 0.2782474 | 0.0811724 | 0.1798933 | 0.2783064 | 0.0025658 | 0.0506536 | 0.001 |
| edge | PS–Dd | PS | Dd | 0.0572379 | 0.0670055 | 0.0543432 | -0.0514484 | 0.1659242 | 0.0000000 | 0.1954095 | 0.0073796 | 0.0793903 | 0.2020731 | 0.0025153 | 0.0501524 | 0.156 |
| edge | PS–PHQA | PS | PHQA | 0.3078669 | 0.3089989 | 0.0634432 | 0.1809806 | 0.4347532 | 0.1776779 | 0.4325152 | 0.1776779 | 0.3089989 | 0.4325152 | 0.0040250 | 0.0634432 | 0.000 |
| edge | PS–PHQD | PS | PHQD | 0.3662954 | 0.3609275 | 0.0744050 | 0.2174854 | 0.5151054 | 0.2128951 | 0.4989610 | 0.2128951 | 0.3609275 | 0.4989610 | 0.0055361 | 0.0744050 | 0.000 |
| edge | R–Dd | R | Dd | 0.0000000 | 0.0222557 | 0.0671108 | -0.1342216 | 0.1342216 | -0.0712516 | 0.1983828 | -0.1039240 | 0.0656511 | 0.2586502 | 0.0104543 | 0.1022464 | 0.661 |
| edge | R–PHQA | R | PHQA | -0.1243392 | -0.1172044 | 0.0737969 | -0.2719330 | 0.0232545 | -0.2618743 | 0.0000000 | -0.2633905 | -0.1292220 | -0.0176121 | 0.0044503 | 0.0667109 | 0.093 |
| edge | R–PHQD | R | PHQD | -0.0297674 | -0.0466931 | 0.0511711 | -0.1321095 | 0.0725747 | -0.1664137 | 0.0000000 | -0.1773554 | -0.0692776 | -0.0032687 | 0.0023199 | 0.0481657 | 0.326 |
| edge | R–PS | R | PS | -0.3769716 | -0.3746776 | 0.0646336 | -0.5062388 | -0.2477045 | -0.4929840 | -0.2433474 | -0.4929840 | -0.3746776 | -0.2433474 | 0.0041775 | 0.0646336 | 0.000 |
| strength | Dd | Dd | 0.2444310 | 0.2940443 | 0.1152654 | 0.0139003 | 0.4749618 | 0.1403310 | 0.6127137 | 0.1413055 | 0.2943387 | 0.6127472 | 0.0132127 | 0.1149465 | 0.001 | |
| strength | PHQA | PHQA | 0.7908612 | 0.7929025 | 0.0814560 | 0.6279492 | 0.9537732 | 0.6485684 | 0.9721893 | 0.6485684 | 0.7929025 | 0.9721893 | 0.0066351 | 0.0814560 | 0.000 | |
| strength | PHQD | PHQD | 0.9419110 | 0.9424642 | 0.0835964 | 0.7747182 | 1.1091039 | 0.7768520 | 1.1030018 | 0.7768520 | 0.9424642 | 1.1030018 | 0.0069884 | 0.0835964 | 0.000 | |
| strength | PS | PS | 1.1083718 | 1.1117597 | 0.0911248 | 0.9261221 | 1.2906215 | 0.9432889 | 1.2942867 | 0.9432889 | 1.1117597 | 1.2942867 | 0.0083037 | 0.0911248 | 0.000 | |
| strength | R | R | 0.5310783 | 0.5741813 | 0.1119661 | 0.3071460 | 0.7550105 | 0.3901286 | 0.8466610 | 0.3901286 | 0.5741813 | 0.8466610 | 0.0125364 | 0.1119661 | 0.000 |
Grafico ICs
plot(EdgeWgt2, "expectedInfluence")
plot(EdgeWgt, labels = T, order = "sample", bootlwd = 0.5, samplelwd = 3, CIstyle = "quantiles", meanColor = "#FA8072")
dev.new()
plot(EdgeWgt, labels = T, order = "sample", bootlwd = 0.5, samplelwd = 3, CIstyle = "quantiles", meanColor = "#FA8072")
dev.copy(png, "bootstrap.png",width = 3000, height = 2400, res = 500)## png
## 5dev.off()## png
## 2El anterior gráfico muestra la media de las ponderaciones de la muestra original y una remuestreada. Acorde al gráfico y a los resultados anteriores, ninguno de los intervalos de confianza de las ponderaciones incluye el cero, se puede tener la seguridad de las ponderaciones son estadísticamente significativas entre sí, no necesariamente de la correlación como sucedía con pruebas anteriores.
Detectando diferencias significativas entre nodos y fuerza
Fuerza
plot(EdgeWgt, "strength")
Se encontraron diferencias significativas en muchos de los nodos.
Diferencia entre los nodos
dev.new()
plot(EdgeWgt, "strength")
dev.copy(png, "strength.png",width = 3000, height = 2400, res = 500)## png
## 6dev.off()## png
## 2plot(EdgeWgt, "edge", plot = "difference", onlyNonZero = TRUE, order = "sample")