Inicio
Probamos nuevos análisis en función de las sugerencias de Miquel, donde indica que no le queda claro por qué usamos distintas metodologías para la parte psicométrica y las relaciones entre variables. En los comentarios él se pregunta por qué no usar ESEM (modelo de ecuaciones estructurales exploratorio) para integrar los análisis. Tiene sentido probar las relaciones en un SEM, aunque no sé por qué lo haríamos de forma exploratoria, cuando todos los instrumentos están validados con una estructura particular y que funciona (el único que podría pensarse es el de id. social, pero el único cambio es el idioma). Sería como hacer un factorial confirmatorio o uno exploratorio (con modelos bastante más complejos).
Otro inconveniente que surge al buscar evaluar las moderaciones con SEM es que la inclusión de la interacción de variables reduce ampliamente los grados de libertad de los modelos. En las regresiones tomamos las medias de cada variable y las multiplicamos, generando una única variable nueva. En los modelos SEM, las interacciones se incluyen multiplicando ítem a ítem, haciendo que sean muchas más variables. Por ejemplo, la interacción entre sexo y liderazgo del coach en una regresión lineal implicaría sumar una única variable (sexo*Liderazgo), mientras que en SEM se sumarían 16 variables (sexo*LT01, sexo*LT02, …).
Hasta ahí, no hay tanto problema con los grados de libertad. El problema es que esas nuevas variables (las de la interacción) no son independientes de las variables originales (los ítems), entonces hay que incluir en el modelo esa falta de independencia, incluyendo la estimación de las covarianzas de error. Ahí es donde se disparan los parámetros a estimar y, por lo tanto, el modelo termina con grados de libertad negativos. Es decir, con la estimación de tantos parámetros el modelo no está identificado (no se puede ajustar).
Dada esta problemática, propongo tres soluciones:
- SEM-MG
- SEM y Reg. Lineal
- SEM con moderación
SEM-MG
Una posibilidad es ajustar modelos de ecuaciones estructurales multigrupos (SEM-MG). La aproximación es la misma que la utilizada para estimar la invarianza de un instrumento, solo que en vez de incluir solo el instrumento de identidad social, se incluiría el modelo completo (con las relaciones con liderazgo). A diferencia del estudio de invarianza, en este caso buscaríamos que haya diferencias significativas entre modelos (que no sea lo mismo en varones que en mujeres, por ejemplo).
A partir de este análisis se podrían obtener estimaciones para cada parámetro en en caso de varones/mujeres o early/late adolescents y presentar las figuras en cada caso.
La desventaja de este abordaje es que no se podrían incluir ambas interacciones (sexo y edad) en simultaneo. PENSAR SI PODRÍA CREARSE UNA NUEVA VARIABLE SEXO*EDAD, CON CUATRO GRUPOS (varon-joven, varon-tardio, mujer-joven, mujer-tardia)
Coach
Sexo
modelo <- '
# Identidad social
it =~ siqs01 + siqs02 + siqs03
cc =~ siqs04 + siqs05 + siqs06
ia =~ siqs07 + siqs08 + siqs09
SIQS =~ it + cc + ia
# Liderazgo transformacional (coach)
LTc =~ tcq02 + tcq05 + tcq12 + tcq16 + tcq01 + tcq09 + tcq11 + tcq14 + tcq03 + tcq07 + tcq10 + tcq13 + tcq04 + tcq06 + tcq08 + tcq15
# REGRESIONES
SIQS ~ LTc + edad
'
# Configural model
sem.config <- datos %>%
sem(modelo, .,
estimator="ULS",
group = "sexo")
#sem.config %>%
# summary(fit.measures = TRUE, standardized = TRUE)
# Metric model
sem.metric <- datos %>%
sem(modelo, .,
estimator="ULS",
group = "sexo",
group.equal = "loadings")
# Scalar model
sem.scalar <- datos %>%
sem(modelo, .,
estimator="ULS",
group = "sexo",
group.equal = c("loadings",
"intercepts"))
# Strict model
sem.strict <- datos %>%
sem(modelo, .,
estimator="ULS",
group = "sexo",
group.equal = c("loadings",
"intercepts",
"residuals"))
# Model comparison
semTools::compareFit(sem.config,
sem.metric,
sem.scalar,
sem.strict) %>%
summary()## ################### Nested Model Comparison #########################
##
## Chi-Squared Difference Test
##
## Df AIC BIC Chisq Chisq diff RMSEA Df diff Pr(>Chisq)
## sem.config 590 2347.6
## sem.metric 613 2729.4 381.81 0.218088 23 < 2e-16 ***
## sem.scalar 633 2763.3 33.90 0.046039 20 0.02678 *
## sem.strict 658 2925.0 161.62 0.129077 25 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## ####################### Model Fit Indices ###########################
## chisq df pvalue rmsea cfi tli srmr
## sem.config 2347.627† 590 NA .095† .975† .972† .057†
## sem.metric 2729.438 613 NA .103 .969 .968 .063
## sem.scalar 2763.342 633 NA .101 .969 .968 .064
## sem.strict 2924.960 658 NA .103 .967 .968 .067
##
## ################## Differences in Fit Indices #######################
## df rmsea cfi tli srmr
## sem.metric - sem.config 23 0.007 -0.005 -0.004 0.006
## sem.scalar - sem.metric 20 -0.001 0.000 0.001 0.001
## sem.strict - sem.scalar 25 0.001 -0.002 -0.001 0.003El modelo métrico ya difiere significativamente del modelo configural, por lo que no se verifica la invariaza en el modelo completo para el entrenador.
Estimaciones para el modelo configural (el que asume menos igualdades entre los grupos)
semPaths(# Argumentos globales
sem.config, what="diagram", whatLabels="std",layout="tree3", rotation = 2, width=50, height=35,exoVar = F,
# Etiquetas
curvePivot=T, curvature = .5
)
Nota. 1=Varones; 2=Mujeres.
# [Grupo 0] SIQS = 0 + .358 * LTc
# [Grupo 1] SIQS = 0 + .567 * LTc
graf <- data.frame(sexo=rep(0:1,2),
"LTc (std)"=c(rep(min(scale(datos$tcq_LT),na.rm = T),2),rep(max(scale(datos$tcq_LT),na.rm = T),2)),
"siqs (std)"=rep(NA,4),
check.names = F)
for(i in 1:4){
graf$`siqs (std)`[i] <- ifelse(graf$sexo[i]==0,
.358*graf$`LTc (std)`[i],
.567*graf$`LTc (std)`[i])
}
graf %>%
ggplot(aes(x=`LTc (std)`, y=`siqs (std)`, color=as.factor(sexo), group=as.factor(sexo)))+
geom_line()+
scale_color_manual(name="Sex",
values = 2:3,
labels = c("females",
"males"))
Edad
modelo <- '
# Identidad social
it =~ siqs01 + siqs02 + siqs03
cc =~ siqs04 + siqs05 + siqs06
ia =~ siqs07 + siqs08 + siqs09
SIQS =~ it + cc + ia
# Liderazgo transformacional (coach)
LTc =~ tcq02 + tcq05 + tcq12 + tcq16 + tcq01 + tcq09 + tcq11 + tcq14 + tcq03 + tcq07 + tcq10 + tcq13 + tcq04 + tcq06 + tcq08 + tcq15
# REGRESIONES
SIQS ~ LTc + sexo
'
# Configural model
sem.config <- datos %>%
sem(modelo, .,
estimator="ULS",
group = "edaddicot")
#sem.config %>%
# summary(fit.measures = TRUE, standardized = TRUE)
# Metric model
sem.metric <- datos %>%
sem(modelo, .,
estimator="ULS",
group = "edaddicot",
group.equal = "loadings")
# Scalar model
sem.scalar <- datos %>%
sem(modelo, .,
estimator="ULS",
group = "edaddicot",
group.equal = c("loadings",
"intercepts"))
# Strict model
sem.strict <- datos %>%
sem(modelo, .,
estimator="ULS",
group = "edaddicot",
group.equal = c("loadings",
"intercepts",
"residuals"))
# Model comparison
semTools::compareFit(sem.config,
sem.metric,
sem.scalar,
sem.strict) %>%
summary()## ################### Nested Model Comparison #########################
##
## Chi-Squared Difference Test
##
## Df AIC BIC Chisq Chisq diff RMSEA Df diff Pr(>Chisq)
## sem.config 590 1428.5
## sem.metric 613 1794.4 365.98 0.213222 23 < 2.2e-16 ***
## sem.scalar 633 1837.4 42.93 0.059124 20 0.002087 **
## sem.strict 658 2031.1 193.75 0.143454 25 < 2.2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## ####################### Model Fit Indices ###########################
## chisq df pvalue rmsea cfi tli srmr
## sem.config 1428.454† 590 NA .066† .987† .986† .055†
## sem.metric 1794.430 613 NA .077 .982 .981 .061
## sem.scalar 1837.362 633 NA .076 .982 .981 .061
## sem.strict 2031.110 658 NA .080 .979 .980 .064
##
## ################## Differences in Fit Indices #######################
## df rmsea cfi tli srmr
## sem.metric - sem.config 23 0.011 -0.005 -0.005 0.005
## sem.scalar - sem.metric 20 0.000 0.000 0.000 0.001
## sem.strict - sem.scalar 25 0.004 -0.003 -0.002 0.003El modelo métrico ya difiere significativamente del modelo configural, por lo que no se verifica la invariaza en el modelo completo para el entrenador.
Estimaciones para el modelo configural (el que asume menos igualdades entre los grupos)
semPaths(# Argumentos globales
sem.config, what="diagram", whatLabels="std",layout="tree3", rotation = 2, width=50, height=35,exoVar = F,
# Etiquetas
curvePivot=T, curvature = .5
)
Nota. 1=Late; 2=Early.
# [Grupo 1] SIQS = 0 + .481 * LTc
# [Grupo 2] SIQS = 0 + .381 * LTc
graf <- data.frame(edad=rep(1:2,2),
"LTc (std)"=c(rep(min(scale(datos$tcq_LT),na.rm = T),2),rep(max(scale(datos$tcq_LT),na.rm = T),2)),
"siqs (std)"=rep(NA,4),
check.names = F)
for(i in 1:4){
graf$`siqs (std)`[i] <- ifelse(graf$edad[i]==1,
.481*graf$`LTc (std)`[i],
.381*graf$`LTc (std)`[i])
}
graf %>%
ggplot(aes(x=`LTc (std)`, y=`siqs (std)`, color=as.factor(edad), group=as.factor(edad)))+
geom_line()+
scale_color_manual(name="Age stage",
values = 2:3,
labels = c("early",
"late"))
Madre
Sexo
modelo <- '
# Identidad social
it =~ siqs01 + siqs02 + siqs03
cc =~ siqs04 + siqs05 + siqs06
ia =~ siqs07 + siqs08 + siqs09
SIQS =~ it + cc + ia
# Liderazgo transformacional (coach)
LTm =~ tpqm02 + tpqm05 + tpqm12 + tpqm16 + tpqm01 + tpqm09 + tpqm11 + tpqm14 + tpqm03 + tpqm07 + tpqm10 + tpqm13 + tpqm04 + tpqm06 + tpqm08 + tpqm15
# REGRESIONES
SIQS ~ LTm + edad
'
# Configural model
sem.config <- datos %>%
sem(modelo, .,
estimator="ULS",
group = "sexo")
#sem.config %>%
# summary(fit.measures = TRUE, standardized = TRUE)
# Metric model
sem.metric <- datos %>%
sem(modelo, .,
estimator="ULS",
group = "sexo",
group.equal = "loadings")
# Scalar model
sem.scalar <- datos %>%
sem(modelo, .,
estimator="ULS",
group = "sexo",
group.equal = c("loadings",
"intercepts"))
# Strict model
sem.strict <- datos %>%
sem(modelo, .,
estimator="ULS",
group = "sexo",
group.equal = c("loadings",
"intercepts",
"residuals"))
# Model comparison
semTools::compareFit(sem.config,
sem.metric,
sem.scalar,
sem.strict) %>%
summary()## ################### Nested Model Comparison #########################
##
## Chi-Squared Difference Test
##
## Df AIC BIC Chisq Chisq diff RMSEA Df diff Pr(>Chisq)
## sem.config 590 705.61
## sem.metric 613 1110.81 405.19 0.225082 23 <2e-16 ***
## sem.scalar 633 1131.03 20.22 0.005758 20 0.4444
## sem.strict 658 1294.55 163.53 0.129976 25 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## ####################### Model Fit Indices ###########################
## chisq df pvalue rmsea cfi tli srmr
## sem.config 705.614† 590 NA .024† 0.997† 0.996† .049†
## sem.metric 1110.808 613 NA .050 .986 .985 .066
## sem.scalar 1131.026 633 NA .049 .986 .986 .066
## sem.strict 1294.554 658 NA .054 .982 .982 .072
##
## ################## Differences in Fit Indices #######################
## df rmsea cfi tli srmr
## sem.metric - sem.config 23 0.025 -0.011 -0.011 0.017
## sem.scalar - sem.metric 20 -0.001 0.000 0.000 0.001
## sem.strict - sem.scalar 25 0.005 -0.004 -0.003 0.006El modelo métrico ya difiere significativamente del modelo configural, por lo que no se verifica la invariaza en el modelo completo para la madre
Estimaciones para el modelo configural (el que asume menos igualdades entre los grupos)
semPaths(# Argumentos globales
sem.config, what="diagram", whatLabels="std",layout="tree3", rotation = 2, width=50, height=35,exoVar = F,
# Etiquetas
curvePivot=T, curvature = .5
)
Nota. 1=Varones; 2=Mujeres.
# [Grupo 0] SIQS = 0 + .4 * LTc
# [Grupo 1] SIQS = 0 + .29 * LTc
graf <- data.frame(sexo=rep(0:1,2),
"LTm (std)"=c(rep(min(scale(datos$tpqm_LT),na.rm = T),2),rep(max(scale(datos$tpqm_LT),na.rm = T),2)),
"siqs (std)"=rep(NA,4),
check.names = F)
for(i in 1:4){
graf$`siqs (std)`[i] <- ifelse(graf$sexo[i]==0,
.4*graf$`LTm (std)`[i],
.29*graf$`LTm (std)`[i])
}
graf %>%
ggplot(aes(x=`LTm (std)`, y=`siqs (std)`, color=as.factor(sexo), group=as.factor(sexo)))+
geom_line()+
scale_color_manual(name="Sex",
values = 2:3,
labels = c("females",
"males"))
Edad
modelo <- '
# Identidad social
it =~ siqs01 + siqs02 + siqs03
cc =~ siqs04 + siqs05 + siqs06
ia =~ siqs07 + siqs08 + siqs09
SIQS =~ it + cc + ia
# Liderazgo transformacional (coach)
LTm =~ tpqm02 + tpqm05 + tpqm12 + tpqm16 + tpqm01 + tpqm09 + tpqm11 + tpqm14 + tpqm03 + tpqm07 + tpqm10 + tpqm13 + tpqm04 + tpqm06 + tpqm08 + tpqm15
# REGRESIONES
SIQS ~ LTm + sexo
'
# Configural model
sem.config <- datos %>%
sem(modelo, .,
estimator="ULS",
group = "edaddicot")
#sem.config %>%
# summary(fit.measures = TRUE, standardized = TRUE)
# Metric model
sem.metric <- datos %>%
sem(modelo, .,
estimator="ULS",
group = "edaddicot",
group.equal = "loadings")
# Scalar model
sem.scalar <- datos %>%
sem(modelo, .,
estimator="ULS",
group = "edaddicot",
group.equal = c("loadings",
"intercepts"))
# Strict model
sem.strict <- datos %>%
sem(modelo, .,
estimator="ULS",
group = "edaddicot",
group.equal = c("loadings",
"intercepts",
"residuals"))
# Model comparison
semTools::compareFit(sem.config,
sem.metric,
sem.scalar,
sem.strict) %>%
summary()## ################### Nested Model Comparison #########################
##
## Chi-Squared Difference Test
##
## Df AIC BIC Chisq Chisq diff RMSEA Df diff Pr(>Chisq)
## sem.config 590 551.16
## sem.metric 613 849.89 298.730 0.191179 23 < 2.2e-16 ***
## sem.scalar 633 865.98 16.089 0.000000 20 0.7111
## sem.strict 658 939.58 73.607 0.076991 25 1.106e-06 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## ####################### Model Fit Indices ###########################
## chisq df pvalue rmsea cfi tli srmr
## sem.config 551.158† 590 NA .000† 1.000† 1.001† .049†
## sem.metric 849.888 613 NA .034 .993 .992 .062
## sem.scalar 865.977 633 NA .034 .993 .993 .063
## sem.strict 939.584 658 NA .036 .991 .991 .066
##
## ################## Differences in Fit Indices #######################
## df rmsea cfi tli srmr
## sem.metric - sem.config 23 0.034 -0.007 -0.009 0.013
## sem.scalar - sem.metric 20 -0.001 0.000 0.000 0.001
## sem.strict - sem.scalar 25 0.003 -0.001 -0.001 0.003El modelo métrico ya difiere significativamente del modelo configural, por lo que no se verifica la invariaza en el modelo completo para la madre.
Estimaciones para el modelo configural (el que asume menos igualdades entre los grupos)
semPaths(# Argumentos globales
sem.config, what="diagram", whatLabels="std",layout="tree3", rotation = 2, width=50, height=35,exoVar = F,
# Etiquetas
curvePivot=T, curvature = .5
)
Nota. 1=Late; 2=Early.
sem.config %>% summary(standardized=T)## lavaan 0.6.16 ended normally after 76 iterations
##
## Estimator ULS
## Optimization method NLMINB
## Number of model parameters 164
##
## Number of observations per group:
## 2 302
## 1 354
##
## Model Test User Model:
##
## Test statistic 551.158
## Degrees of freedom 590
## P-value (Unknown) NA
## Test statistic for each group:
## 2 322.487
## 1 228.671
##
## Parameter Estimates:
##
## Standard errors Standard
## Information Expected
## Information saturated (h1) model Unstructured
##
##
## Group 1 [2]:
##
## Latent Variables:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## it =~
## siqs01 1.000 0.906 0.844
## siqs02 0.852 0.043 19.866 0.000 0.772 0.664
## siqs03 1.126 0.052 21.487 0.000 1.021 0.863
## cc =~
## siqs04 1.000 1.054 0.769
## siqs05 1.002 0.059 16.922 0.000 1.056 0.760
## siqs06 0.717 0.044 16.278 0.000 0.756 0.639
## ia =~
## siqs07 1.000 0.850 0.852
## siqs08 1.011 0.051 19.819 0.000 0.860 0.892
## siqs09 0.969 0.050 19.561 0.000 0.824 0.744
## SIQS =~
## it 1.000 0.893 0.893
## cc 0.640 0.040 15.875 0.000 0.491 0.491
## ia 0.925 0.056 16.440 0.000 0.880 0.880
## LTm =~
## tpqm02 1.000 0.705 0.674
## tpqm05 0.825 0.038 21.793 0.000 0.582 0.695
## tpqm12 1.143 0.045 25.243 0.000 0.806 0.720
## tpqm16 0.854 0.038 22.193 0.000 0.602 0.789
## tpqm01 0.826 0.038 21.805 0.000 0.582 0.692
## tpqm09 1.001 0.042 23.937 0.000 0.706 0.758
## tpqm11 1.125 0.045 25.094 0.000 0.793 0.752
## tpqm14 1.075 0.044 24.656 0.000 0.758 0.852
## tpqm03 0.755 0.036 20.735 0.000 0.532 0.523
## tpqm07 1.001 0.042 23.940 0.000 0.706 0.662
## tpqm10 1.031 0.043 24.244 0.000 0.727 0.844
## tpqm13 1.208 0.047 25.735 0.000 0.851 0.741
## tpqm04 0.936 0.040 23.221 0.000 0.660 0.761
## tpqm06 1.111 0.044 24.979 0.000 0.783 0.820
## tpqm08 0.851 0.038 22.148 0.000 0.600 0.704
## tpqm15 1.218 0.047 25.808 0.000 0.859 0.787
##
## Regressions:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## SIQS ~
## LTm 0.422 0.023 18.553 0.000 0.368 0.368
## sexo 0.286 0.112 2.562 0.010 0.354 0.177
##
## Intercepts:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .siqs01 4.739 0.082 57.485 0.000 4.739 4.414
## .siqs02 4.847 0.076 63.378 0.000 4.847 4.169
## .siqs03 4.568 0.088 51.977 0.000 4.568 3.862
## .siqs04 4.208 0.069 61.042 0.000 4.208 3.070
## .siqs05 4.165 0.069 60.380 0.000 4.165 2.996
## .siqs06 4.771 0.064 74.869 0.000 4.771 4.034
## .siqs07 5.124 0.079 64.551 0.000 5.124 5.134
## .siqs08 5.109 0.080 64.034 0.000 5.109 5.299
## .siqs09 4.853 0.078 62.049 0.000 4.853 4.384
## .tpqm02 5.411 0.058 93.883 0.000 5.411 5.175
## .tpqm05 5.672 0.058 98.422 0.000 5.672 6.782
## .tpqm12 5.328 0.058 92.446 0.000 5.328 4.763
## .tpqm16 5.666 0.058 98.307 0.000 5.666 7.426
## .tpqm01 5.546 0.058 96.238 0.000 5.546 6.596
## .tpqm09 5.626 0.058 97.617 0.000 5.626 6.045
## .tpqm11 5.371 0.058 93.193 0.000 5.371 5.095
## .tpqm14 5.606 0.058 97.273 0.000 5.606 6.303
## .tpqm03 5.268 0.058 91.412 0.000 5.268 5.181
## .tpqm07 5.278 0.058 91.584 0.000 5.278 4.947
## .tpqm10 5.586 0.058 96.928 0.000 5.586 6.486
## .tpqm13 5.301 0.058 91.987 0.000 5.301 4.613
## .tpqm04 5.566 0.058 96.583 0.000 5.566 6.421
## .tpqm06 5.560 0.058 96.468 0.000 5.560 5.823
## .tpqm08 5.566 0.058 96.583 0.000 5.566 6.538
## .tpqm15 5.325 0.058 92.389 0.000 5.325 4.881
## sexo 0.507 0.058 8.791 0.000 0.507 1.012
## .it 0.000 0.000 0.000
## .cc 0.000 0.000 0.000
## .ia 0.000 0.000 0.000
## .SIQS 0.000 0.000 0.000
## LTm 0.000 0.000 0.000
##
## Variances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .siqs01 0.332 0.080 4.122 0.000 0.332 0.288
## .siqs02 0.756 0.073 10.346 0.000 0.756 0.559
## .siqs03 0.358 0.089 4.013 0.000 0.358 0.256
## .siqs04 0.768 0.101 7.607 0.000 0.768 0.409
## .siqs05 0.818 0.101 8.081 0.000 0.818 0.423
## .siqs06 0.828 0.077 10.808 0.000 0.828 0.592
## .siqs07 0.273 0.079 3.457 0.001 0.273 0.274
## .siqs08 0.191 0.080 2.395 0.017 0.191 0.205
## .siqs09 0.547 0.077 7.078 0.000 0.547 0.446
## .tpqm02 0.596 0.064 9.248 0.000 0.596 0.545
## .tpqm05 0.361 0.062 5.805 0.000 0.361 0.517
## .tpqm12 0.602 0.067 9.022 0.000 0.602 0.481
## .tpqm16 0.220 0.063 3.512 0.000 0.220 0.378
## .tpqm01 0.368 0.062 5.916 0.000 0.368 0.521
## .tpqm09 0.368 0.064 5.711 0.000 0.368 0.425
## .tpqm11 0.482 0.066 7.265 0.000 0.482 0.434
## .tpqm14 0.217 0.066 3.309 0.001 0.217 0.274
## .tpqm03 0.751 0.061 12.220 0.000 0.751 0.726
## .tpqm07 0.640 0.064 9.925 0.000 0.640 0.562
## .tpqm10 0.213 0.065 3.283 0.001 0.213 0.287
## .tpqm13 0.596 0.068 8.791 0.000 0.596 0.451
## .tpqm04 0.316 0.064 4.970 0.000 0.316 0.421
## .tpqm06 0.298 0.066 4.504 0.000 0.298 0.327
## .tpqm08 0.365 0.063 5.841 0.000 0.365 0.504
## .tpqm15 0.453 0.068 6.661 0.000 0.453 0.381
## .it 0.167 0.050 3.354 0.001 0.203 0.203
## .cc 0.843 0.071 11.869 0.000 0.759 0.759
## .ia 0.163 0.047 3.495 0.000 0.225 0.225
## .SIQS 0.545 0.046 11.905 0.000 0.833 0.833
## LTm 0.497 0.029 17.182 0.000 1.000 1.000
## sexo 0.251 0.058 4.352 0.000 0.251 1.000
##
##
## Group 2 [1]:
##
## Latent Variables:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## it =~
## siqs01 1.000 0.979 0.871
## siqs02 0.957 0.036 26.874 0.000 0.936 0.708
## siqs03 1.176 0.042 28.178 0.000 1.151 0.932
## cc =~
## siqs04 1.000 0.854 0.696
## siqs05 1.112 0.049 22.708 0.000 0.949 0.725
## siqs06 1.073 0.048 22.513 0.000 0.916 0.778
## ia =~
## siqs07 1.000 0.813 0.877
## siqs08 1.075 0.044 24.249 0.000 0.874 0.890
## siqs09 0.920 0.040 22.902 0.000 0.748 0.683
## SIQS =~
## it 1.000 0.856 0.856
## cc 0.737 0.036 20.702 0.000 0.723 0.723
## ia 0.887 0.041 21.391 0.000 0.914 0.914
## LTm =~
## tpqm02 1.000 0.607 0.639
## tpqm05 1.249 0.050 25.151 0.000 0.758 0.816
## tpqm12 1.447 0.055 26.321 0.000 0.878 0.789
## tpqm16 0.667 0.036 18.370 0.000 0.405 0.662
## tpqm01 1.012 0.044 23.145 0.000 0.614 0.643
## tpqm09 1.398 0.054 26.066 0.000 0.848 0.843
## tpqm11 1.287 0.051 25.407 0.000 0.781 0.760
## tpqm14 1.115 0.046 24.113 0.000 0.676 0.835
## tpqm03 0.824 0.039 20.879 0.000 0.500 0.502
## tpqm07 1.005 0.044 23.073 0.000 0.610 0.589
## tpqm10 1.081 0.045 23.811 0.000 0.656 0.819
## tpqm13 1.530 0.057 26.707 0.000 0.928 0.800
## tpqm04 0.890 0.041 21.755 0.000 0.540 0.685
## tpqm06 1.222 0.049 24.961 0.000 0.741 0.826
## tpqm08 1.259 0.050 25.222 0.000 0.764 0.797
## tpqm15 1.326 0.052 25.654 0.000 0.805 0.733
##
## Regressions:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## SIQS ~
## LTm 0.563 0.026 21.582 0.000 0.407 0.407
## sexo 0.404 0.120 3.371 0.001 0.482 0.238
##
## Intercepts:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .siqs01 4.817 0.076 63.612 0.000 4.817 4.286
## .siqs02 4.680 0.074 63.171 0.000 4.680 3.537
## .siqs03 4.657 0.083 56.330 0.000 4.657 3.771
## .siqs04 4.524 0.066 68.100 0.000 4.524 3.689
## .siqs05 4.491 0.069 64.927 0.000 4.491 3.431
## .siqs06 4.939 0.068 72.434 0.000 4.939 4.195
## .siqs07 5.259 0.072 73.483 0.000 5.259 5.672
## .siqs08 5.191 0.074 70.147 0.000 5.191 5.285
## .siqs09 4.907 0.069 71.030 0.000 4.907 4.486
## .tpqm02 5.492 0.053 103.165 0.000 5.492 5.782
## .tpqm05 5.585 0.053 104.916 0.000 5.585 6.018
## .tpqm12 5.373 0.053 100.936 0.000 5.373 4.830
## .tpqm16 5.729 0.053 107.623 0.000 5.729 9.370
## .tpqm01 5.446 0.053 102.316 0.000 5.446 5.707
## .tpqm09 5.551 0.053 104.279 0.000 5.551 5.515
## .tpqm11 5.373 0.053 100.936 0.000 5.373 5.228
## .tpqm14 5.619 0.053 105.553 0.000 5.619 6.938
## .tpqm03 5.325 0.053 100.034 0.000 5.325 5.350
## .tpqm07 5.331 0.053 100.140 0.000 5.331 5.149
## .tpqm10 5.633 0.053 105.818 0.000 5.633 7.036
## .tpqm13 5.356 0.053 100.618 0.000 5.356 4.616
## .tpqm04 5.621 0.053 105.606 0.000 5.621 7.132
## .tpqm06 5.582 0.053 104.863 0.000 5.582 6.222
## .tpqm08 5.508 0.053 103.483 0.000 5.508 5.746
## .tpqm15 5.359 0.053 100.671 0.000 5.359 4.881
## sexo 0.412 0.053 7.748 0.000 0.412 0.837
## .it 0.000 0.000 0.000
## .cc 0.000 0.000 0.000
## .ia 0.000 0.000 0.000
## .SIQS 0.000 0.000 0.000
## LTm 0.000 0.000 0.000
##
## Variances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .siqs01 0.306 0.073 4.197 0.000 0.306 0.242
## .siqs02 0.874 0.071 12.346 0.000 0.874 0.499
## .siqs03 0.201 0.083 2.418 0.016 0.201 0.132
## .siqs04 0.775 0.071 10.888 0.000 0.775 0.516
## .siqs05 0.812 0.077 10.573 0.000 0.812 0.474
## .siqs06 0.548 0.075 7.332 0.000 0.548 0.395
## .siqs07 0.199 0.069 2.862 0.004 0.199 0.231
## .siqs08 0.200 0.073 2.744 0.006 0.200 0.208
## .siqs09 0.638 0.066 9.634 0.000 0.638 0.533
## .tpqm02 0.534 0.058 9.254 0.000 0.534 0.592
## .tpqm05 0.287 0.060 4.758 0.000 0.287 0.334
## .tpqm12 0.467 0.063 7.416 0.000 0.467 0.377
## .tpqm16 0.210 0.055 3.806 0.000 0.210 0.562
## .tpqm01 0.534 0.058 9.225 0.000 0.534 0.586
## .tpqm09 0.294 0.062 4.717 0.000 0.294 0.290
## .tpqm11 0.446 0.061 7.340 0.000 0.446 0.423
## .tpqm14 0.198 0.059 3.372 0.001 0.198 0.303
## .tpqm03 0.741 0.056 13.162 0.000 0.741 0.748
## .tpqm07 0.700 0.058 12.112 0.000 0.700 0.653
## .tpqm10 0.211 0.059 3.607 0.000 0.211 0.329
## .tpqm13 0.485 0.064 7.556 0.000 0.485 0.360
## .tpqm04 0.330 0.057 5.806 0.000 0.330 0.531
## .tpqm06 0.255 0.060 4.252 0.000 0.255 0.317
## .tpqm08 0.336 0.060 5.552 0.000 0.336 0.365
## .tpqm15 0.558 0.061 9.098 0.000 0.558 0.463
## .it 0.256 0.041 6.216 0.000 0.267 0.267
## .cc 0.348 0.037 9.464 0.000 0.477 0.477
## .ia 0.108 0.039 2.780 0.005 0.164 0.164
## .SIQS 0.546 0.039 14.169 0.000 0.778 0.778
## LTm 0.368 0.022 16.448 0.000 1.000 1.000
## sexo 0.243 0.053 4.565 0.000 0.243 1.000# [Grupo 1] SIQS = 0 + .407 * LTc
# [Grupo 2] SIQS = 0 + .368 * LTc
graf <- data.frame(edad=rep(1:2,2),
"LTm (std)"=c(rep(min(scale(datos$tpqm_LT),na.rm = T),2),rep(max(scale(datos$tpqm_LT),na.rm = T),2)),
"siqs (std)"=rep(NA,4),
check.names = F)
for(i in 1:4){
graf$`siqs (std)`[i] <- ifelse(graf$edad[i]==1,
.481*graf$`LTm (std)`[i],
.381*graf$`LTm (std)`[i])
}
graf %>%
ggplot(aes(x=`LTm (std)`, y=`siqs (std)`, color=as.factor(edad), group=as.factor(edad)))+
geom_line()+
scale_color_manual(name="Age stage",
values = 2:3,
labels = c("early",
"late"))
Padre
Sexo
modelo <- '
# Identidad social
it =~ siqs01 + siqs02 + siqs03
cc =~ siqs04 + siqs05 + siqs06
ia =~ siqs07 + siqs08 + siqs09
SIQS =~ it + cc + ia
# Liderazgo transformacional (coach)
LTp =~ tpqp02 + tpqp05 + tpqp12 + tpqp16 + tpqp01 + tpqp09 + tpqp11 + tpqp14 + tpqp03 + tpqp07 + tpqp10 + tpqp13 + tpqp04 + tpqp06 + tpqp08 + tpqp15
# REGRESIONES
SIQS ~ LTp + edad
'
# Configural model
sem.config <- datos %>%
sem(modelo, .,
estimator="ULS",
group = "sexo")
#sem.config %>%
# summary(fit.measures = TRUE, standardized = TRUE)
# Metric model
sem.metric <- datos %>%
sem(modelo, .,
estimator="ULS",
group = "sexo",
group.equal = "loadings")
# Scalar model
sem.scalar <- datos %>%
sem(modelo, .,
estimator="ULS",
group = "sexo",
group.equal = c("loadings",
"intercepts"))
# Strict model
sem.strict <- datos %>%
sem(modelo, .,
estimator="ULS",
group = "sexo",
group.equal = c("loadings",
"intercepts",
"residuals"))
# Model comparison
semTools::compareFit(sem.config,
sem.metric,
sem.scalar,
sem.strict) %>%
summary()## ################### Nested Model Comparison #########################
##
## Chi-Squared Difference Test
##
## Df AIC BIC Chisq Chisq diff RMSEA Df diff Pr(>Chisq)
## sem.config 590 925.72
## sem.metric 613 1210.66 284.94 0.186336 23 <2e-16 ***
## sem.scalar 633 1234.98 24.32 0.025661 20 0.2287
## sem.strict 658 1383.73 148.75 0.122849 25 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## ####################### Model Fit Indices ###########################
## chisq df pvalue rmsea cfi tli srmr
## sem.config 925.724† 590 NA .042† .994† .994† .045†
## sem.metric 1210.659 613 NA .055 .990 .989 .053
## sem.scalar 1234.978 633 NA .054 .990 .990 .054
## sem.strict 1383.731 658 NA .058 .988 .988 .058
##
## ################## Differences in Fit Indices #######################
## df rmsea cfi tli srmr
## sem.metric - sem.config 23 0.013 -0.004 -0.004 0.008
## sem.scalar - sem.metric 20 -0.001 0.000 0.000 0.001
## sem.strict - sem.scalar 25 0.004 -0.002 -0.002 0.005El modelo métrico ya difiere significativamente del modelo configural, por lo que no se verifica la invariaza en el modelo completo para el padre.
Estimaciones para el modelo configural (el que asume menos igualdades entre los grupos)
# sem.config %>%
# summary(standardized=T)
semPaths(# Argumentos globales
sem.config, what="diagram", whatLabels="std",layout="tree3", rotation = 2, width=50, height=35,exoVar = F,
# Etiquetas
curvePivot=T, curvature = .5
)
Nota. 1=Varones; 2=Mujeres.
# [Grupo 0] SIQS = 0 + .329 * LTc
# [Grupo 1] SIQS = 0 + .175 * LTc
graf <- data.frame(sexo=rep(0:1,2),
"LTp (std)"=c(rep(min(scale(datos$tpqp_LT),na.rm = T),2),rep(max(scale(datos$tpqp_LT),na.rm = T),2)),
"siqs (std)"=rep(NA,4),
check.names = F)
for(i in 1:4){
graf$`siqs (std)`[i] <- ifelse(graf$sexo[i]==0,
.329*graf$`LTp (std)`[i],
.175*graf$`LTp (std)`[i])
}
graf %>%
ggplot(aes(x=`LTp (std)`, y=`siqs (std)`, color=as.factor(sexo), group=as.factor(sexo)))+
geom_line()+
scale_color_manual(name="Sex",
values = 2:3,
labels = c("females",
"males"))
Edad
modelo <- '
# Identidad social
it =~ siqs01 + siqs02 + siqs03
cc =~ siqs04 + siqs05 + siqs06
ia =~ siqs07 + siqs08 + siqs09
SIQS =~ it + cc + ia
# Liderazgo transformacional (coach)
LTp =~ tpqp02 + tpqp05 + tpqp12 + tpqp16 + tpqp01 + tpqp09 + tpqp11 + tpqp14 + tpqp03 + tpqp07 + tpqp10 + tpqp13 + tpqp04 + tpqp06 + tpqp08 + tpqp15
# REGRESIONES
SIQS ~ LTp + sexo
'
# Configural model
sem.config <- datos %>%
sem(modelo, .,
estimator="ULS",
group = "edaddicot")
#sem.config %>%
# summary(fit.measures = TRUE, standardized = TRUE)
# Metric model
sem.metric <- datos %>%
sem(modelo, .,
estimator="ULS",
group = "edaddicot",
group.equal = "loadings")
# Scalar model
sem.scalar <- datos %>%
sem(modelo, .,
estimator="ULS",
group = "edaddicot",
group.equal = c("loadings",
"intercepts"))
# Strict model
sem.strict <- datos %>%
sem(modelo, .,
estimator="ULS",
group = "edaddicot",
group.equal = c("loadings",
"intercepts",
"residuals"))
# Model comparison
semTools::compareFit(sem.config,
sem.metric,
sem.scalar,
sem.strict) %>%
summary()## ################### Nested Model Comparison #########################
##
## Chi-Squared Difference Test
##
## Df AIC BIC Chisq Chisq diff RMSEA Df diff Pr(>Chisq)
## sem.config 590 799.73
## sem.metric 613 1135.56 335.83 0.203637 23 < 2.2e-16 ***
## sem.scalar 633 1163.09 27.53 0.033877 20 0.121
## sem.strict 658 1278.27 115.18 0.104871 25 1.552e-13 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## ####################### Model Fit Indices ###########################
## chisq df pvalue rmsea cfi tli srmr
## sem.config 799.727† 590 NA .033† 0.996† 0.996† .050†
## sem.metric 1135.561 613 NA .051 .991 .990 .057
## sem.scalar 1163.089 633 NA .051 .991 .991 .058
## sem.strict 1278.272 658 NA .054 .989 .989 .060
##
## ################## Differences in Fit Indices #######################
## df rmsea cfi tli srmr
## sem.metric - sem.config 23 0.018 -0.005 -0.006 0.007
## sem.scalar - sem.metric 20 0.000 0.000 0.000 0.001
## sem.strict - sem.scalar 25 0.003 -0.002 -0.001 0.002El modelo métrico ya difiere significativamente del modelo configural, por lo que no se verifica la invariaza en el modelo completo para el padre
Estimaciones para el modelo configural (el que asume menos igualdades entre los grupos)
# sem.config %>%
# summary(standardized=T)
semPaths(# Argumentos globales
sem.config, what="diagram", whatLabels="std",layout="tree3", rotation = 2, width=50, height=35,exoVar = F,
# Etiquetas
curvePivot=T, curvature = .5
)
Nota. 1=Late; 2=Early.
# [Grupo 1] SIQS = 0 + .377 * LTc
# [Grupo 2] SIQS = 0 + .218 * LTc
graf <- data.frame(edad=rep(1:2,2),
"LTp (std)"=c(rep(min(scale(datos$tpqp_LT),na.rm = T),2),rep(max(scale(datos$tpqp_LT),na.rm = T),2)),
"siqs (std)"=rep(NA,4),
check.names = F)
for(i in 1:4){
graf$`siqs (std)`[i] <- ifelse(graf$edad[i]==1,
.377*graf$`LTp (std)`[i],
.218*graf$`LTp (std)`[i])
}
graf %>%
ggplot(aes(x=`LTp (std)`, y=`siqs (std)`, color=as.factor(edad), group=as.factor(edad)))+
geom_line()+
scale_color_manual(name="Age stage",
values = 2:3,
labels = c("early",
"late"))
Madre y Padre
Probé también el modelo de contexto familiar (i.e., cada LT por separado aporta a un LT “del hogar” y eso como predictor de la identidad social).
Sexo
modelo <- '
# Identidad social
it =~ siqs01 + siqs02 + siqs03
cc =~ siqs04 + siqs05 + siqs06
ia =~ siqs07 + siqs08 + siqs09
SIQS =~ it + cc + ia
# Liderazgo transformacional (coach)
LTp =~ tpqp02 + tpqp05 + tpqp12 + tpqp16 + tpqp01 + tpqp09 + tpqp11 + tpqp14 + tpqp03 + tpqp07 + tpqp10 + tpqp13 + tpqp04 + tpqp06 + tpqp08 + tpqp15
LTm =~ tpqm02 + tpqm05 + tpqm12 + tpqm16 + tpqm01 + tpqm09 + tpqm11 + tpqm14 + tpqm03 + tpqm07 + tpqm10 + tpqm13 + tpqm04 + tpqm06 + tpqm08 + tpqm15
LT =~ LTm + LTp
# REGRESIONES
SIQS ~ LT + edad
'
# Configural model
sem.config <- datos %>%
sem(modelo, .,
estimator="ULS",
group = "sexo")
#sem.config %>%
# summary(fit.measures = TRUE, standardized = TRUE)
# Metric model
sem.metric <- datos %>%
sem(modelo, .,
estimator="ULS",
group = "sexo",
group.equal = "loadings")
# Scalar model
sem.scalar <- datos %>%
sem(modelo, .,
estimator="ULS",
group = "sexo",
group.equal = c("loadings",
"intercepts"))
# Strict model
sem.strict <- datos %>%
sem(modelo, .,
estimator="ULS",
group = "sexo",
group.equal = c("loadings",
"intercepts",
"residuals"))
# Model comparison
semTools::compareFit(sem.config,
sem.metric,
sem.scalar,
sem.strict) %>%
summary()## ################### Nested Model Comparison #########################
##
## Chi-Squared Difference Test
##
## Df AIC BIC Chisq Chisq diff RMSEA Df diff Pr(>Chisq)
## sem.config 1626 2623.5
## sem.metric 1665 3329.9 706.38 0.228411 39 <2e-16 ***
## sem.scalar 1699 3371.0 41.09 0.025219 34 0.1877
## sem.strict 1740 3650.7 279.72 0.133235 41 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## ####################### Model Fit Indices ###########################
## chisq df pvalue rmsea cfi tli srmr
## sem.config 2623.548† 1626 NA .043† .991† .991† .057†
## sem.metric 3329.927 1665 NA .055 .985 .985 .067
## sem.scalar 3371.019 1699 NA .055 .985 .985 .068
## sem.strict 3650.741 1740 NA .058 .983 .983 .072
##
## ################## Differences in Fit Indices #######################
## df rmsea cfi tli srmr
## sem.metric - sem.config 39 0.012 -0.006 -0.006 0.010
## sem.scalar - sem.metric 34 0.000 0.000 0.000 0.000
## sem.strict - sem.scalar 41 0.003 -0.002 -0.002 0.004El modelo métrico ya difiere significativamente del modelo configural, por lo que no se verifica la invariaza en el modelo completo para la parentalidad transformacional
Estimaciones para el modelo configural (el que asume menos igualdades entre los grupos)
# sem.config %>%
# summary(standardized=T)
semPaths(# Argumentos globales
sem.config, what="diagram", whatLabels="std",layout="tree2", residuals = F,
rotation = 2, width=50, height=35,exoVar = F,
# Etiquetas
curvePivot=T, curvature = .5
)
Nota. 1=Varones; 2=Mujeres.
# [Grupo 0] SIQS = 0 + .44 * LTc
# [Grupo 1] SIQS = 0 + .312 * LTc
graf <- data.frame(sexo=rep(0:1,2),
"LT (std)"=c(rep(min(c(scale(datos$tpqm_LT),scale(datos$tpqp_LT)),na.rm = T),2),rep(max(c(scale(datos$tpqm_LT),scale(datos$tpqp_LT)),na.rm = T),2)),
"siqs (std)"=rep(NA,4),
check.names = F)
for(i in 1:4){
graf$`siqs (std)`[i] <- ifelse(graf$sexo[i]==0,
.44*graf$`LT (std)`[i],
.312*graf$`LT (std)`[i])
}
graf %>%
ggplot(aes(x=`LT (std)`, y=`siqs (std)`, color=as.factor(sexo), group=as.factor(sexo)))+
geom_line()+
scale_color_manual(name="Sex",
values = 2:3,
labels = c("females",
"males"))
Edad
modelo <- '
# Identidad social
it =~ siqs01 + siqs02 + siqs03
cc =~ siqs04 + siqs05 + siqs06
ia =~ siqs07 + siqs08 + siqs09
SIQS =~ it + cc + ia
# Liderazgo transformacional (coach)
LTp =~ tpqp02 + tpqp05 + tpqp12 + tpqp16 + tpqp01 + tpqp09 + tpqp11 + tpqp14 + tpqp03 + tpqp07 + tpqp10 + tpqp13 + tpqp04 + tpqp06 + tpqp08 + tpqp15
LTm =~ tpqm02 + tpqm05 + tpqm12 + tpqm16 + tpqm01 + tpqm09 + tpqm11 + tpqm14 + tpqm03 + tpqm07 + tpqm10 + tpqm13 + tpqm04 + tpqm06 + tpqm08 + tpqm15
LT =~ LTm + LTp
# REGRESIONES
SIQS ~ LT + sexo
'
# Configural model
sem.config <- datos %>%
sem(modelo, .,
estimator="ULS",
group = "edaddicot")
#sem.config %>%
# summary(fit.measures = TRUE, standardized = TRUE)
# Metric model
sem.metric <- datos %>%
sem(modelo, .,
estimator="ULS",
group = "edaddicot",
group.equal = "loadings")
# Scalar model
sem.scalar <- datos %>%
sem(modelo, .,
estimator="ULS",
group = "edaddicot",
group.equal = c("loadings",
"intercepts"))
# Strict model
sem.strict <- datos %>%
sem(modelo, .,
estimator="ULS",
group = "edaddicot",
group.equal = c("loadings",
"intercepts",
"residuals"))
# Model comparison
semTools::compareFit(sem.config,
sem.metric,
sem.scalar,
sem.strict) %>%
summary()## ################### Nested Model Comparison #########################
##
## Chi-Squared Difference Test
##
## Df AIC BIC Chisq Chisq diff RMSEA Df diff Pr(>Chisq)
## sem.config 1626 2363.4
## sem.metric 1665 3260.2 896.81 0.258956 39 < 2.2e-16 ***
## sem.scalar 1699 3296.2 36.06 0.013606 34 0.3722
## sem.strict 1740 3454.8 158.52 0.093480 41 9.943e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## ####################### Model Fit Indices ###########################
## chisq df pvalue rmsea cfi tli srmr
## sem.config 2363.374† 1626 NA .037† .993† .993† .058†
## sem.metric 3260.181 1665 NA .054 .985 .985 .068
## sem.scalar 3296.245 1699 NA .054 .985 .985 .068
## sem.strict 3454.761 1740 NA .055 .984 .984 .070
##
## ################## Differences in Fit Indices #######################
## df rmsea cfi tli srmr
## sem.metric - sem.config 39 0.017 -0.008 -0.008 0.010
## sem.scalar - sem.metric 34 -0.001 0.000 0.000 0.000
## sem.strict - sem.scalar 41 0.001 -0.001 -0.001 0.002El modelo métrico ya difiere significativamente del modelo configural, por lo que no se verifica la invariaza en el modelo completo para el padre
Estimaciones para el modelo configural (el que asume menos igualdades entre los grupos)
# sem.config %>%
# summary(standardized=T)
semPaths(# Argumentos globales
sem.config, what="diagram", whatLabels="std",layout="tree2", residuals = F,
rotation = 2, width=50, height=35,exoVar = F,
# Etiquetas
curvePivot=T, curvature = .5
)
Nota. 1=Late; 2=Early.
# [Grupo 1] SIQS = 0 + .487 * LTc
# [Grupo 2] SIQS = 0 + .359 * LTc
graf <- data.frame(edad=rep(1:2,2),
"LT (std)"=c(rep(min(c(scale(datos$tpqm_LT),scale(datos$tpqp_LT)),na.rm = T),2),rep(max(c(scale(datos$tpqm_LT),scale(datos$tpqp_LT)),na.rm = T),2)),
"siqs (std)"=rep(NA,4),
check.names = F)
for(i in 1:4){
graf$`siqs (std)`[i] <- ifelse(graf$edad[i]==1,
.487*graf$`LT (std)`[i],
.359*graf$`LT (std)`[i])
}
graf %>%
ggplot(aes(x=`LT (std)`, y=`siqs (std)`, color=as.factor(edad), group=as.factor(edad)))+
geom_line()+
scale_color_manual(name="Age stage",
values = 2:3,
labels = c("early",
"late"))
Contexto Total
Probé también el modelo de contexto familiar y entrenador en simultaneo.
Sexo
modelo <- '
# Identidad social
it =~ siqs01 + siqs02 + siqs03
cc =~ siqs04 + siqs05 + siqs06
ia =~ siqs07 + siqs08 + siqs09
SIQS =~ it + cc + ia
# Liderazgo transformacional (coach)
LTp =~ tpqp02 + tpqp05 + tpqp12 + tpqp16 + tpqp01 + tpqp09 + tpqp11 + tpqp14 + tpqp03 + tpqp07 + tpqp10 + tpqp13 + tpqp04 + tpqp06 + tpqp08 + tpqp15
LTm =~ tpqm02 + tpqm05 + tpqm12 + tpqm16 + tpqm01 + tpqm09 + tpqm11 + tpqm14 + tpqm03 + tpqm07 + tpqm10 + tpqm13 + tpqm04 + tpqm06 + tpqm08 + tpqm15
LT =~ LTm + LTp
LTc =~ tcq02 + tcq05 + tcq12 + tcq16 + tcq01 + tcq09 + tcq11 + tcq14 + tcq03 + tcq07 + tcq10 + tcq13 + tcq04 + tcq06 + tcq08 + tcq15
# REGRESIONES
SIQS ~ LT + LTc + edad
'
# Configural model
sem.config.sexo <- datos %>%
sem(modelo, .,
estimator="ULS",
group = "sexo")
#sem.config %>%
# summary(fit.measures = TRUE, standardized = TRUE)
# Metric model
sem.metric <- datos %>%
sem(modelo, .,
estimator="ULS",
group = "sexo",
group.equal = "loadings")
# Scalar model
sem.scalar <- datos %>%
sem(modelo, .,
estimator="ULS",
group = "sexo",
group.equal = c("loadings",
"intercepts"))
# Strict model
sem.strict <- datos %>%
sem(modelo, .,
estimator="ULS",
group = "sexo",
group.equal = c("loadings",
"intercepts",
"residuals"))
# Model comparison
semTools::compareFit(sem.config.sexo,
sem.metric,
sem.scalar,
sem.strict) %>%
summary()## ################### Nested Model Comparison #########################
##
## Chi-Squared Difference Test
##
## Df AIC BIC Chisq Chisq diff RMSEA Df diff Pr(>Chisq)
## sem.config.sexo 3174 6591.7
## sem.metric 3228 7629.8 1038.08 0.23571 54 < 2e-16 ***
## sem.scalar 3277 7701.0 71.13 0.03711 49 0.02106 *
## sem.strict 3334 8095.0 394.04 0.13427 57 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## ####################### Model Fit Indices ###########################
## chisq df pvalue rmsea cfi tli srmr
## sem.config.sexo 6591.747† 3174 NA .057† .981† .980† .057†
## sem.metric 7629.830 3228 NA .065 .975 .975 .064
## sem.scalar 7700.964 3277 NA .064 .975 .975 .064
## sem.strict 8095.001 3334 NA .066 .973 .973 .067
##
## ################## Differences in Fit Indices #######################
## df rmsea cfi tli srmr
## sem.metric - sem.config.sexo 54 0.007 -0.006 -0.005 0.007
## sem.scalar - sem.metric 49 0.000 0.000 0.000 0.000
## sem.strict - sem.scalar 57 0.002 -0.002 -0.001 0.003El modelo métrico ya difiere significativamente del modelo configural, por lo que no se verifica la invariaza en el modelo completo para ambos contextos sociales
Estimaciones para el modelo configural (el que asume menos igualdades entre los grupos)
# sem.config %>%
# summary(standardized=T)
semPaths(# Argumentos globales
sem.config.sexo, what="diagram", whatLabels="std",layout="tree2", residuals = F,
rotation = 2, width=50, height=35,exoVar = F,
# Etiquetas
curvePivot=T, curvature = .5
)
Nota. 1=Varones; 2=Mujeres.
# [Grupo 0] SIQS = 0 + .390 * LT
# [Grupo 1] SIQS = 0 + .201 * LT
graf <- data.frame(sexo=rep(0:1,2),
"LT (std)"=c(rep(min(c(scale(datos$tpqm_LT),scale(datos$tpqp_LT)),na.rm = T),2),rep(max(c(scale(datos$tpqm_LT),scale(datos$tpqp_LT)),na.rm = T),2)),
"siqs (std)"=rep(NA,4),
check.names = F)
for(i in 1:4){
graf$`siqs (std)`[i] <- ifelse(graf$sexo[i]==0,
.390*graf$`LT (std)`[i],
.201*graf$`LT (std)`[i])
}
p1 <- graf %>%
ggplot(aes(x=`LT (std)`, y=`siqs (std)`, color=as.factor(sexo), group=as.factor(sexo)))+
geom_line()+
scale_color_manual(name="Sex",
values = 2:3,
labels = c("females",
"males"))+
ylim(c(min(scale(datos$total_IS)),max(scale(datos$total_IS))))
# [Grupo 0] SIQS = 0 + .284 * LTc
# [Grupo 1] SIQS = 0 + .525 * LTc
graf <- data.frame(sexo=rep(0:1,2),
"LTc (std)"=c(rep(min(scale(datos$tcq_LT),na.rm = T),2),rep(max(scale(datos$tcq_LT),na.rm = T),2)),
"siqs (std)"=rep(NA,4),
check.names = F)
for(i in 1:4){
graf$`siqs (std)`[i] <- ifelse(graf$sexo[i]==0,
.284*graf$`LTc (std)`[i],
.525*graf$`LTc (std)`[i])
}
p2 <- graf %>%
ggplot(aes(x=`LTc (std)`, y=`siqs (std)`, color=as.factor(sexo), group=as.factor(sexo)))+
geom_line()+
scale_color_manual(name="Sex",
values = 2:3,
labels = c("females",
"males"))+
ylim(c(min(scale(datos$total_IS)),max(scale(datos$total_IS))))
grid.arrange(p1,p2, nrow=1)
Edad
modelo <- '
# Identidad social
it =~ siqs01 + siqs02 + siqs03
cc =~ siqs04 + siqs05 + siqs06
ia =~ siqs07 + siqs08 + siqs09
SIQS =~ it + cc + ia
# Liderazgo transformacional (coach)
LTp =~ tpqp02 + tpqp05 + tpqp12 + tpqp16 + tpqp01 + tpqp09 + tpqp11 + tpqp14 + tpqp03 + tpqp07 + tpqp10 + tpqp13 + tpqp04 + tpqp06 + tpqp08 + tpqp15
LTm =~ tpqm02 + tpqm05 + tpqm12 + tpqm16 + tpqm01 + tpqm09 + tpqm11 + tpqm14 + tpqm03 + tpqm07 + tpqm10 + tpqm13 + tpqm04 + tpqm06 + tpqm08 + tpqm15
LT =~ LTm + LTp
LTc =~ tcq02 + tcq05 + tcq12 + tcq16 + tcq01 + tcq09 + tcq11 + tcq14 + tcq03 + tcq07 + tcq10 + tcq13 + tcq04 + tcq06 + tcq08 + tcq15
# REGRESIONES
SIQS ~ LT + LTc + sexo
'
# Configural model
sem.config.edad <- datos %>%
sem(modelo, .,
estimator="ULS",
group = "edaddicot")
#sem.config %>%
# summary(fit.measures = TRUE, standardized = TRUE)
# Metric model
sem.metric <- datos %>%
sem(modelo, .,
estimator="ULS",
group = "edaddicot",
group.equal = "loadings")
# Scalar model
sem.scalar <- datos %>%
sem(modelo, .,
estimator="ULS",
group = "edaddicot",
group.equal = c("loadings",
"intercepts"))
# Strict model
sem.strict <- datos %>%
sem(modelo, .,
estimator="ULS",
group = "edaddicot",
group.equal = c("loadings",
"intercepts",
"residuals"))
# Model comparison
semTools::compareFit(sem.config.edad,
sem.metric,
sem.scalar,
sem.strict) %>%
summary()## ################### Nested Model Comparison #########################
##
## Chi-Squared Difference Test
##
## Df AIC BIC Chisq Chisq diff RMSEA Df diff Pr(>Chisq)
## sem.config.edad 3174 5373.0
## sem.metric 3228 6720.6 1347.57 0.270248 54 < 2e-16 ***
## sem.scalar 3277 6793.4 72.75 0.038442 49 0.01542 *
## sem.strict 3334 7115.7 322.32 0.119127 57 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## ####################### Model Fit Indices ###########################
## chisq df pvalue rmsea cfi tli srmr
## sem.config.edad 5373.037† 3174 NA .046† .987† .987† .057†
## sem.metric 6720.611 3228 NA .058 .980 .979 .064
## sem.scalar 6793.361 3277 NA .057 .980 .979 .064
## sem.strict 7115.682 3334 NA .059 .978 .978 .066
##
## ################## Differences in Fit Indices #######################
## df rmsea cfi tli srmr
## sem.metric - sem.config.edad 54 0.011 -0.008 -0.007 0.007
## sem.scalar - sem.metric 49 0.000 0.000 0.000 0.000
## sem.strict - sem.scalar 57 0.002 -0.002 -0.001 0.001El modelo métrico ya difiere significativamente del modelo configural, por lo que no se verifica la invariaza en el modelo completo para ambos contextos
Estimaciones para el modelo configural (el que asume menos igualdades entre los grupos)
# sem.config %>%
# summary(standardized=T)
semPaths(# Argumentos globales
sem.config.edad, what="diagram", whatLabels="std",layout="tree2", residuals = F,
rotation = 2, width=50, height=35,exoVar = F,
# Etiquetas
curvePivot=T, curvature = .5
)
Nota. 1=Late; 2=Early.
sem.config.edad %>% summary(standardized=T)## lavaan 0.6.16 ended normally after 116 iterations
##
## Estimator ULS
## Optimization method NLMINB
## Number of model parameters 364
##
## Number of observations per group:
## 2 302
## 1 354
##
## Model Test User Model:
##
## Test statistic 5373.037
## Degrees of freedom 3174
## P-value (Unknown) NA
## Test statistic for each group:
## 2 2700.585
## 1 2672.453
##
## Parameter Estimates:
##
## Standard errors Standard
## Information Expected
## Information saturated (h1) model Unstructured
##
##
## Group 1 [2]:
##
## Latent Variables:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## it =~
## siqs01 1.000 0.907 0.845
## siqs02 0.858 0.034 24.869 0.000 0.778 0.669
## siqs03 1.117 0.041 27.395 0.000 1.013 0.856
## cc =~
## siqs04 1.000 1.068 0.779
## siqs05 0.931 0.047 19.896 0.000 0.995 0.715
## siqs06 0.745 0.039 18.874 0.000 0.796 0.673
## ia =~
## siqs07 1.000 0.857 0.858
## siqs08 1.014 0.042 23.961 0.000 0.869 0.901
## siqs09 0.945 0.040 23.340 0.000 0.809 0.731
## SIQS =~
## it 1.000 0.916 0.916
## cc 0.655 0.031 20.822 0.000 0.509 0.509
## ia 0.868 0.038 22.679 0.000 0.842 0.842
## LTp =~
## tpqp02 1.000 0.724 0.685
## tpqp05 1.107 0.031 35.800 0.000 0.801 0.760
## tpqp12 1.601 0.040 40.367 0.000 1.158 0.812
## tpqp16 1.120 0.031 35.966 0.000 0.810 0.749
## tpqp01 1.063 0.030 35.177 0.000 0.769 0.723
## tpqp09 1.244 0.033 37.437 0.000 0.900 0.794
## tpqp11 1.200 0.032 36.949 0.000 0.868 0.738
## tpqp14 1.180 0.032 36.712 0.000 0.854 0.868
## tpqp03 0.863 0.027 31.784 0.000 0.624 0.580
## tpqp07 1.352 0.035 38.503 0.000 0.978 0.764
## tpqp10 1.139 0.031 36.213 0.000 0.824 0.850
## tpqp13 1.405 0.036 38.964 0.000 1.017 0.792
## tpqp04 1.160 0.032 36.470 0.000 0.839 0.742
## tpqp06 1.193 0.032 36.871 0.000 0.864 0.826
## tpqp08 1.209 0.033 37.056 0.000 0.875 0.821
## tpqp15 1.418 0.036 39.067 0.000 1.026 0.840
## LTm =~
## tpqm02 1.000 0.674 0.645
## tpqm05 0.750 0.029 25.421 0.000 0.505 0.604
## tpqm12 1.329 0.040 33.199 0.000 0.896 0.801
## tpqm16 0.842 0.031 27.199 0.000 0.568 0.744
## tpqm01 0.813 0.030 26.667 0.000 0.548 0.652
## tpqm09 0.975 0.033 29.332 0.000 0.658 0.707
## tpqm11 1.162 0.037 31.649 0.000 0.784 0.744
## tpqm14 1.136 0.036 31.367 0.000 0.766 0.861
## tpqm03 0.832 0.031 27.018 0.000 0.561 0.552
## tpqm07 1.028 0.034 30.056 0.000 0.693 0.650
## tpqm10 1.090 0.035 30.841 0.000 0.735 0.854
## tpqm13 1.310 0.040 33.045 0.000 0.884 0.769
## tpqm04 0.901 0.032 28.193 0.000 0.607 0.701
## tpqm06 1.162 0.037 31.647 0.000 0.784 0.821
## tpqm08 0.892 0.032 28.050 0.000 0.601 0.706
## tpqm15 1.349 0.040 33.359 0.000 0.910 0.834
## LT =~
## LTm 1.000 0.987 0.987
## LTp 0.694 0.033 21.153 0.000 0.638 0.638
## LTc =~
## tcq02 1.000 1.012 0.734
## tcq05 0.982 0.023 42.347 0.000 0.994 0.777
## tcq12 0.717 0.020 36.109 0.000 0.726 0.487
## tcq16 1.064 0.024 43.729 0.000 1.077 0.776
## tcq01 1.036 0.024 43.282 0.000 1.048 0.827
## tcq09 0.926 0.022 41.264 0.000 0.937 0.694
## tcq11 0.915 0.022 41.035 0.000 0.926 0.713
## tcq14 0.849 0.021 39.583 0.000 0.859 0.724
## tcq03 0.720 0.020 36.179 0.000 0.728 0.603
## tcq07 0.706 0.020 35.778 0.000 0.715 0.632
## tcq10 0.579 0.018 31.493 0.000 0.586 0.439
## tcq13 0.823 0.021 38.967 0.000 0.833 0.709
## tcq04 1.022 0.024 43.053 0.000 1.034 0.800
## tcq06 0.993 0.023 42.544 0.000 1.005 0.801
## tcq08 0.858 0.022 39.802 0.000 0.868 0.757
## tcq15 0.876 0.022 40.211 0.000 0.887 0.746
##
## Regressions:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## SIQS ~
## LT 0.390 0.019 20.166 0.000 0.312 0.312
## LTc 0.272 0.011 25.280 0.000 0.332 0.332
## sexo 0.292 0.114 2.557 0.011 0.352 0.176
##
## Covariances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## LT ~~
## LTc 0.099 0.004 22.917 0.000 0.147 0.147
##
## Intercepts:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .siqs01 4.736 0.083 56.783 0.000 4.736 4.411
## .siqs02 4.843 0.077 62.535 0.000 4.843 4.166
## .siqs03 4.566 0.089 51.533 0.000 4.566 3.860
## .siqs04 4.204 0.070 60.152 0.000 4.204 3.067
## .siqs05 4.168 0.068 60.940 0.000 4.168 2.998
## .siqs06 4.766 0.065 73.602 0.000 4.766 4.029
## .siqs07 5.130 0.078 65.890 0.000 5.130 5.140
## .siqs08 5.115 0.078 65.285 0.000 5.115 5.304
## .siqs09 4.862 0.076 64.017 0.000 4.862 4.392
## .tpqp02 5.308 0.058 92.102 0.000 5.308 5.021
## .tpqp05 5.464 0.058 94.802 0.000 5.464 5.178
## .tpqp12 4.851 0.058 84.173 0.000 4.851 3.401
## .tpqp16 5.440 0.058 94.400 0.000 5.440 5.028
## .tpqp01 5.371 0.058 93.193 0.000 5.371 5.050
## .tpqp09 5.411 0.058 93.883 0.000 5.411 4.771
## .tpqp11 5.192 0.058 90.091 0.000 5.192 4.412
## .tpqp14 5.523 0.058 95.836 0.000 5.523 5.614
## .tpqp03 5.275 0.058 91.527 0.000 5.275 4.904
## .tpqp07 5.053 0.058 87.677 0.000 5.053 3.948
## .tpqp10 5.520 0.058 95.779 0.000 5.520 5.689
## .tpqp13 5.132 0.058 89.056 0.000 5.132 3.995
## .tpqp04 5.285 0.058 91.699 0.000 5.285 4.670
## .tpqp06 5.447 0.058 94.515 0.000 5.447 5.213
## .tpqp08 5.424 0.058 94.113 0.000 5.424 5.090
## .tpqp15 5.126 0.058 88.941 0.000 5.126 4.196
## .tpqm02 5.411 0.058 93.883 0.000 5.411 5.175
## .tpqm05 5.672 0.058 98.422 0.000 5.672 6.782
## .tpqm12 5.328 0.058 92.446 0.000 5.328 4.763
## .tpqm16 5.666 0.058 98.307 0.000 5.666 7.426
## .tpqm01 5.546 0.058 96.238 0.000 5.546 6.596
## .tpqm09 5.626 0.058 97.617 0.000 5.626 6.045
## .tpqm11 5.371 0.058 93.193 0.000 5.371 5.095
## .tpqm14 5.606 0.058 97.273 0.000 5.606 6.303
## .tpqm03 5.268 0.058 91.412 0.000 5.268 5.181
## .tpqm07 5.278 0.058 91.584 0.000 5.278 4.947
## .tpqm10 5.586 0.058 96.928 0.000 5.586 6.486
## .tpqm13 5.301 0.058 91.987 0.000 5.301 4.613
## .tpqm04 5.566 0.058 96.583 0.000 5.566 6.421
## .tpqm06 5.560 0.058 96.468 0.000 5.560 5.823
## .tpqm08 5.566 0.058 96.583 0.000 5.566 6.538
## .tpqm15 5.325 0.058 92.389 0.000 5.325 4.881
## .tcq02 3.977 0.058 69.004 0.000 3.977 2.883
## .tcq05 4.788 0.058 83.081 0.000 4.788 3.743
## .tcq12 3.854 0.058 66.878 0.000 3.854 2.588
## .tcq16 4.682 0.058 81.242 0.000 4.682 3.374
## .tcq01 4.603 0.058 79.863 0.000 4.603 3.628
## .tcq09 4.897 0.058 84.977 0.000 4.897 3.630
## .tcq11 4.589 0.058 79.634 0.000 4.589 3.536
## .tcq14 4.772 0.058 82.794 0.000 4.772 4.024
## .tcq03 4.646 0.058 80.610 0.000 4.646 3.846
## .tcq07 4.818 0.058 83.598 0.000 4.818 4.260
## .tcq10 4.384 0.058 76.071 0.000 4.384 3.281
## .tcq13 4.517 0.058 78.370 0.000 4.517 3.845
## .tcq04 4.795 0.058 83.196 0.000 4.795 3.708
## .tcq06 4.712 0.058 81.760 0.000 4.712 3.755
## .tcq08 5.073 0.058 88.022 0.000 5.073 4.420
## .tcq15 4.844 0.058 84.058 0.000 4.844 4.074
## sexo 0.507 0.058 8.791 0.000 0.507 1.012
## .it 0.000 0.000 0.000
## .cc 0.000 0.000 0.000
## .ia 0.000 0.000 0.000
## .SIQS 0.000 0.000 0.000
## .LTp 0.000 0.000 0.000
## .LTm 0.000 0.000 0.000
## LT 0.000 0.000 0.000
## LTc 0.000 0.000 0.000
##
## Variances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .siqs01 0.330 0.076 4.350 0.000 0.330 0.286
## .siqs02 0.746 0.070 10.693 0.000 0.746 0.552
## .siqs03 0.373 0.082 4.525 0.000 0.373 0.266
## .siqs04 0.737 0.095 7.750 0.000 0.737 0.392
## .siqs05 0.944 0.089 10.649 0.000 0.944 0.488
## .siqs06 0.766 0.076 10.122 0.000 0.766 0.547
## .siqs07 0.262 0.076 3.460 0.001 0.262 0.263
## .siqs08 0.175 0.077 2.290 0.022 0.175 0.189
## .siqs09 0.571 0.073 7.793 0.000 0.571 0.466
## .tpqp02 0.594 0.061 9.657 0.000 0.594 0.531
## .tpqp05 0.471 0.062 7.550 0.000 0.471 0.423
## .tpqp12 0.692 0.068 10.195 0.000 0.692 0.340
## .tpqp16 0.514 0.063 8.223 0.000 0.514 0.439
## .tpqp01 0.540 0.062 8.708 0.000 0.540 0.477
## .tpqp09 0.476 0.064 7.470 0.000 0.476 0.370
## .tpqp11 0.631 0.063 9.974 0.000 0.631 0.456
## .tpqp14 0.239 0.063 3.791 0.000 0.239 0.247
## .tpqp03 0.767 0.060 12.680 0.000 0.767 0.663
## .tpqp07 0.681 0.065 10.506 0.000 0.681 0.416
## .tpqp10 0.262 0.063 4.180 0.000 0.262 0.278
## .tpqp13 0.616 0.065 9.417 0.000 0.616 0.373
## .tpqp04 0.576 0.063 9.169 0.000 0.576 0.450
## .tpqp06 0.346 0.063 5.479 0.000 0.346 0.317
## .tpqp08 0.370 0.063 5.833 0.000 0.370 0.325
## .tpqp15 0.440 0.066 6.708 0.000 0.440 0.295
## .tpqm02 0.639 0.062 10.379 0.000 0.639 0.584
## .tpqm05 0.444 0.060 7.426 0.000 0.444 0.635
## .tpqm12 0.448 0.065 6.926 0.000 0.448 0.358
## .tpqm16 0.260 0.060 4.301 0.000 0.260 0.446
## .tpqm01 0.407 0.060 6.757 0.000 0.407 0.575
## .tpqm09 0.434 0.061 7.073 0.000 0.434 0.501
## .tpqm11 0.497 0.063 7.892 0.000 0.497 0.447
## .tpqm14 0.204 0.063 3.253 0.001 0.204 0.258
## .tpqm03 0.719 0.060 11.928 0.000 0.719 0.696
## .tpqm07 0.658 0.062 10.654 0.000 0.658 0.578
## .tpqm10 0.201 0.062 3.228 0.001 0.201 0.271
## .tpqm13 0.540 0.064 8.380 0.000 0.540 0.409
## .tpqm04 0.383 0.061 6.296 0.000 0.383 0.509
## .tpqm06 0.298 0.063 4.729 0.000 0.298 0.327
## .tpqm08 0.363 0.061 5.982 0.000 0.363 0.501
## .tpqm15 0.363 0.065 5.588 0.000 0.363 0.305
## .tcq02 0.879 0.067 13.214 0.000 0.879 0.462
## .tcq05 0.648 0.066 9.794 0.000 0.648 0.396
## .tcq12 1.691 0.062 27.228 0.000 1.691 0.762
## .tcq16 0.766 0.068 11.305 0.000 0.766 0.398
## .tcq01 0.510 0.067 7.585 0.000 0.510 0.317
## .tcq09 0.942 0.065 14.449 0.000 0.942 0.518
## .tcq11 0.828 0.065 12.735 0.000 0.828 0.492
## .tcq14 0.668 0.064 10.451 0.000 0.668 0.475
## .tcq03 0.928 0.062 14.942 0.000 0.928 0.636
## .tcq07 0.768 0.062 12.399 0.000 0.768 0.601
## .tcq10 1.442 0.061 23.821 0.000 1.442 0.808
## .tcq13 0.687 0.064 10.800 0.000 0.687 0.497
## .tcq04 0.602 0.067 8.989 0.000 0.602 0.360
## .tcq06 0.565 0.066 8.506 0.000 0.565 0.359
## .tcq08 0.563 0.064 8.779 0.000 0.563 0.427
## .tcq15 0.628 0.064 9.753 0.000 0.628 0.444
## .it 0.132 0.044 3.008 0.003 0.161 0.161
## .cc 0.846 0.065 12.924 0.000 0.741 0.741
## .ia 0.214 0.042 5.114 0.000 0.292 0.292
## .SIQS 0.505 0.036 14.116 0.000 0.731 0.731
## .LTp 0.311 0.015 20.370 0.000 0.593 0.593
## .LTm 0.012 0.017 0.696 0.487 0.026 0.026
## LT 0.443 0.026 16.860 0.000 1.000 1.000
## LTc 1.024 0.033 30.799 0.000 1.000 1.000
## sexo 0.251 0.058 4.352 0.000 0.251 1.000
##
##
## Group 2 [1]:
##
## Latent Variables:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## it =~
## siqs01 1.000 1.021 0.908
## siqs02 0.892 0.026 33.852 0.000 0.910 0.688
## siqs03 1.109 0.030 36.529 0.000 1.133 0.917
## cc =~
## siqs04 1.000 0.817 0.666
## siqs05 1.204 0.038 31.386 0.000 0.984 0.752
## siqs06 1.123 0.037 30.691 0.000 0.917 0.779
## ia =~
## siqs07 1.000 0.812 0.876
## siqs08 1.057 0.033 31.891 0.000 0.858 0.873
## siqs09 0.942 0.031 30.377 0.000 0.765 0.699
## SIQS =~
## it 1.000 0.791 0.791
## cc 0.815 0.027 29.835 0.000 0.807 0.807
## ia 0.903 0.029 31.068 0.000 0.899 0.899
## LTp =~
## tpqp02 1.000 0.631 0.647
## tpqp05 1.480 0.038 38.774 0.000 0.934 0.824
## tpqp12 1.662 0.042 39.985 0.000 1.049 0.806
## tpqp16 1.162 0.033 35.678 0.000 0.733 0.742
## tpqp01 1.193 0.033 36.054 0.000 0.753 0.676
## tpqp09 1.611 0.041 39.676 0.000 1.016 0.830
## tpqp11 1.584 0.040 39.506 0.000 1.000 0.792
## tpqp14 1.256 0.034 36.763 0.000 0.793 0.825
## tpqp03 0.956 0.029 32.693 0.000 0.603 0.545
## tpqp07 1.176 0.033 35.852 0.000 0.742 0.666
## tpqp10 1.324 0.035 37.441 0.000 0.835 0.824
## tpqp13 1.672 0.042 40.038 0.000 1.055 0.835
## tpqp04 1.354 0.036 37.725 0.000 0.854 0.785
## tpqp06 1.260 0.034 36.806 0.000 0.795 0.786
## tpqp08 1.441 0.037 38.471 0.000 0.909 0.794
## tpqp15 1.544 0.039 39.239 0.000 0.974 0.795
## LTm =~
## tpqm02 1.000 0.581 0.612
## tpqm05 1.258 0.040 31.598 0.000 0.731 0.788
## tpqm12 1.480 0.044 33.317 0.000 0.860 0.773
## tpqm16 0.702 0.030 23.506 0.000 0.408 0.667
## tpqm01 1.023 0.035 29.009 0.000 0.594 0.623
## tpqm09 1.456 0.044 33.153 0.000 0.846 0.841
## tpqm11 1.443 0.044 33.067 0.000 0.839 0.816
## tpqm14 1.172 0.038 30.763 0.000 0.681 0.841
## tpqm03 0.888 0.033 27.025 0.000 0.516 0.518
## tpqm07 1.074 0.036 29.662 0.000 0.624 0.603
## tpqm10 1.124 0.037 30.239 0.000 0.653 0.816
## tpqm13 1.577 0.047 33.902 0.000 0.916 0.790
## tpqm04 0.937 0.034 27.793 0.000 0.544 0.691
## tpqm06 1.216 0.039 31.199 0.000 0.706 0.787
## tpqm08 1.303 0.041 31.999 0.000 0.758 0.790
## tpqm15 1.450 0.044 33.119 0.000 0.843 0.768
## LT =~
## LTm 1.000 0.821 0.821
## LTp 1.043 0.040 26.381 0.000 0.789 0.789
## LTc =~
## tcq02 1.000 0.931 0.634
## tcq05 0.930 0.022 42.958 0.000 0.866 0.727
## tcq12 0.873 0.021 41.648 0.000 0.813 0.501
## tcq16 1.081 0.024 45.833 0.000 1.006 0.805
## tcq01 0.924 0.022 42.834 0.000 0.861 0.692
## tcq09 0.851 0.021 41.126 0.000 0.793 0.710
## tcq11 1.013 0.023 44.644 0.000 0.943 0.754
## tcq14 0.994 0.022 44.269 0.000 0.925 0.805
## tcq03 0.831 0.020 40.611 0.000 0.774 0.626
## tcq07 0.898 0.021 42.234 0.000 0.836 0.676
## tcq10 0.772 0.020 39.005 0.000 0.719 0.540
## tcq13 1.037 0.023 45.074 0.000 0.965 0.750
## tcq04 0.938 0.022 43.132 0.000 0.873 0.744
## tcq06 1.020 0.023 44.774 0.000 0.950 0.768
## tcq08 0.916 0.021 42.653 0.000 0.853 0.784
## tcq15 1.031 0.023 44.979 0.000 0.960 0.775
##
## Regressions:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## SIQS ~
## LT 0.650 0.025 25.834 0.000 0.384 0.384
## LTc 0.327 0.011 30.500 0.000 0.377 0.377
## sexo 0.390 0.116 3.368 0.001 0.482 0.238
##
## Covariances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## LT ~~
## LTc 0.121 0.004 28.884 0.000 0.272 0.272
##
## Intercepts:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .siqs01 4.822 0.074 64.785 0.000 4.822 4.291
## .siqs02 4.696 0.071 66.487 0.000 4.696 3.549
## .siqs03 4.675 0.079 59.549 0.000 4.675 3.785
## .siqs04 4.516 0.068 66.317 0.000 4.516 3.682
## .siqs05 4.469 0.074 60.570 0.000 4.469 3.414
## .siqs06 4.923 0.071 68.918 0.000 4.923 4.182
## .siqs07 5.262 0.071 74.080 0.000 5.262 5.675
## .siqs08 5.197 0.073 71.383 0.000 5.197 5.291
## .siqs09 4.906 0.069 70.844 0.000 4.906 4.485
## .tpqp02 5.455 0.053 102.475 0.000 5.455 5.593
## .tpqp05 5.342 0.053 100.352 0.000 5.342 4.712
## .tpqp12 5.008 0.053 94.090 0.000 5.008 3.851
## .tpqp16 5.506 0.053 103.430 0.000 5.506 5.573
## .tpqp01 5.325 0.053 100.034 0.000 5.325 4.782
## .tpqp09 5.325 0.053 100.034 0.000 5.325 4.347
## .tpqp11 5.178 0.053 97.274 0.000 5.178 4.105
## .tpqp14 5.537 0.053 104.014 0.000 5.537 5.761
## .tpqp03 5.232 0.053 98.283 0.000 5.232 4.724
## .tpqp07 5.229 0.053 98.229 0.000 5.229 4.692
## .tpqp10 5.486 0.053 103.059 0.000 5.486 5.414
## .tpqp13 5.192 0.053 97.540 0.000 5.192 4.111
## .tpqp04 5.311 0.053 99.768 0.000 5.311 4.881
## .tpqp06 5.449 0.053 102.369 0.000 5.449 5.384
## .tpqp08 5.331 0.053 100.140 0.000 5.331 4.658
## .tpqp15 5.189 0.053 97.487 0.000 5.189 4.234
## .tpqm02 5.492 0.053 103.165 0.000 5.492 5.782
## .tpqm05 5.585 0.053 104.916 0.000 5.585 6.018
## .tpqm12 5.373 0.053 100.936 0.000 5.373 4.830
## .tpqm16 5.729 0.053 107.623 0.000 5.729 9.370
## .tpqm01 5.446 0.053 102.316 0.000 5.446 5.707
## .tpqm09 5.551 0.053 104.279 0.000 5.551 5.515
## .tpqm11 5.373 0.053 100.936 0.000 5.373 5.228
## .tpqm14 5.619 0.053 105.553 0.000 5.619 6.938
## .tpqm03 5.325 0.053 100.034 0.000 5.325 5.350
## .tpqm07 5.331 0.053 100.140 0.000 5.331 5.149
## .tpqm10 5.633 0.053 105.818 0.000 5.633 7.036
## .tpqm13 5.356 0.053 100.618 0.000 5.356 4.616
## .tpqm04 5.621 0.053 105.606 0.000 5.621 7.132
## .tpqm06 5.582 0.053 104.863 0.000 5.582 6.222
## .tpqm08 5.508 0.053 103.483 0.000 5.508 5.746
## .tpqm15 5.359 0.053 100.671 0.000 5.359 4.881
## .tcq02 4.280 0.053 80.399 0.000 4.280 2.914
## .tcq05 5.124 0.053 96.266 0.000 5.124 4.303
## .tcq12 3.698 0.053 69.466 0.000 3.698 2.282
## .tcq16 5.040 0.053 94.674 0.000 5.040 4.032
## .tcq01 4.833 0.053 90.800 0.000 4.833 3.889
## .tcq09 5.257 0.053 98.760 0.000 5.257 4.711
## .tcq11 5.020 0.053 94.302 0.000 5.020 4.011
## .tcq14 5.073 0.053 95.311 0.000 5.073 4.415
## .tcq03 4.938 0.053 92.763 0.000 4.938 3.997
## .tcq07 4.915 0.053 92.339 0.000 4.915 3.976
## .tcq10 4.562 0.053 85.705 0.000 4.562 3.428
## .tcq13 4.749 0.053 89.208 0.000 4.749 3.690
## .tcq04 5.017 0.053 94.249 0.000 5.017 4.276
## .tcq06 4.952 0.053 93.029 0.000 4.952 4.003
## .tcq08 5.311 0.053 99.768 0.000 5.311 4.881
## .tcq15 5.040 0.053 94.674 0.000 5.040 4.069
## sexo 0.412 0.053 7.748 0.000 0.412 0.837
## .it 0.000 0.000 0.000
## .cc 0.000 0.000 0.000
## .ia 0.000 0.000 0.000
## .SIQS 0.000 0.000 0.000
## .LTp 0.000 0.000 0.000
## .LTm 0.000 0.000 0.000
## LT 0.000 0.000 0.000
## LTc 0.000 0.000 0.000
##
## Variances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .siqs01 0.221 0.070 3.144 0.002 0.221 0.175
## .siqs02 0.921 0.066 13.997 0.000 0.921 0.526
## .siqs03 0.242 0.076 3.191 0.001 0.242 0.159
## .siqs04 0.836 0.065 12.959 0.000 0.836 0.556
## .siqs05 0.745 0.072 10.321 0.000 0.745 0.435
## .siqs06 0.545 0.069 7.927 0.000 0.545 0.393
## .siqs07 0.200 0.066 3.029 0.002 0.200 0.233
## .siqs08 0.229 0.068 3.344 0.001 0.229 0.237
## .siqs09 0.612 0.064 9.555 0.000 0.612 0.512
## .tpqp02 0.553 0.056 9.913 0.000 0.553 0.582
## .tpqp05 0.413 0.059 7.008 0.000 0.413 0.322
## .tpqp12 0.591 0.061 9.764 0.000 0.591 0.350
## .tpqp16 0.439 0.057 7.738 0.000 0.439 0.450
## .tpqp01 0.673 0.057 11.832 0.000 0.673 0.543
## .tpqp09 0.467 0.060 7.779 0.000 0.467 0.312
## .tpqp11 0.592 0.060 9.898 0.000 0.592 0.372
## .tpqp14 0.295 0.057 5.152 0.000 0.295 0.320
## .tpqp03 0.863 0.056 15.524 0.000 0.863 0.703
## .tpqp07 0.692 0.057 12.179 0.000 0.692 0.557
## .tpqp10 0.329 0.058 5.699 0.000 0.329 0.321
## .tpqp13 0.483 0.061 7.959 0.000 0.483 0.303
## .tpqp04 0.454 0.058 7.826 0.000 0.454 0.384
## .tpqp06 0.392 0.057 6.835 0.000 0.392 0.383
## .tpqp08 0.483 0.059 8.236 0.000 0.483 0.369
## .tpqp15 0.553 0.060 9.292 0.000 0.553 0.368
## .tpqm02 0.564 0.056 10.121 0.000 0.564 0.626
## .tpqm05 0.327 0.057 5.700 0.000 0.327 0.379
## .tpqm12 0.497 0.059 8.437 0.000 0.497 0.402
## .tpqm16 0.208 0.054 3.811 0.000 0.208 0.555
## .tpqm01 0.557 0.056 9.974 0.000 0.557 0.612
## .tpqm09 0.297 0.059 5.062 0.000 0.297 0.293
## .tpqm11 0.353 0.059 6.015 0.000 0.353 0.334
## .tpqm14 0.192 0.057 3.378 0.001 0.192 0.292
## .tpqm03 0.724 0.055 13.118 0.000 0.724 0.731
## .tpqm07 0.682 0.056 12.140 0.000 0.682 0.636
## .tpqm10 0.214 0.056 3.798 0.000 0.214 0.334
## .tpqm13 0.506 0.060 8.476 0.000 0.506 0.376
## .tpqm04 0.325 0.055 5.859 0.000 0.325 0.523
## .tpqm06 0.306 0.057 5.363 0.000 0.306 0.380
## .tpqm08 0.345 0.058 5.994 0.000 0.345 0.376
## .tpqm15 0.494 0.059 8.426 0.000 0.494 0.410
## .tcq02 1.290 0.060 21.580 0.000 1.290 0.598
## .tcq05 0.668 0.059 11.356 0.000 0.668 0.471
## .tcq12 1.965 0.058 33.780 0.000 1.965 0.749
## .tcq16 0.550 0.061 9.029 0.000 0.550 0.352
## .tcq01 0.804 0.059 13.675 0.000 0.804 0.521
## .tcq09 0.617 0.058 10.652 0.000 0.617 0.495
## .tcq11 0.676 0.060 11.279 0.000 0.676 0.432
## .tcq14 0.464 0.060 7.782 0.000 0.464 0.352
## .tcq03 0.927 0.058 16.068 0.000 0.927 0.608
## .tcq07 0.830 0.058 14.195 0.000 0.830 0.543
## .tcq10 1.254 0.057 21.973 0.000 1.254 0.708
## .tcq13 0.724 0.060 12.019 0.000 0.724 0.437
## .tcq04 0.614 0.059 10.412 0.000 0.614 0.446
## .tcq06 0.628 0.060 10.457 0.000 0.628 0.410
## .tcq08 0.456 0.059 7.773 0.000 0.456 0.385
## .tcq15 0.612 0.060 10.160 0.000 0.612 0.399
## .it 0.389 0.038 10.205 0.000 0.374 0.374
## .cc 0.233 0.030 7.898 0.000 0.349 0.349
## .ia 0.126 0.035 3.603 0.000 0.192 0.192
## .SIQS 0.376 0.026 14.522 0.000 0.576 0.576
## .LTp 0.150 0.009 17.255 0.000 0.377 0.377
## .LTm 0.110 0.008 13.742 0.000 0.326 0.326
## LT 0.228 0.012 18.832 0.000 1.000 1.000
## LTc 0.867 0.027 31.889 0.000 1.000 1.000
## sexo 0.243 0.053 4.565 0.000 0.243 1.000# [Grupo 1] SIQS = 0 + .384 * LT
# [Grupo 2] SIQS = 0 + .312 * LT
graf <- data.frame(edad=rep(1:2,2),
"LT (std)"=c(rep(min(c(scale(datos$tpqm_LT),scale(datos$tpqp_LT)),na.rm = T),2),rep(max(c(scale(datos$tpqm_LT),scale(datos$tpqp_LT)),na.rm = T),2)),
"siqs (std)"=rep(NA,4),
check.names = F)
for(i in 1:4){
graf$`siqs (std)`[i] <- ifelse(graf$edad[i]==1,
.384*graf$`LT (std)`[i],
.312*graf$`LT (std)`[i])
}
p3 <- graf %>%
ggplot(aes(x=`LT (std)`, y=`siqs (std)`, color=as.factor(edad), group=as.factor(edad)))+
geom_line()+
scale_color_manual(name="Age stage",
values = 2:3,
labels = c("early",
"late"))+
ylim(c(min(scale(datos$total_IS)),max(scale(datos$total_IS))))
# [Grupo 1] SIQS = 0 + .377 * LTc
# [Grupo 2] SIQS = 0 + .332 * LTc
graf <- data.frame(edad=rep(1:2,2),
"LTc (std)"=c(rep(min(scale(datos$tcq_LT),na.rm = T),2),rep(max(scale(datos$tcq_LT),na.rm = T),2)),
"siqs (std)"=rep(NA,4),
check.names = F)
for(i in 1:4){
graf$`siqs (std)`[i] <- ifelse(graf$edad[i]==1,
.377*graf$`LTc (std)`[i],
.332*graf$`LTc (std)`[i])
}
p4 <- graf %>%
ggplot(aes(x=`LTc (std)`, y=`siqs (std)`, color=as.factor(edad), group=as.factor(edad)))+
geom_line()+
scale_color_manual(name="Age stage",
values = 2:3,
labels = c("early",
"late"))+
ylim(c(min(scale(datos$total_IS)),max(scale(datos$total_IS))))
grid.arrange(p3, p4, nrow=1)
Conclusión
El mejor modelo en esta línea de análisis es el más complejo, que incluye el estilo de liderazgo del coach y el estilo de parentalidad en el hogar (compuesto por cada estilo de parentalidad por separado) en la predicción de la identidad social. La utilización de este modelo permite interpretar el funcionamiendo de cada una de las escalas, el peso de cada figura paterna en la concepción de parentalidad transformacional y el peso de cada contexto social en la predicción de la identidad social.
indicadores <- bind_rows(sem.config.sexo %>%
fitmeasures(c("chisq","df", "cfi","tli","rmsea","srmr")),
sem.config.edad %>%
fitmeasures(c("chisq","df", "cfi","tli","rmsea","srmr")))
indicadores <- cbind("modelo"=c("sexo","edad"), indicadores)
indicadores[,4:7] <- round(indicadores[,4:7],3)
kable(indicadores,
"html",
booktabs = T,
caption = "Indicadores para cada modelo") %>%
kable_styling(full_width = F,
position = "center", font_size = 12)| modelo | chisq | df | cfi | tli | rmsea | srmr |
|---|---|---|---|---|---|---|
| sexo | 6591.747 | 3174 | 0.981 | 0.980 | 0.057 | 0.057 |
| edad | 5373.037 | 3174 | 0.987 | 0.987 | 0.046 | 0.057 |
semPaths(# Argumentos globales
sem.config.sexo, what="diagram", whatLabels="std",layout="tree2", residuals = F,
rotation = 2, width=50, height=35,exoVar = F,
# Etiquetas
curvePivot=T, curvature = .5
)
Nota. 1=Varones; 2=Mujeres.
Tanto en varones como en mujeres, el peso del estilo transformacional de la madre (0.91 y 0.87) es mayor que el peso del estilo transformacional del padre (0.59 y 0.77) a la hora de evaluar el liderazgo en el contexto familiar.
semPaths(# Argumentos globales
sem.config.edad, what="diagram", whatLabels="std",layout="tree2", residuals = F,
rotation = 2, width=50, height=35,exoVar = F,
# Etiquetas
curvePivot=T, curvature = .5
)
Nota. 1=Late; 2=Early.
A su vez, ambas etapas vitales también ponderan más el liderazgo de la madre (0.99 y 0.82) que el del padre (0.64 y 0.79).
grid.arrange(p1, p2, nrow=1)
Respecto al sexo, en las mujeres la figura de los padres tiene mayor influencia sobre su identidad social, mientras que la figura del entrenador se relaciona más intensamente con la identidad social en varones.
grid.arrange(p3, p4, nrow=1)
El efecto del estilo de liderazgo del coach resulta significativo pero bastante constante entre las distintas etapas de la adolescencia, mientras que el estilo de parentalidad de las familias tiene una relación más intensa con la identidad social cuando los adolescentes son más jóvenes.
SEM y Reg. Lineal
Otra opción es armar el modelo completo en SEM, sacar conclusiones de las relaciones controlando por sexo y edad, pero sin la interacción. Y finalmente, justificado en la imposibilidad de identificar el modelo SEM con la interacción, hacer una regresión lineal múltiple para evaluar la interacción.
Coach
modelo <- '
# Identidad social
it =~ siqs01 + siqs02 + siqs03
cc =~ siqs04 + siqs05 + siqs06
ia =~ siqs07 + siqs08 + siqs09
SIQS =~ it + cc + ia
# Liderazgo transformacional (coach)
LTc =~ tcq02 + tcq05 + tcq12 + tcq16 + tcq01 + tcq09 + tcq11 + tcq14 + tcq03 + tcq07 + tcq10 + tcq13 + tcq04 + tcq06 + tcq08 + tcq15
# REGRESIONES
SIQS ~ LTc + sexo + edad
'
# Configural model
sem.coach <- datos %>%
sem(modelo, .,
estimator="ULS")
# sem.coach %>%
# summary(fit.measures=T, standardized=T)
semPaths(# Argumentos globales
sem.coach, what="diagram", whatLabels="std",layout="tree3", rotation = 2, width=50, height=35,exoVar = F,
# Etiquetas
curvePivot=T, curvature = .5
)
mod <- datos %>%
lm(total_IS~tcq_LT*(sexo+edaddicot),.)
mod %>%
summary()##
## Call:
## lm(formula = total_IS ~ tcq_LT * (sexo + edaddicot), data = .)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.81294 -0.48916 0.09818 0.53117 1.95378
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.65285 0.49321 5.379 1.05e-07 ***
## tcq_LT 0.48089 0.10101 4.761 2.38e-06 ***
## sexo -0.42005 0.33909 -1.239 0.2159
## edaddicot 0.58013 0.32104 1.807 0.0712 .
## tcq_LT:sexo 0.12685 0.06935 1.829 0.0678 .
## tcq_LT:edaddicot -0.14046 0.06644 -2.114 0.0349 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.7559 on 650 degrees of freedom
## Multiple R-squared: 0.1679, Adjusted R-squared: 0.1615
## F-statistic: 26.23 on 5 and 650 DF, p-value: < 2.2e-16graf <- data.frame(sexo = rep(1, 4),
edaddicot = rep(1:2, 2),
tcq_LT = c(rep(min(datos$tcq_LT, na.rm = TRUE), 2), rep(max(datos$tcq_LT, na.rm = TRUE), 2)),
total_IS = rep(NA, 4),
check.names = FALSE)
graf$total_IS <- predict(mod, newdata = graf)
p5 <- graf %>%
ggplot(aes(x=tcq_LT, y=total_IS, color=as.factor(edaddicot), group=as.factor(edaddicot)))+
geom_line()+
scale_color_manual(name="Age stage",
values = 2:3,
labels = c("early",
"late"))+
ylim(c(min(datos$total_IS), max(datos$total_IS)))
p5
El entrenador tiene más peso en los early que en los late.
Madre
modelo <- '
# Identidad social
it =~ siqs01 + siqs02 + siqs03
cc =~ siqs04 + siqs05 + siqs06
ia =~ siqs07 + siqs08 + siqs09
SIQS =~ it + cc + ia
# Liderazgo transformacional (coach)
LTm =~ tpqm02 + tpqm05 + tpqm12 + tpqm16 + tpqm01 + tpqm09 + tpqm11 + tpqm14 + tpqm03 + tpqm07 + tpqm10 + tpqm13 + tpqm04 + tpqm06 + tpqm08 + tpqm15
# REGRESIONES
SIQS ~ LTm + sexo + edad
'
# Configural model
sem.mom <- datos %>%
sem(modelo, .,
estimator="ULS")
# sem.coach %>%
# summary(fit.measures=T, standardized=T)
semPaths(# Argumentos globales
sem.mom, what="diagram", whatLabels="std",layout="tree3", rotation = 2, width=50, height=35,exoVar = F,
# Etiquetas
curvePivot=T, curvature = .5
)
mod <- datos %>%
lm(total_IS~tpqm_LT*(sexo+edaddicot),.)
mod %>%
summary()##
## Call:
## lm(formula = total_IS ~ tpqm_LT * (sexo + edaddicot), data = .)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.6623 -0.5137 0.1154 0.5871 1.8389
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.12015 0.73566 2.882 0.00408 **
## tpqm_LT 0.53433 0.13294 4.019 6.52e-05 ***
## sexo 0.63647 0.52745 1.207 0.22799
## edaddicot 0.41303 0.46589 0.887 0.37565
## tpqm_LT:sexo -0.07462 0.09466 -0.788 0.43082
## tpqm_LT:edaddicot -0.10540 0.08424 -1.251 0.21131
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.77 on 650 degrees of freedom
## Multiple R-squared: 0.1367, Adjusted R-squared: 0.1301
## F-statistic: 20.59 on 5 and 650 DF, p-value: < 2.2e-16graf <- data.frame(sexo = rep(1, 4),
edaddicot = rep(1:2, 2),
tpqm_LT = c(rep(min(datos$tpqm_LT, na.rm = TRUE), 2), rep(max(datos$tpqm_LT, na.rm = TRUE), 2)),
total_IS = rep(NA, 4),
check.names = FALSE)
graf$total_IS <- predict(mod, newdata = graf)
# graf %>%
# ggplot(aes(x=tpqm_LT, y=total_IS, color=as.factor(edaddicot), group=as.factor(edaddicot)))+
# geom_line()+
# scale_color_manual(name="Age stage",
# values = 2:3,
# labels = c("early",
# "late"))El efecto de la madre es constante, más allá de la edad y del sexo.
Padre
modelo <- '
# Identidad social
it =~ siqs01 + siqs02 + siqs03
cc =~ siqs04 + siqs05 + siqs06
ia =~ siqs07 + siqs08 + siqs09
SIQS =~ it + cc + ia
# Liderazgo transformacional (coach)
LTp =~ tpqp02 + tpqp05 + tpqp12 + tpqp16 + tpqp01 + tpqp09 + tpqp11 + tpqp14 + tpqp03 + tpqp07 + tpqp10 + tpqp13 + tpqp04 + tpqp06 + tpqp08 + tpqp15
# REGRESIONES
SIQS ~ LTp + sexo + edad
'
# Configural model
sem.dad <- datos %>%
sem(modelo, .,
estimator="ULS")
# sem.coach %>%
# summary(fit.measures=T, standardized=T)
semPaths(# Argumentos globales
sem.dad, what="diagram", whatLabels="std",layout="tree3", rotation = 2, width=50, height=35,exoVar = F,
# Etiquetas
curvePivot=T, curvature = .5
)
mod <- datos %>%
lm(total_IS~tpqp_LT*(sexo+edaddicot),.)
mod %>%
summary()##
## Call:
## lm(formula = total_IS ~ tpqp_LT * (sexo + edaddicot), data = .)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.8162 -0.5043 0.1139 0.6073 1.6018
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.15053 0.60338 3.564 0.000392 ***
## tpqp_LT 0.54498 0.11196 4.868 1.42e-06 ***
## sexo 0.95358 0.40412 2.360 0.018586 *
## edaddicot 0.83389 0.37523 2.222 0.026605 *
## tpqp_LT:sexo -0.13339 0.07445 -1.792 0.073673 .
## tpqp_LT:edaddicot -0.18766 0.06975 -2.690 0.007318 **
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.7825 on 650 degrees of freedom
## Multiple R-squared: 0.1084, Adjusted R-squared: 0.1016
## F-statistic: 15.81 on 5 and 650 DF, p-value: 1.037e-14graf <- data.frame(sexo = rep(1, 4),
edaddicot = rep(1:2, 2),
tpqp_LT = c(rep(min(datos$tpqp_LT, na.rm = TRUE), 2), rep(max(datos$tpqp_LT, na.rm = TRUE), 2)),
total_IS = rep(NA, 4),
check.names = FALSE)
graf$total_IS <- predict(mod, newdata = graf)
p6 <- graf %>%
ggplot(aes(x=tpqp_LT, y=total_IS, color=as.factor(edaddicot), group=as.factor(edaddicot)))+
geom_line()+
scale_color_manual(name="Age stage",
values = 2:3,
labels = c("early",
"late"))+
ylim(c(min(datos$total_IS), max(datos$total_IS)))
p6
El efecto del padre (como el del coach) es mucho más marcado en early que en late.
Madre y Padre
modelo <- '
# Identidad social
it =~ siqs01 + siqs02 + siqs03
cc =~ siqs04 + siqs05 + siqs06
ia =~ siqs07 + siqs08 + siqs09
SIQS =~ it + cc + ia
# Liderazgo transformacional (coach)
LTp =~ tpqp02 + tpqp05 + tpqp12 + tpqp16 + tpqp01 + tpqp09 + tpqp11 + tpqp14 + tpqp03 + tpqp07 + tpqp10 + tpqp13 + tpqp04 + tpqp06 + tpqp08 + tpqp15
LTm =~ tpqm02 + tpqm05 + tpqm12 + tpqm16 + tpqm01 + tpqm09 + tpqm11 + tpqm14 + tpqm03 + tpqm07 + tpqm10 + tpqm13 + tpqm04 + tpqm06 + tpqm08 + tpqm15
LT =~ LTm+LTp
# REGRESIONES
SIQS ~ LT + sexo + edad
'
# Configural model
sem.parents <- datos %>%
sem(modelo, .,
estimator="ULS")
# sem.coach %>%
# summary(fit.measures=T, standardized=T)
semPaths(# Argumentos globales
sem.parents, what="diagram", whatLabels="std",layout="tree3", rotation = 2, width=50, height=35,exoVar = F,
# Etiquetas
curvePivot=T, curvature = .5
)
mod <- datos %>%
lm(total_IS~(tpqp_LT+tpqm_LT)*(sexo+edaddicot),.)
mod %>%
summary()##
## Call:
## lm(formula = total_IS ~ (tpqp_LT + tpqm_LT) * (sexo + edaddicot),
## data = .)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.7226 -0.4897 0.1163 0.5919 1.8810
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.697499 0.748524 2.268 0.0237 *
## tpqp_LT 0.410799 0.144470 2.843 0.0046 **
## tpqm_LT 0.213313 0.173772 1.228 0.2201
## sexo 0.661874 0.541299 1.223 0.2219
## edaddicot 0.631328 0.476135 1.326 0.1853
## tpqp_LT:sexo -0.087224 0.088968 -0.980 0.3273
## tpqp_LT:edaddicot -0.202599 0.086557 -2.341 0.0196 *
## tpqm_LT:sexo 0.004825 0.113694 0.042 0.9662
## tpqm_LT:edaddicot 0.051532 0.105971 0.486 0.6269
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.7664 on 647 degrees of freedom
## Multiple R-squared: 0.1487, Adjusted R-squared: 0.1381
## F-statistic: 14.12 on 8 and 647 DF, p-value: < 2.2e-16graf <- data.frame(sexo = rep(1, 4),
edaddicot = rep(1:2, 2),
tpqm_LT = rep(mean(datos$tpqm_LT),4),
tpqp_LT = c(rep(min(datos$tpqp_LT, na.rm = TRUE), 2), rep(max(datos$tpqp_LT, na.rm = TRUE), 2)),
total_IS = rep(NA, 4),
check.names = FALSE)
graf$total_IS <- predict(mod, newdata = graf)
graf %>%
ggplot(aes(x=tpqp_LT, y=total_IS, color=as.factor(edaddicot), group=as.factor(edaddicot)))+
geom_line()+
scale_color_manual(name="Age stage",
values = 2:3,
labels = c("early",
"late"))+
ylim(c(min(datos$total_IS), max(datos$total_IS)))
Cuando se incluyen madre, padre, edad y sexo (con sus interacciones) sólo resulta significativo el padre y su interacción con la edad. Dejando las variables sexo y madre fijas, da este resultado medio contraintuitivo para los lates.
Contexto Total
modelo <- '
# Identidad social
it =~ siqs01 + siqs02 + siqs03
cc =~ siqs04 + siqs05 + siqs06
ia =~ siqs07 + siqs08 + siqs09
SIQS =~ it + cc + ia
# Liderazgo transformacional (coach)
LTp =~ tpqp02 + tpqp05 + tpqp12 + tpqp16 + tpqp01 + tpqp09 + tpqp11 + tpqp14 + tpqp03 + tpqp07 + tpqp10 + tpqp13 + tpqp04 + tpqp06 + tpqp08 + tpqp15
LTm =~ tpqm02 + tpqm05 + tpqm12 + tpqm16 + tpqm01 + tpqm09 + tpqm11 + tpqm14 + tpqm03 + tpqm07 + tpqm10 + tpqm13 + tpqm04 + tpqm06 + tpqm08 + tpqm15
LT =~ LTm+LTp
LTc =~ tcq02 + tcq05 + tcq12 + tcq16 + tcq01 + tcq09 + tcq11 + tcq14 + tcq03 + tcq07 + tcq10 + tcq13 + tcq04 + tcq06 + tcq08 + tcq15
# REGRESIONES
SIQS ~ LTc + LT + sexo + edad
'
# Configural model
sem.todes <- datos %>%
sem(modelo, .,
estimator="ULS")
# sem.coach %>%
# summary(fit.measures=T, standardized=T)
semPaths(# Argumentos globales
sem.todes, what="diagram", whatLabels="std",layout="tree3", rotation = 2, width=50, height=35,exoVar = F,
# Etiquetas
curvePivot=T, curvature = .5
)
mod <- datos %>%
lm(total_IS~(tcq_LT+tpqp_LT+tpqm_LT)*(sexo+edaddicot),.)
mod %>%
summary()##
## Call:
## lm(formula = total_IS ~ (tcq_LT + tpqp_LT + tpqm_LT) * (sexo +
## edaddicot), data = .)
##
## Residuals:
## Min 1Q Median 3Q Max
## -2.78482 -0.48365 0.07396 0.52589 1.99956
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.53189 0.78004 0.682 0.495567
## tcq_LT 0.37917 0.09913 3.825 0.000143 ***
## tpqp_LT 0.32115 0.13735 2.338 0.019681 *
## tpqm_LT 0.16750 0.16470 1.017 0.309553
## sexo 0.25546 0.56309 0.454 0.650212
## edaddicot 0.86484 0.50464 1.714 0.087050 .
## tcq_LT:sexo 0.13890 0.06749 2.058 0.039994 *
## tcq_LT:edaddicot -0.10208 0.06481 -1.575 0.115698
## tpqp_LT:sexo -0.08101 0.08425 -0.962 0.336638
## tpqp_LT:edaddicot -0.15893 0.08212 -1.935 0.053376 .
## tpqm_LT:sexo -0.06202 0.10820 -0.573 0.566738
## tpqm_LT:edaddicot 0.06947 0.10052 0.691 0.489755
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.7241 on 644 degrees of freedom
## Multiple R-squared: 0.2435, Adjusted R-squared: 0.2306
## F-statistic: 18.85 on 11 and 644 DF, p-value: < 2.2e-16graf <- data.frame(edaddicot = rep(1, 4),
sexo = rep(0:1, 2),
tpqm_LT = rep(mean(datos$tpqm_LT),4),
tpqp_LT = rep(mean(datos$tpqp_LT),4),
tcq_LT = c(rep(min(datos$tcq_LT, na.rm = TRUE), 2), rep(max(datos$tcq_LT, na.rm = TRUE), 2)),
total_IS = rep(NA, 4),
check.names = FALSE)
graf$total_IS <- predict(mod, newdata = graf)
graf %>%
ggplot(aes(x=tcq_LT, y=total_IS, color=as.factor(sexo), group=as.factor(sexo)))+
geom_line()+
scale_color_manual(name="sex",
values = 2:3,
labels = c("female",
"male"))+
ylim(c(min(datos$total_IS), max(datos$total_IS)))
Cuando se incluyen coach, madre, padre, edad y sexo (con sus interacciones) sólo resulta significativo el padre, el coach y la interacción de coach con sexo.
Conclusión
Creo que se podría reportar el sem completo para interpretarlo, pero los modelos que evaluan interacción los haría por separado.
indicadores <- sem.todes %>%
fitmeasures(c("chisq","df", "cfi","tli","rmsea","srmr"))
indicadores[3:6] <- round(indicadores[3:6],3)
kable(indicadores,
"html",
booktabs = T,
caption = "Indicadores para el modelo completo") %>%
kable_styling(full_width = F,
position = "center", font_size = 12)| x | |
|---|---|
| chisq | 4679.133 |
| df | 1643.000 |
| cfi | 0.982 |
| tli | 0.982 |
| rmsea | 0.053 |
| srmr | 0.050 |
semPaths(# Argumentos globales
sem.todes, what="diagram", whatLabels="std",layout="tree2", residuals = F,
rotation = 2, width=50, height=35,exoVar = F,
# Etiquetas
curvePivot=T, curvature = .5
)
El peso del estilo transformacional de la madre (0.87) es mayor que el peso del estilo transformacional del padre (0.73) a la hora de evaluar el liderazgo en el contexto familiar. En la predicción de la identidad social, ambos contextos tienen importancia similar (coach: 0.37, familia: 0.35).
grid.arrange(p5, p6, nrow=1)
Nota. tcq_LT = Liderazgo Coach, tpqp_LT = Liderazgo Padre.
La etapa vital resultó un moderador significativo en el modelo del padre y del coach. El liderazgo de ambas figuras presenta una relación más fuerte con la identidad social entre los early que entre los late. En el caso del padre, esa relación en lates parece casi nula.
SEM con moderación
La última opción no la recomiendo realmente, pero es también una opción. Se podrían ajustar los modelos sin incluir la covarianza de los errores. En las limitaciones se podría indicar que debido a problemas con la identificación del modelo no fue posible ajustarlo con la covarianza de los errores entre los términos de la interacción y las variables originales y que, por lo tanto, la estimación de los parámentros de la interacción podrían estar sesgados.
Coach
Sexo
datos_mod <- semTools::indProd(datos, var1 = 24:39, var2 = c(2,4), match = FALSE,
meanC = TRUE, residualC = FALSE, doubleMC = TRUE)
modelo <- '
# Identidad social
it =~ siqs01 + siqs02 + siqs03
cc =~ siqs04 + siqs05 + siqs06
ia =~ siqs07 + siqs08 + siqs09
SIQS =~ it + cc + ia
# Liderazgo transformacional (coach)
LTc =~ tcq02 + tcq05 + tcq12 + tcq16 + tcq01 + tcq09 + tcq11 + tcq14 + tcq03 + tcq07 + tcq10 + tcq13 + tcq04 + tcq06 + tcq08 + tcq15
# MODERADORES
int.sex_LTc =~ tcq01.sexo + tcq02.sexo + tcq03.sexo + tcq04.sexo + tcq05.sexo + tcq06.sexo + tcq07.sexo + tcq08.sexo + tcq09.sexo + tcq10.sexo + tcq11.sexo + tcq12.sexo + tcq13.sexo + tcq14.sexo + tcq15.sexo + tcq16.sexo
# REGRESIONES
SIQS ~ LTc + sexo + edaddicot + int.sex_LTc
'
ajuste_LTc <- datos_mod %>%
sem(modelo, ., estimator="ULS")
# ajuste_LTc %>%
# summary(fit.measures = TRUE, standardized = TRUE)
semPaths(# Argumentos globales
ajuste_LTc, what="diagram", whatLabels="std",layout="tree3", rotation = 2, width=50, height=35,exoVar = F,
# Etiquetas
curvePivot=T, curvature = .5
)
Edad
modelo <- '
# Identidad social
it =~ siqs01 + siqs02 + siqs03
cc =~ siqs04 + siqs05 + siqs06
ia =~ siqs07 + siqs08 + siqs09
SIQS =~ it + cc + ia
# Liderazgo transformacional (coach)
LTc =~ tcq02 + tcq05 + tcq12 + tcq16 + tcq01 + tcq09 + tcq11 + tcq14 + tcq03 + tcq07 + tcq10 + tcq13 + tcq04 + tcq06 + tcq08 + tcq15
# MODERADORES
int.age_LTc =~ tcq01.edaddicot + tcq02.edaddicot + tcq03.edaddicot + tcq04.edaddicot + tcq05.edaddicot + tcq06.edaddicot + tcq07.edaddicot + tcq08.edaddicot + tcq09.edaddicot + tcq10.edaddicot + tcq11.edaddicot + tcq12.edaddicot + tcq13.edaddicot + tcq14.edaddicot + tcq15.edaddicot + tcq16.edaddicot
# REGRESIONES
SIQS ~ LTc + sexo + edaddicot + int.age_LTc
'
ajuste_LTc <- datos_mod %>%
sem(modelo, ., estimator="ULS")
# ajuste_LTc %>%
# summary(fit.measures = TRUE, standardized = TRUE)
semPaths(# Argumentos globales
ajuste_LTc, what="diagram", whatLabels="std",layout="tree3", rotation = 2, width=50, height=35,exoVar = F,
# Etiquetas
curvePivot=T, curvature = .5
)
Sexo y edad
modelo <- '
# Identidad social
it =~ siqs01 + siqs02 + siqs03
cc =~ siqs04 + siqs05 + siqs06
ia =~ siqs07 + siqs08 + siqs09
SIQS =~ it + cc + ia
# Liderazgo transformacional (coach)
LTc =~ tcq02 + tcq05 + tcq12 + tcq16 + tcq01 + tcq09 + tcq11 + tcq14 + tcq03 + tcq07 + tcq10 + tcq13 + tcq04 + tcq06 + tcq08 + tcq15
# MODERADORES
int.sex_LTc =~ tcq01.sexo + tcq02.sexo + tcq03.sexo + tcq04.sexo + tcq05.sexo + tcq06.sexo + tcq07.sexo + tcq08.sexo + tcq09.sexo + tcq10.sexo + tcq11.sexo + tcq12.sexo + tcq13.sexo + tcq14.sexo + tcq15.sexo + tcq16.sexo
int.age_LTc =~ tcq01.edaddicot + tcq02.edaddicot + tcq03.edaddicot + tcq04.edaddicot + tcq05.edaddicot + tcq06.edaddicot + tcq07.edaddicot + tcq08.edaddicot + tcq09.edaddicot + tcq10.edaddicot + tcq11.edaddicot + tcq12.edaddicot + tcq13.edaddicot + tcq14.edaddicot + tcq15.edaddicot + tcq16.edaddicot
# REGRESIONES
SIQS ~ LTc + sexo + edaddicot + int.sex_LTc + int.age_LTc
'
ajuste_LTc <- datos_mod %>%
sem(modelo, ., estimator="ULS")
ajuste_LTc %>%
summary(fit.measures = TRUE, standardized = TRUE)## lavaan 0.6.16 ended normally after 100 iterations
##
## Estimator ULS
## Optimization method NLMINB
## Number of model parameters 128
##
## Number of observations 656
##
## Model Test User Model:
##
## Test statistic 1818.133
## Degrees of freedom 1642
## P-value (Unknown) NA
##
## Model Test Baseline Model:
##
## Test statistic 76859.422
## Degrees of freedom 1710
## P-value NA
##
## User Model versus Baseline Model:
##
## Comparative Fit Index (CFI) 0.998
## Tucker-Lewis Index (TLI) 0.998
##
## Root Mean Square Error of Approximation:
##
## RMSEA 0.013
## 90 Percent confidence interval - lower 0.008
## 90 Percent confidence interval - upper 0.016
## P-value H_0: RMSEA <= 0.050 1.000
## P-value H_0: RMSEA >= 0.080 0.000
##
## Standardized Root Mean Square Residual:
##
## SRMR 0.045
##
## Parameter Estimates:
##
## Standard errors Standard
## Information Expected
## Information saturated (h1) model Unstructured
##
## Latent Variables:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## it =~
## siqs01 1.000 0.989 0.898
## siqs02 0.792 0.023 34.447 0.000 0.783 0.625
## siqs03 1.131 0.029 38.524 0.000 1.119 0.923
## cc =~
## siqs04 1.000 0.931 0.713
## siqs05 1.095 0.032 34.611 0.000 1.019 0.750
## siqs06 0.935 0.028 33.141 0.000 0.870 0.734
## ia =~
## siqs07 1.000 0.830 0.863
## siqs08 1.053 0.031 34.088 0.000 0.874 0.897
## siqs09 0.939 0.029 32.688 0.000 0.779 0.709
## SIQS =~
## it 1.000 0.797 0.797
## cc 0.856 0.027 31.567 0.000 0.725 0.725
## ia 0.906 0.029 31.676 0.000 0.860 0.860
## LTc =~
## tcq02 1.000 0.992 0.691
## tcq05 0.965 0.016 61.239 0.000 0.958 0.771
## tcq12 0.743 0.014 53.623 0.000 0.737 0.472
## tcq16 1.072 0.017 63.938 0.000 1.064 0.803
## tcq01 0.961 0.016 61.124 0.000 0.954 0.757
## tcq09 0.879 0.015 58.665 0.000 0.873 0.703
## tcq11 0.965 0.016 61.248 0.000 0.958 0.743
## tcq14 0.905 0.015 59.479 0.000 0.898 0.764
## tcq03 0.797 0.014 55.763 0.000 0.790 0.642
## tcq07 0.774 0.014 54.877 0.000 0.768 0.646
## tcq10 0.632 0.013 48.565 0.000 0.627 0.470
## tcq13 0.915 0.015 59.780 0.000 0.908 0.731
## tcq04 0.966 0.016 61.256 0.000 0.958 0.777
## tcq06 0.990 0.016 61.921 0.000 0.983 0.786
## tcq08 0.873 0.015 58.445 0.000 0.866 0.772
## tcq15 0.931 0.015 60.263 0.000 0.924 0.758
## int.sex_LTc =~
## tcq01.sexo 1.000 0.447 0.732
## tcq02.sexo 1.052 0.070 15.016 0.000 0.470 0.663
## tcq03.sexo 0.878 0.064 13.816 0.000 0.393 0.654
## tcq04.sexo 1.034 0.069 14.909 0.000 0.462 0.763
## tcq05.sexo 1.034 0.069 14.909 0.000 0.462 0.758
## tcq06.sexo 1.038 0.070 14.931 0.000 0.464 0.751
## tcq07.sexo 0.813 0.061 13.268 0.000 0.364 0.625
## tcq08.sexo 0.965 0.067 14.463 0.000 0.432 0.795
## tcq09.sexo 0.925 0.065 14.175 0.000 0.414 0.682
## tcq10.sexo 0.598 0.055 10.967 0.000 0.268 0.405
## tcq11.sexo 1.024 0.069 14.847 0.000 0.458 0.722
## tcq12.sexo 0.813 0.061 13.271 0.000 0.364 0.470
## tcq13.sexo 0.992 0.068 14.646 0.000 0.444 0.736
## tcq14.sexo 1.003 0.068 14.717 0.000 0.449 0.796
## tcq15.sexo 1.043 0.070 14.962 0.000 0.466 0.779
## tcq16.sexo 1.170 0.075 15.645 0.000 0.523 0.822
## int.age_LTc =~
## tcq01.edaddict 1.000 0.465 0.743
## tcq02.edaddict 1.056 0.072 14.625 0.000 0.491 0.694
## tcq03.edaddict 0.839 0.064 13.161 0.000 0.390 0.642
## tcq04.edaddict 1.047 0.072 14.575 0.000 0.487 0.789
## tcq05.edaddict 1.055 0.072 14.620 0.000 0.491 0.795
## tcq06.edaddict 1.054 0.072 14.614 0.000 0.490 0.790
## tcq07.edaddict 0.823 0.063 13.035 0.000 0.383 0.651
## tcq08.edaddict 0.934 0.067 13.872 0.000 0.434 0.778
## tcq09.edaddict 0.920 0.067 13.777 0.000 0.428 0.689
## tcq10.edaddict 0.644 0.057 11.275 0.000 0.299 0.451
## tcq11.edaddict 1.018 0.071 14.406 0.000 0.473 0.744
## tcq12.edaddict 0.778 0.062 12.642 0.000 0.362 0.469
## tcq13.edaddict 0.920 0.067 13.777 0.000 0.428 0.700
## tcq14.edaddict 0.931 0.067 13.852 0.000 0.433 0.744
## tcq15.edaddict 0.982 0.069 14.188 0.000 0.457 0.756
## tcq16.edaddict 1.143 0.076 15.064 0.000 0.532 0.805
##
## Regressions:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## SIQS ~
## LTc 0.372 0.010 38.409 0.000 0.468 0.468
## sexo 0.323 0.085 3.783 0.000 0.409 0.204
## edaddicot -0.194 0.085 -2.282 0.023 -0.246 -0.123
## int.sex_LTc 0.179 0.024 7.508 0.000 0.101 0.101
## int.age_LTc -0.164 0.022 -7.625 0.000 -0.097 -0.097
##
## Covariances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## LTc ~~
## int.sex_LTc -0.091 0.005 -17.624 0.000 -0.205 -0.205
## int.age_LTc 0.010 0.003 3.595 0.000 0.022 0.022
## int.sex_LTc ~~
## int.age_LTc 0.045 0.004 11.227 0.000 0.214 0.214
## sexo ~~
## edaddicot 0.023 0.039 0.600 0.549 0.023 0.094
##
## Variances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .siqs01 0.234 0.055 4.296 0.000 0.234 0.193
## .siqs02 0.955 0.048 19.994 0.000 0.955 0.609
## .siqs03 0.217 0.061 3.575 0.000 0.217 0.148
## .siqs04 0.838 0.054 15.583 0.000 0.838 0.492
## .siqs05 0.808 0.058 13.986 0.000 0.808 0.438
## .siqs06 0.647 0.051 12.562 0.000 0.647 0.461
## .siqs07 0.237 0.051 4.645 0.000 0.237 0.256
## .siqs08 0.186 0.053 3.509 0.000 0.186 0.195
## .siqs09 0.602 0.049 12.239 0.000 0.602 0.498
## .tcq02 1.075 0.045 23.983 0.000 1.075 0.522
## .tcq05 0.627 0.044 14.120 0.000 0.627 0.406
## .tcq12 1.897 0.042 44.977 0.000 1.897 0.778
## .tcq16 0.626 0.046 13.684 0.000 0.626 0.356
## .tcq01 0.676 0.044 15.232 0.000 0.676 0.426
## .tcq09 0.778 0.043 17.901 0.000 0.778 0.505
## .tcq11 0.747 0.044 16.812 0.000 0.747 0.449
## .tcq14 0.574 0.044 13.119 0.000 0.574 0.416
## .tcq03 0.889 0.043 20.841 0.000 0.889 0.587
## .tcq07 0.824 0.042 19.421 0.000 0.824 0.583
## .tcq10 1.389 0.041 33.635 0.000 1.389 0.779
## .tcq13 0.716 0.044 16.330 0.000 0.716 0.465
## .tcq04 0.604 0.044 13.604 0.000 0.604 0.397
## .tcq06 0.597 0.045 13.356 0.000 0.597 0.382
## .tcq08 0.507 0.043 11.688 0.000 0.507 0.404
## .tcq15 0.633 0.044 14.369 0.000 0.633 0.426
## .tcq01.sexo 0.174 0.043 3.999 0.000 0.174 0.465
## .tcq02.sexo 0.282 0.044 6.424 0.000 0.282 0.560
## .tcq03.sexo 0.206 0.042 4.864 0.000 0.206 0.572
## .tcq04.sexo 0.154 0.044 3.513 0.000 0.154 0.418
## .tcq05.sexo 0.158 0.044 3.610 0.000 0.158 0.425
## .tcq06.sexo 0.167 0.044 3.809 0.000 0.167 0.436
## .tcq07.sexo 0.206 0.042 4.913 0.000 0.206 0.609
## .tcq08.sexo 0.108 0.043 2.509 0.012 0.108 0.367
## .tcq09.sexo 0.197 0.043 4.603 0.000 0.197 0.535
## .tcq10.sexo 0.365 0.041 8.999 0.000 0.365 0.836
## .tcq11.sexo 0.192 0.044 4.408 0.000 0.192 0.479
## .tcq12.sexo 0.465 0.042 11.101 0.000 0.465 0.779
## .tcq13.sexo 0.167 0.043 3.845 0.000 0.167 0.458
## .tcq14.sexo 0.116 0.043 2.671 0.008 0.116 0.366
## .tcq15.sexo 0.141 0.044 3.207 0.001 0.141 0.393
## .tcq16.sexo 0.132 0.045 2.919 0.004 0.132 0.325
## .tcq01.edaddict 0.175 0.044 3.954 0.000 0.175 0.448
## .tcq02.edaddict 0.260 0.045 5.771 0.000 0.260 0.518
## .tcq03.edaddict 0.217 0.043 5.086 0.000 0.217 0.588
## .tcq04.edaddict 0.144 0.045 3.202 0.001 0.144 0.377
## .tcq05.edaddict 0.140 0.045 3.113 0.002 0.140 0.368
## .tcq06.edaddict 0.145 0.045 3.225 0.001 0.145 0.376
## .tcq07.edaddict 0.199 0.043 4.670 0.000 0.199 0.576
## .tcq08.edaddict 0.123 0.044 2.814 0.005 0.123 0.394
## .tcq09.edaddict 0.203 0.044 4.661 0.000 0.203 0.526
## .tcq10.edaddict 0.352 0.041 8.530 0.000 0.352 0.797
## .tcq11.edaddict 0.181 0.045 4.051 0.000 0.181 0.446
## .tcq12.edaddict 0.466 0.042 11.027 0.000 0.466 0.780
## .tcq13.edaddict 0.191 0.044 4.383 0.000 0.191 0.510
## .tcq14.edaddict 0.152 0.044 3.475 0.001 0.152 0.447
## .tcq15.edaddict 0.156 0.044 3.530 0.000 0.156 0.428
## .tcq16.edaddict 0.154 0.046 3.333 0.001 0.154 0.352
## .it 0.357 0.031 11.389 0.000 0.365 0.365
## .cc 0.411 0.029 13.975 0.000 0.474 0.474
## .ia 0.179 0.028 6.452 0.000 0.260 0.260
## .SIQS 0.457 0.024 19.136 0.000 0.735 0.735
## LTc 0.985 0.022 44.834 0.000 1.000 1.000
## int.sex_LTc 0.200 0.019 10.536 0.000 1.000 1.000
## int.age_LTc 0.216 0.021 10.293 0.000 1.000 1.000
## sexo 0.248 0.039 6.358 0.000 0.248 1.000
## edaddicot 0.249 0.039 6.368 0.000 0.249 1.000semPaths(# Argumentos globales
ajuste_LTc, what="diagram", whatLabels="std",layout="tree3", rotation = 2, width=50, height=35,exoVar = F,
# Etiquetas
curvePivot=T, curvature = .5
)
Madre
Sexo
datos_mod <- semTools::indProd(datos, var1 = 40:55, var2 = c(2,4), match = FALSE,
meanC = TRUE, residualC = FALSE, doubleMC = TRUE)
modelo <- '
# Identidad social
it =~ siqs01 + siqs02 + siqs03
cc =~ siqs04 + siqs05 + siqs06
ia =~ siqs07 + siqs08 + siqs09
SIQS =~ it + cc + ia
# Liderazgo transformacional (coach)
LTm =~ tpqm02 + tpqm05 + tpqm12 + tpqm16 + tpqm01 + tpqm09 + tpqm11 + tpqm14 + tpqm03 + tpqm07 + tpqm10 + tpqm13 + tpqm04 + tpqm06 + tpqm08 + tpqm15
# MODERADORES
int.sex_LTm =~ tpqm01.sexo + tpqm02.sexo + tpqm03.sexo + tpqm04.sexo + tpqm05.sexo + tpqm06.sexo + tpqm07.sexo + tpqm08.sexo + tpqm09.sexo + tpqm10.sexo + tpqm11.sexo + tpqm12.sexo + tpqm13.sexo + tpqm14.sexo + tpqm15.sexo + tpqm16.sexo
# REGRESIONES
SIQS ~ LTm + sexo + edaddicot + int.sex_LTm
'
ajuste_LTm <- datos_mod %>%
sem(modelo, ., estimator="ULS")
# ajuste_LTc %>%
# summary(fit.measures = TRUE, standardized = TRUE)
semPaths(# Argumentos globales
ajuste_LTm, what="diagram", whatLabels="std",layout="tree3", rotation = 2, width=50, height=35,exoVar = F,
# Etiquetas
curvePivot=T, curvature = .5
)
Edad
modelo <- '
# Identidad social
it =~ siqs01 + siqs02 + siqs03
cc =~ siqs04 + siqs05 + siqs06
ia =~ siqs07 + siqs08 + siqs09
SIQS =~ it + cc + ia
# Liderazgo transformacional (coach)
LTm =~ tpqm02 + tpqm05 + tpqm12 + tpqm16 + tpqm01 + tpqm09 + tpqm11 + tpqm14 + tpqm03 + tpqm07 + tpqm10 + tpqm13 + tpqm04 + tpqm06 + tpqm08 + tpqm15
# MODERADORES
int.age_LTm =~ tpqm01.edaddicot + tpqm02.edaddicot + tpqm03.edaddicot + tpqm04.edaddicot + tpqm05.edaddicot + tpqm06.edaddicot + tpqm07.edaddicot + tpqm08.edaddicot + tpqm09.edaddicot + tpqm10.edaddicot + tpqm11.edaddicot + tpqm12.edaddicot + tpqm13.edaddicot + tpqm14.edaddicot + tpqm15.edaddicot + tpqm16.edaddicot
# REGRESIONES
SIQS ~ LTm + sexo + edaddicot + int.age_LTm
'
ajuste_LTm <- datos_mod %>%
sem(modelo, ., estimator="ULS")
# ajuste_LTc %>%
# summary(fit.measures = TRUE, standardized = TRUE)
semPaths(# Argumentos globales
ajuste_LTm, what="diagram", whatLabels="std",layout="tree3", rotation = 2, width=50, height=35,exoVar = F,
# Etiquetas
curvePivot=T, curvature = .5
)
Sexo y edad
modelo <- '
# Identidad social
it =~ siqs01 + siqs02 + siqs03
cc =~ siqs04 + siqs05 + siqs06
ia =~ siqs07 + siqs08 + siqs09
SIQS =~ it + cc + ia
# Liderazgo transformacional (coach)
LTm =~ tpqm02 + tpqm05 + tpqm12 + tpqm16 + tpqm01 + tpqm09 + tpqm11 + tpqm14 + tpqm03 + tpqm07 + tpqm10 + tpqm13 + tpqm04 + tpqm06 + tpqm08 + tpqm15
# MODERADORES
int.sex_LTm =~ tpqm01.sexo + tpqm02.sexo + tpqm03.sexo + tpqm04.sexo + tpqm05.sexo + tpqm06.sexo + tpqm07.sexo + tpqm08.sexo + tpqm09.sexo + tpqm10.sexo + tpqm11.sexo + tpqm12.sexo + tpqm13.sexo + tpqm14.sexo + tpqm15.sexo + tpqm16.sexo
int.age_LTm =~ tpqm01.edaddicot + tpqm02.edaddicot + tpqm03.edaddicot + tpqm04.edaddicot + tpqm05.edaddicot + tpqm06.edaddicot + tpqm07.edaddicot + tpqm08.edaddicot + tpqm09.edaddicot + tpqm10.edaddicot + tpqm11.edaddicot + tpqm12.edaddicot + tpqm13.edaddicot + tpqm14.edaddicot + tpqm15.edaddicot + tpqm16.edaddicot
# REGRESIONES
SIQS ~ LTm + sexo + edaddicot + int.sex_LTm + int.age_LTm
'
ajuste_LTm <- datos_mod %>%
sem(modelo, ., estimator="ULS")
ajuste_LTm %>%
summary(fit.measures = TRUE, standardized = TRUE)## lavaan 0.6.16 ended normally after 98 iterations
##
## Estimator ULS
## Optimization method NLMINB
## Number of model parameters 128
##
## Number of observations 656
##
## Model Test User Model:
##
## Test statistic 778.784
## Degrees of freedom 1642
## P-value (Unknown) NA
##
## Model Test Baseline Model:
##
## Test statistic 36954.497
## Degrees of freedom 1710
## P-value NA
##
## User Model versus Baseline Model:
##
## Comparative Fit Index (CFI) 1.000
## Tucker-Lewis Index (TLI) 1.026
##
## Root Mean Square Error of Approximation:
##
## RMSEA 0.000
## 90 Percent confidence interval - lower 0.000
## 90 Percent confidence interval - upper 0.000
## P-value H_0: RMSEA <= 0.050 1.000
## P-value H_0: RMSEA >= 0.080 0.000
##
## Standardized Root Mean Square Residual:
##
## SRMR 0.054
##
## Parameter Estimates:
##
## Standard errors Standard
## Information Expected
## Information saturated (h1) model Unstructured
##
## Latent Variables:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## it =~
## siqs01 1.000 0.947 0.860
## siqs02 0.897 0.027 33.194 0.000 0.849 0.678
## siqs03 1.163 0.033 35.465 0.000 1.101 0.909
## cc =~
## siqs04 1.000 0.950 0.728
## siqs05 1.057 0.037 28.816 0.000 1.004 0.739
## siqs06 0.912 0.033 28.014 0.000 0.866 0.731
## ia =~
## siqs07 1.000 0.834 0.866
## siqs08 1.040 0.033 31.622 0.000 0.867 0.890
## siqs09 0.939 0.031 30.523 0.000 0.783 0.712
## SIQS =~
## it 1.000 0.868 0.868
## cc 0.708 0.027 26.480 0.000 0.613 0.613
## ia 0.919 0.034 27.241 0.000 0.906 0.906
## LTm =~
## tpqm02 1.000 0.651 0.654
## tpqm05 1.049 0.031 34.333 0.000 0.683 0.770
## tpqm12 1.298 0.035 37.146 0.000 0.845 0.758
## tpqm16 0.765 0.026 29.277 0.000 0.498 0.726
## tpqm01 0.913 0.028 32.207 0.000 0.594 0.657
## tpqm09 1.209 0.033 36.275 0.000 0.787 0.809
## tpqm11 1.219 0.034 36.375 0.000 0.793 0.763
## tpqm14 1.091 0.031 34.900 0.000 0.710 0.839
## tpqm03 0.778 0.026 29.555 0.000 0.506 0.504
## tpqm07 0.988 0.030 33.434 0.000 0.643 0.612
## tpqm10 1.059 0.031 34.466 0.000 0.689 0.832
## tpqm13 1.376 0.036 37.807 0.000 0.896 0.776
## tpqm04 0.918 0.028 32.305 0.000 0.598 0.724
## tpqm06 1.163 0.033 35.771 0.000 0.757 0.820
## tpqm08 1.046 0.031 34.295 0.000 0.681 0.748
## tpqm15 1.274 0.035 36.920 0.000 0.829 0.758
## int.sex_LTm =~
## tpqm01.sexo 1.000 0.270 0.613
## tpqm02.sexo 1.114 0.124 8.995 0.000 0.301 0.624
## tpqm03.sexo 0.711 0.100 7.079 0.000 0.192 0.385
## tpqm04.sexo 1.096 0.123 8.934 0.000 0.296 0.745
## tpqm05.sexo 1.380 0.142 9.739 0.000 0.373 0.885
## tpqm06.sexo 1.296 0.136 9.537 0.000 0.350 0.783
## tpqm07.sexo 0.956 0.114 8.382 0.000 0.258 0.492
## tpqm08.sexo 1.091 0.122 8.915 0.000 0.295 0.655
## tpqm09.sexo 1.536 0.153 10.054 0.000 0.415 0.904
## tpqm10.sexo 1.239 0.132 9.382 0.000 0.335 0.834
## tpqm11.sexo 1.447 0.146 9.884 0.000 0.391 0.787
## tpqm12.sexo 1.452 0.147 9.894 0.000 0.392 0.730
## tpqm13.sexo 1.636 0.160 10.222 0.000 0.442 0.808
## tpqm14.sexo 1.196 0.129 9.257 0.000 0.323 0.784
## tpqm15.sexo 1.412 0.144 9.809 0.000 0.381 0.722
## tpqm16.sexo 0.909 0.111 8.170 0.000 0.246 0.739
## int.age_LTm =~
## tpqm01.edaddct 1.000 0.296 0.663
## tpqm02.edaddct 1.104 0.147 7.526 0.000 0.326 0.654
## tpqm03.edaddct 0.856 0.128 6.684 0.000 0.253 0.505
## tpqm04.edaddct 1.023 0.140 7.287 0.000 0.302 0.730
## tpqm05.edaddct 1.108 0.147 7.536 0.000 0.328 0.748
## tpqm06.edaddct 1.296 0.162 7.974 0.000 0.383 0.828
## tpqm07.edaddct 1.124 0.148 7.580 0.000 0.332 0.634
## tpqm08.edaddct 1.140 0.150 7.621 0.000 0.337 0.750
## tpqm09.edaddct 1.291 0.162 7.964 0.000 0.382 0.793
## tpqm10.edaddct 1.166 0.152 7.688 0.000 0.345 0.830
## tpqm11.edaddct 1.318 0.164 8.018 0.000 0.390 0.751
## tpqm12.edaddct 1.442 0.175 8.231 0.000 0.426 0.767
## tpqm13.edaddct 1.515 0.182 8.338 0.000 0.448 0.779
## tpqm14.edaddct 1.220 0.156 7.814 0.000 0.361 0.848
## tpqm15.edaddct 1.376 0.169 8.124 0.000 0.407 0.747
## tpqm16.edaddct 0.851 0.128 6.662 0.000 0.252 0.724
##
## Regressions:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## SIQS ~
## LTm 0.476 0.019 24.609 0.000 0.377 0.377
## sexo 0.343 0.088 3.893 0.000 0.417 0.208
## edaddicot -0.184 0.088 -2.095 0.036 -0.224 -0.112
## int.sex_LTm -0.083 0.052 -1.607 0.108 -0.027 -0.027
## int.age_LTm -0.115 0.038 -2.980 0.003 -0.041 -0.041
##
## Covariances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## LTm ~~
## int.sex_LTm -0.070 0.006 -11.447 0.000 -0.398 -0.398
## int.age_LTm 0.001 0.002 0.383 0.702 0.004 0.004
## int.sex_LTm ~~
## int.age_LTm 0.000 0.002 0.221 0.825 0.005 0.005
## sexo ~~
## edaddicot 0.023 0.039 0.600 0.549 0.023 0.094
##
## Variances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .siqs01 0.316 0.054 5.872 0.000 0.316 0.261
## .siqs02 0.847 0.050 16.797 0.000 0.847 0.540
## .siqs03 0.256 0.061 4.177 0.000 0.256 0.174
## .siqs04 0.802 0.058 13.809 0.000 0.802 0.471
## .siqs05 0.838 0.061 13.739 0.000 0.838 0.454
## .siqs06 0.653 0.054 12.059 0.000 0.653 0.466
## .siqs07 0.232 0.052 4.452 0.000 0.232 0.250
## .siqs08 0.198 0.054 3.697 0.000 0.198 0.208
## .siqs09 0.596 0.050 11.910 0.000 0.596 0.493
## .tpqm02 0.567 0.043 13.222 0.000 0.567 0.572
## .tpqm05 0.321 0.043 7.426 0.000 0.321 0.408
## .tpqm12 0.528 0.046 11.588 0.000 0.528 0.425
## .tpqm16 0.222 0.041 5.382 0.000 0.222 0.472
## .tpqm01 0.465 0.042 11.024 0.000 0.465 0.569
## .tpqm09 0.326 0.045 7.296 0.000 0.326 0.345
## .tpqm11 0.450 0.045 10.061 0.000 0.450 0.417
## .tpqm14 0.212 0.044 4.870 0.000 0.212 0.296
## .tpqm03 0.754 0.041 18.234 0.000 0.754 0.746
## .tpqm07 0.688 0.043 16.097 0.000 0.688 0.625
## .tpqm10 0.212 0.043 4.892 0.000 0.212 0.309
## .tpqm13 0.531 0.046 11.436 0.000 0.531 0.398
## .tpqm04 0.324 0.042 7.657 0.000 0.324 0.475
## .tpqm06 0.280 0.044 6.320 0.000 0.280 0.328
## .tpqm08 0.366 0.043 8.457 0.000 0.366 0.441
## .tpqm15 0.509 0.045 11.236 0.000 0.509 0.426
## .tpqm01.sexo 0.121 0.041 2.967 0.003 0.121 0.624
## .tpqm02.sexo 0.142 0.041 3.445 0.001 0.142 0.611
## .tpqm03.sexo 0.212 0.040 5.292 0.000 0.212 0.851
## .tpqm04.sexo 0.070 0.041 1.701 0.089 0.070 0.444
## .tpqm05.sexo 0.039 0.043 0.907 0.364 0.039 0.218
## .tpqm06.sexo 0.078 0.042 1.841 0.066 0.078 0.388
## .tpqm07.sexo 0.209 0.041 5.128 0.000 0.209 0.758
## .tpqm08.sexo 0.116 0.041 2.803 0.005 0.116 0.571
## .tpqm09.sexo 0.039 0.043 0.889 0.374 0.039 0.183
## .tpqm10.sexo 0.049 0.042 1.166 0.243 0.049 0.304
## .tpqm11.sexo 0.094 0.043 2.183 0.029 0.094 0.380
## .tpqm12.sexo 0.135 0.043 3.139 0.002 0.135 0.467
## .tpqm13.sexo 0.104 0.044 2.361 0.018 0.104 0.347
## .tpqm14.sexo 0.065 0.042 1.569 0.117 0.065 0.385
## .tpqm15.sexo 0.133 0.043 3.118 0.002 0.133 0.478
## .tpqm16.sexo 0.050 0.041 1.233 0.218 0.050 0.453
## .tpqm01.edaddct 0.111 0.043 2.612 0.009 0.111 0.560
## .tpqm02.edaddct 0.143 0.043 3.292 0.001 0.143 0.573
## .tpqm03.edaddct 0.187 0.042 4.503 0.000 0.187 0.745
## .tpqm04.edaddct 0.080 0.043 1.874 0.061 0.080 0.467
## .tpqm05.edaddct 0.085 0.043 1.951 0.051 0.085 0.441
## .tpqm06.edaddct 0.067 0.045 1.490 0.136 0.067 0.314
## .tpqm07.edaddct 0.164 0.044 3.773 0.000 0.164 0.598
## .tpqm08.edaddct 0.088 0.044 2.024 0.043 0.088 0.438
## .tpqm09.edaddct 0.086 0.045 1.910 0.056 0.086 0.371
## .tpqm10.edaddct 0.054 0.044 1.219 0.223 0.054 0.310
## .tpqm11.edaddct 0.117 0.045 2.592 0.010 0.117 0.436
## .tpqm12.edaddct 0.127 0.047 2.725 0.006 0.127 0.411
## .tpqm13.edaddct 0.130 0.047 2.738 0.006 0.130 0.393
## .tpqm14.edaddct 0.051 0.044 1.142 0.254 0.051 0.280
## .tpqm15.edaddct 0.131 0.046 2.860 0.004 0.131 0.442
## .tpqm16.edaddct 0.057 0.042 1.378 0.168 0.057 0.475
## .it 0.221 0.031 7.016 0.000 0.246 0.246
## .cc 0.563 0.035 16.236 0.000 0.625 0.625
## .ia 0.125 0.030 4.160 0.000 0.179 0.179
## .SIQS 0.538 0.030 18.185 0.000 0.796 0.796
## LTm 0.424 0.018 24.068 0.000 1.000 1.000
## int.sex_LTm 0.073 0.012 6.073 0.000 1.000 1.000
## int.age_LTm 0.087 0.017 5.159 0.000 1.000 1.000
## sexo 0.248 0.039 6.358 0.000 0.248 1.000
## edaddicot 0.249 0.039 6.368 0.000 0.249 1.000semPaths(# Argumentos globales
ajuste_LTm, what="diagram", whatLabels="std",layout="tree3", rotation = 2, width=50, height=35,exoVar = F,
# Etiquetas
curvePivot=T, curvature = .5
)
Padre
Sexo
datos_mod <- semTools::indProd(datos, var1 = 56:71, var2 = c(2,4), match = FALSE,
meanC = TRUE, residualC = FALSE, doubleMC = TRUE)
modelo <- '
# Identidad social
it =~ siqs01 + siqs02 + siqs03
cc =~ siqs04 + siqs05 + siqs06
ia =~ siqs07 + siqs08 + siqs09
SIQS =~ it + cc + ia
# Liderazgo transformacional (coach)
LTp =~ tpqp02 + tpqp05 + tpqp12 + tpqp16 + tpqp01 + tpqp09 + tpqp11 + tpqp14 + tpqp03 + tpqp07 + tpqp10 + tpqp13 + tpqp04 + tpqp06 + tpqp08 + tpqp15
# MODERADORES
int.sex_LTp =~ tpqp01.sexo + tpqp02.sexo + tpqp03.sexo + tpqp04.sexo + tpqp05.sexo + tpqp06.sexo + tpqp07.sexo + tpqp08.sexo + tpqp09.sexo + tpqp10.sexo + tpqp11.sexo + tpqp12.sexo + tpqp13.sexo + tpqp14.sexo + tpqp15.sexo + tpqp16.sexo
# REGRESIONES
SIQS ~ LTp + sexo + edaddicot + int.sex_LTp
'
ajuste_LTp <- datos_mod %>%
sem(modelo, ., estimator="ULS")
# ajuste_LTc %>%
# summary(fit.measures = TRUE, standardized = TRUE)
semPaths(# Argumentos globales
ajuste_LTp, what="diagram", whatLabels="std",layout="tree3", rotation = 2, width=50, height=35,exoVar = F,
# Etiquetas
curvePivot=T, curvature = .5
)
Edad
modelo <- '
# Identidad social
it =~ siqs01 + siqs02 + siqs03
cc =~ siqs04 + siqs05 + siqs06
ia =~ siqs07 + siqs08 + siqs09
SIQS =~ it + cc + ia
# Liderazgo transformacional (coach)
LTp =~ tpqp02 + tpqp05 + tpqp12 + tpqp16 + tpqp01 + tpqp09 + tpqp11 + tpqp14 + tpqp03 + tpqp07 + tpqp10 + tpqp13 + tpqp04 + tpqp06 + tpqp08 + tpqp15
# MODERADORES
int.age_LTp =~ tpqp01.edaddicot + tpqp02.edaddicot + tpqp03.edaddicot + tpqp04.edaddicot + tpqp05.edaddicot + tpqp06.edaddicot + tpqp07.edaddicot + tpqp08.edaddicot + tpqp09.edaddicot + tpqp10.edaddicot + tpqp11.edaddicot + tpqp12.edaddicot + tpqp13.edaddicot + tpqp14.edaddicot + tpqp15.edaddicot + tpqp16.edaddicot
# REGRESIONES
SIQS ~ LTp + sexo + edaddicot + int.age_LTp
'
ajuste_LTp <- datos_mod %>%
sem(modelo, ., estimator="ULS")
# ajuste_LTc %>%
# summary(fit.measures = TRUE, standardized = TRUE)
semPaths(# Argumentos globales
ajuste_LTp, what="diagram", whatLabels="std",layout="tree3", rotation = 2, width=50, height=35,exoVar = F,
# Etiquetas
curvePivot=T, curvature = .5
)
Sexo y edad
modelo <- '
# Identidad social
it =~ siqs01 + siqs02 + siqs03
cc =~ siqs04 + siqs05 + siqs06
ia =~ siqs07 + siqs08 + siqs09
SIQS =~ it + cc + ia
# Liderazgo transformacional (coach)
LTp =~ tpqp02 + tpqp05 + tpqp12 + tpqp16 + tpqp01 + tpqp09 + tpqp11 + tpqp14 + tpqp03 + tpqp07 + tpqp10 + tpqp13 + tpqp04 + tpqp06 + tpqp08 + tpqp15
# MODERADORES
int.sex_LTp =~ tpqp01.sexo + tpqp02.sexo + tpqp03.sexo + tpqp04.sexo + tpqp05.sexo + tpqp06.sexo + tpqp07.sexo + tpqp08.sexo + tpqp09.sexo + tpqp10.sexo + tpqp11.sexo + tpqp12.sexo + tpqp13.sexo + tpqp14.sexo + tpqp15.sexo + tpqp16.sexo
int.age_LTp =~ tpqp01.edaddicot + tpqp02.edaddicot + tpqp03.edaddicot + tpqp04.edaddicot + tpqp05.edaddicot + tpqp06.edaddicot + tpqp07.edaddicot + tpqp08.edaddicot + tpqp09.edaddicot + tpqp10.edaddicot + tpqp11.edaddicot + tpqp12.edaddicot + tpqp13.edaddicot + tpqp14.edaddicot + tpqp15.edaddicot + tpqp16.edaddicot
# REGRESIONES
SIQS ~ LTp + sexo + edaddicot + int.sex_LTp + int.age_LTp
'
ajuste_LTp <- datos_mod %>%
sem(modelo, ., estimator="ULS")
ajuste_LTp %>%
summary(fit.measures = TRUE, standardized = TRUE)## lavaan 0.6.16 ended normally after 93 iterations
##
## Estimator ULS
## Optimization method NLMINB
## Number of model parameters 128
##
## Number of observations 656
##
## Model Test User Model:
##
## Test statistic 959.745
## Degrees of freedom 1642
## P-value (Unknown) NA
##
## Model Test Baseline Model:
##
## Test statistic 65269.344
## Degrees of freedom 1710
## P-value NA
##
## User Model versus Baseline Model:
##
## Comparative Fit Index (CFI) 1.000
## Tucker-Lewis Index (TLI) 1.011
##
## Root Mean Square Error of Approximation:
##
## RMSEA 0.000
## 90 Percent confidence interval - lower 0.000
## 90 Percent confidence interval - upper 0.000
## P-value H_0: RMSEA <= 0.050 1.000
## P-value H_0: RMSEA >= 0.080 0.000
##
## Standardized Root Mean Square Residual:
##
## SRMR 0.043
##
## Parameter Estimates:
##
## Standard errors Standard
## Information Expected
## Information saturated (h1) model Unstructured
##
## Latent Variables:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## it =~
## siqs01 1.000 0.955 0.867
## siqs02 0.903 0.028 32.701 0.000 0.863 0.689
## siqs03 1.131 0.033 34.573 0.000 1.080 0.891
## cc =~
## siqs04 1.000 0.928 0.711
## siqs05 1.074 0.036 29.733 0.000 0.997 0.734
## siqs06 0.960 0.033 29.067 0.000 0.891 0.752
## ia =~
## siqs07 1.000 0.836 0.869
## siqs08 1.041 0.033 31.129 0.000 0.871 0.894
## siqs09 0.929 0.031 29.935 0.000 0.777 0.706
## SIQS =~
## it 1.000 0.835 0.835
## cc 0.772 0.029 26.604 0.000 0.664 0.664
## ia 0.935 0.035 27.051 0.000 0.891 0.891
## LTp =~
## tpqp02 1.000 0.664 0.654
## tpqp05 1.336 0.029 46.334 0.000 0.887 0.807
## tpqp12 1.627 0.033 48.929 0.000 1.080 0.794
## tpqp16 1.204 0.027 44.698 0.000 0.799 0.775
## tpqp01 1.173 0.027 44.268 0.000 0.779 0.715
## tpqp09 1.431 0.030 47.312 0.000 0.950 0.803
## tpqp11 1.408 0.030 47.089 0.000 0.935 0.765
## tpqp14 1.230 0.027 45.051 0.000 0.817 0.841
## tpqp03 0.877 0.023 38.769 0.000 0.582 0.533
## tpqp07 1.263 0.028 45.470 0.000 0.839 0.701
## tpqp10 1.274 0.028 45.608 0.000 0.846 0.852
## tpqp13 1.564 0.032 48.463 0.000 1.039 0.816
## tpqp04 1.313 0.028 46.071 0.000 0.872 0.787
## tpqp06 1.271 0.028 45.571 0.000 0.844 0.822
## tpqp08 1.352 0.029 46.513 0.000 0.898 0.810
## tpqp15 1.443 0.030 47.430 0.000 0.958 0.783
## int.sex_LTp =~
## tpqp01.sexo 1.000 0.396 0.761
## tpqp02.sexo 0.825 0.069 11.896 0.000 0.327 0.658
## tpqp03.sexo 0.632 0.063 10.097 0.000 0.250 0.460
## tpqp04.sexo 1.081 0.080 13.548 0.000 0.428 0.803
## tpqp05.sexo 1.132 0.082 13.804 0.000 0.448 0.856
## tpqp06.sexo 1.061 0.079 13.444 0.000 0.420 0.840
## tpqp07.sexo 1.019 0.077 13.212 0.000 0.403 0.685
## tpqp08.sexo 1.079 0.080 13.540 0.000 0.427 0.786
## tpqp09.sexo 1.183 0.084 14.035 0.000 0.468 0.817
## tpqp10.sexo 1.037 0.078 13.312 0.000 0.410 0.851
## tpqp11.sexo 1.130 0.082 13.795 0.000 0.447 0.747
## tpqp12.sexo 1.259 0.088 14.347 0.000 0.498 0.751
## tpqp13.sexo 1.258 0.088 14.344 0.000 0.498 0.817
## tpqp14.sexo 0.978 0.075 12.967 0.000 0.387 0.815
## tpqp15.sexo 1.138 0.082 13.832 0.000 0.450 0.758
## tpqp16.sexo 0.941 0.074 12.730 0.000 0.372 0.738
## int.age_LTp =~
## tpqp01.edaddct 1.000 0.395 0.731
## tpqp02.edaddct 0.839 0.077 10.900 0.000 0.332 0.653
## tpqp03.edaddct 0.741 0.073 10.161 0.000 0.293 0.540
## tpqp04.edaddct 1.104 0.089 12.377 0.000 0.437 0.788
## tpqp05.edaddct 1.113 0.090 12.418 0.000 0.440 0.809
## tpqp06.edaddct 1.089 0.088 12.311 0.000 0.431 0.840
## tpqp07.edaddct 1.084 0.088 12.288 0.000 0.429 0.713
## tpqp08.edaddct 1.130 0.090 12.490 0.000 0.447 0.814
## tpqp09.edaddct 1.180 0.093 12.689 0.000 0.466 0.796
## tpqp10.edaddct 1.068 0.087 12.213 0.000 0.422 0.856
## tpqp11.edaddct 1.132 0.091 12.497 0.000 0.448 0.739
## tpqp12.edaddct 1.365 0.103 13.300 0.000 0.540 0.791
## tpqp13.edaddct 1.310 0.100 13.139 0.000 0.518 0.816
## tpqp14.edaddct 1.038 0.086 12.071 0.000 0.411 0.847
## tpqp15.edaddct 1.184 0.093 12.706 0.000 0.468 0.768
## tpqp16.edaddct 1.038 0.086 12.072 0.000 0.411 0.793
##
## Regressions:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## SIQS ~
## LTp 0.345 0.012 28.085 0.000 0.287 0.287
## sexo 0.331 0.086 3.849 0.000 0.414 0.207
## edaddicot -0.185 0.086 -2.164 0.030 -0.232 -0.116
## int.sex_LTp -0.151 0.025 -5.943 0.000 -0.075 -0.075
## int.age_LTp -0.202 0.025 -8.082 0.000 -0.100 -0.100
##
## Covariances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## LTp ~~
## int.sex_LTp -0.066 0.004 -16.315 0.000 -0.250 -0.250
## int.age_LTp 0.004 0.002 2.484 0.013 0.016 0.016
## int.sex_LTp ~~
## int.age_LTp -0.007 0.002 -3.099 0.002 -0.043 -0.043
## sexo ~~
## edaddicot 0.023 0.039 0.600 0.549 0.023 0.094
##
## Variances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .siqs01 0.301 0.055 5.500 0.000 0.301 0.248
## .siqs02 0.824 0.051 16.100 0.000 0.824 0.526
## .siqs03 0.302 0.061 4.971 0.000 0.302 0.205
## .siqs04 0.843 0.056 14.972 0.000 0.843 0.495
## .siqs05 0.852 0.060 14.255 0.000 0.852 0.462
## .siqs06 0.609 0.055 11.157 0.000 0.609 0.434
## .siqs07 0.227 0.052 4.336 0.000 0.227 0.245
## .siqs08 0.192 0.054 3.552 0.000 0.192 0.202
## .siqs09 0.606 0.050 12.109 0.000 0.606 0.501
## .tpqp02 0.591 0.042 14.119 0.000 0.591 0.573
## .tpqp05 0.421 0.044 9.545 0.000 0.421 0.348
## .tpqp12 0.685 0.047 14.688 0.000 0.685 0.370
## .tpqp16 0.426 0.043 9.880 0.000 0.426 0.400
## .tpqp01 0.581 0.043 13.552 0.000 0.581 0.489
## .tpqp09 0.499 0.045 11.120 0.000 0.499 0.356
## .tpqp11 0.620 0.045 13.885 0.000 0.620 0.415
## .tpqp14 0.275 0.043 6.360 0.000 0.275 0.292
## .tpqp03 0.854 0.041 20.734 0.000 0.854 0.716
## .tpqp07 0.727 0.044 16.698 0.000 0.727 0.508
## .tpqp10 0.271 0.044 6.206 0.000 0.271 0.274
## .tpqp13 0.540 0.046 11.727 0.000 0.540 0.334
## .tpqp04 0.467 0.044 10.629 0.000 0.467 0.380
## .tpqp06 0.342 0.044 7.836 0.000 0.342 0.324
## .tpqp08 0.423 0.044 9.577 0.000 0.423 0.344
## .tpqp15 0.578 0.045 12.858 0.000 0.578 0.386
## .tpqp01.sexo 0.113 0.043 2.670 0.008 0.113 0.420
## .tpqp02.sexo 0.139 0.041 3.366 0.001 0.139 0.566
## .tpqp03.sexo 0.233 0.040 5.771 0.000 0.233 0.788
## .tpqp04.sexo 0.100 0.043 2.330 0.020 0.100 0.355
## .tpqp05.sexo 0.073 0.044 1.679 0.093 0.073 0.267
## .tpqp06.sexo 0.073 0.043 1.705 0.088 0.073 0.294
## .tpqp07.sexo 0.184 0.043 4.309 0.000 0.184 0.531
## .tpqp08.sexo 0.113 0.043 2.622 0.009 0.113 0.383
## .tpqp09.sexo 0.109 0.044 2.485 0.013 0.109 0.333
## .tpqp10.sexo 0.064 0.043 1.502 0.133 0.064 0.276
## .tpqp11.sexo 0.158 0.043 3.639 0.000 0.158 0.442
## .tpqp12.sexo 0.192 0.045 4.308 0.000 0.192 0.436
## .tpqp13.sexo 0.124 0.045 2.777 0.005 0.124 0.333
## .tpqp14.sexo 0.076 0.042 1.793 0.073 0.076 0.337
## .tpqp15.sexo 0.150 0.044 3.453 0.001 0.150 0.426
## .tpqp16.sexo 0.116 0.042 2.746 0.006 0.116 0.455
## .tpqp01.edaddct 0.137 0.043 3.163 0.002 0.137 0.466
## .tpqp02.edaddct 0.148 0.042 3.533 0.000 0.148 0.574
## .tpqp03.edaddct 0.209 0.041 5.062 0.000 0.209 0.709
## .tpqp04.edaddct 0.116 0.044 2.630 0.009 0.116 0.378
## .tpqp05.edaddct 0.102 0.044 2.307 0.021 0.102 0.345
## .tpqp06.edaddct 0.078 0.044 1.767 0.077 0.078 0.295
## .tpqp07.edaddct 0.177 0.044 4.039 0.000 0.177 0.491
## .tpqp08.edaddct 0.102 0.044 2.296 0.022 0.102 0.338
## .tpqp09.edaddct 0.126 0.045 2.806 0.005 0.126 0.366
## .tpqp10.edaddct 0.065 0.044 1.485 0.137 0.065 0.267
## .tpqp11.edaddct 0.167 0.044 3.758 0.000 0.167 0.454
## .tpqp12.edaddct 0.174 0.047 3.708 0.000 0.174 0.374
## .tpqp13.edaddct 0.135 0.046 2.909 0.004 0.135 0.334
## .tpqp14.edaddct 0.066 0.044 1.527 0.127 0.066 0.283
## .tpqp15.edaddct 0.152 0.045 3.388 0.001 0.152 0.410
## .tpqp16.edaddct 0.100 0.044 2.288 0.022 0.100 0.371
## .it 0.276 0.032 8.721 0.000 0.302 0.302
## .cc 0.481 0.032 15.012 0.000 0.559 0.559
## .ia 0.144 0.030 4.774 0.000 0.205 0.205
## .SIQS 0.535 0.029 18.378 0.000 0.841 0.841
## LTp 0.441 0.015 29.514 0.000 1.000 1.000
## int.sex_LTp 0.157 0.017 9.359 0.000 1.000 1.000
## int.age_LTp 0.156 0.018 8.511 0.000 1.000 1.000
## sexo 0.248 0.039 6.358 0.000 0.248 1.000
## edaddicot 0.249 0.039 6.368 0.000 0.249 1.000semPaths(# Argumentos globales
ajuste_LTp, what="diagram", whatLabels="std",layout="tree3", rotation = 2, width=50, height=35,exoVar = F,
# Etiquetas
curvePivot=T, curvature = .5
)
Madre y Padre
Sexo
datos_mod <- semTools::indProd(datos, var1 = 40:71, var2 = c(2,4), match = FALSE,
meanC = TRUE, residualC = FALSE, doubleMC = TRUE)
modelo <- '
# Identidad social
it =~ siqs01 + siqs02 + siqs03
cc =~ siqs04 + siqs05 + siqs06
ia =~ siqs07 + siqs08 + siqs09
SIQS =~ it + cc + ia
# Liderazgo transformacional (coach)
LTp =~ tpqp02 + tpqp05 + tpqp12 + tpqp16 + tpqp01 + tpqp09 + tpqp11 + tpqp14 + tpqp03 + tpqp07 + tpqp10 + tpqp13 + tpqp04 + tpqp06 + tpqp08 + tpqp15
LTm =~ tpqm02 + tpqm05 + tpqm12 + tpqm16 + tpqm01 + tpqm09 + tpqm11 + tpqm14 + tpqm03 + tpqm07 + tpqm10 + tpqm13 + tpqm04 + tpqm06 + tpqm08 + tpqm15
# MODERADORES
int.sex_LTp =~ tpqp01.sexo + tpqp02.sexo + tpqp03.sexo + tpqp04.sexo + tpqp05.sexo + tpqp06.sexo + tpqp07.sexo + tpqp08.sexo + tpqp09.sexo + tpqp10.sexo + tpqp11.sexo + tpqp12.sexo + tpqp13.sexo + tpqp14.sexo + tpqp15.sexo + tpqp16.sexo
int.sex_LTm =~ tpqm01.sexo + tpqm02.sexo + tpqm03.sexo + tpqm04.sexo + tpqm05.sexo + tpqm06.sexo + tpqm07.sexo + tpqm08.sexo + tpqm09.sexo + tpqm10.sexo + tpqm11.sexo + tpqm12.sexo + tpqm13.sexo + tpqm14.sexo + tpqm15.sexo + tpqm16.sexo
# REGRESIONES
SIQS ~ LTp + LTm + sexo + edaddicot + int.sex_LTp + int.sex_LTm
'
ajuste_LT <- datos_mod %>%
sem(modelo, ., estimator="ULS")
# ajuste_LTc %>%
# summary(fit.measures = TRUE, standardized = TRUE)
semPaths(# Argumentos globales
ajuste_LT, what="diagram", whatLabels="std",layout="tree3", rotation = 2, width=50, height=35,exoVar = F,
# Etiquetas
curvePivot=T, curvature = .5
)
Edad
modelo <- '
# Identidad social
it =~ siqs01 + siqs02 + siqs03
cc =~ siqs04 + siqs05 + siqs06
ia =~ siqs07 + siqs08 + siqs09
SIQS =~ it + cc + ia
# Liderazgo transformacional (coach)
LTp =~ tpqp02 + tpqp05 + tpqp12 + tpqp16 + tpqp01 + tpqp09 + tpqp11 + tpqp14 + tpqp03 + tpqp07 + tpqp10 + tpqp13 + tpqp04 + tpqp06 + tpqp08 + tpqp15
LTm =~ tpqm02 + tpqm05 + tpqm12 + tpqm16 + tpqm01 + tpqm09 + tpqm11 + tpqm14 + tpqm03 + tpqm07 + tpqm10 + tpqm13 + tpqm04 + tpqm06 + tpqm08 + tpqm15
# MODERADORES
int.age_LTp =~ tpqp01.edaddicot + tpqp02.edaddicot + tpqp03.edaddicot + tpqp04.edaddicot + tpqp05.edaddicot + tpqp06.edaddicot + tpqp07.edaddicot + tpqp08.edaddicot + tpqp09.edaddicot + tpqp10.edaddicot + tpqp11.edaddicot + tpqp12.edaddicot + tpqp13.edaddicot + tpqp14.edaddicot + tpqp15.edaddicot + tpqp16.edaddicot
int.age_LTm =~ tpqm01.edaddicot + tpqm02.edaddicot + tpqm03.edaddicot + tpqm04.edaddicot + tpqm05.edaddicot + tpqm06.edaddicot + tpqm07.edaddicot + tpqm08.edaddicot + tpqm09.edaddicot + tpqm10.edaddicot + tpqm11.edaddicot + tpqm12.edaddicot + tpqm13.edaddicot + tpqm14.edaddicot + tpqm15.edaddicot + tpqm16.edaddicot
# REGRESIONES
SIQS ~ LTp + LTm + sexo + edaddicot + int.age_LTp + int.age_LTm
'
ajuste_LT <- datos_mod %>%
sem(modelo, ., estimator="ULS")
# ajuste_LTc %>%
# summary(fit.measures = TRUE, standardized = TRUE)
semPaths(# Argumentos globales
ajuste_LT, what="diagram", whatLabels="std",layout="tree3", rotation = 2, width=50, height=35,exoVar = F,
# Etiquetas
curvePivot=T, curvature = .5
)
Sexo y edad
modelo <- '
# Identidad social
it =~ siqs01 + siqs02 + siqs03
cc =~ siqs04 + siqs05 + siqs06
ia =~ siqs07 + siqs08 + siqs09
SIQS =~ it + cc + ia
# Liderazgo transformacional (coach)
LTp =~ tpqp02 + tpqp05 + tpqp12 + tpqp16 + tpqp01 + tpqp09 + tpqp11 + tpqp14 + tpqp03 + tpqp07 + tpqp10 + tpqp13 + tpqp04 + tpqp06 + tpqp08 + tpqp15
LTm =~ tpqm02 + tpqm05 + tpqm12 + tpqm16 + tpqm01 + tpqm09 + tpqm11 + tpqm14 + tpqm03 + tpqm07 + tpqm10 + tpqm13 + tpqm04 + tpqm06 + tpqm08 + tpqm15
# MODERADORES
int.sex_LTp =~ tpqp01.sexo + tpqp02.sexo + tpqp03.sexo + tpqp04.sexo + tpqp05.sexo + tpqp06.sexo + tpqp07.sexo + tpqp08.sexo + tpqp09.sexo + tpqp10.sexo + tpqp11.sexo + tpqp12.sexo + tpqp13.sexo + tpqp14.sexo + tpqp15.sexo + tpqp16.sexo
int.age_LTp =~ tpqp01.edaddicot + tpqp02.edaddicot + tpqp03.edaddicot + tpqp04.edaddicot + tpqp05.edaddicot + tpqp06.edaddicot + tpqp07.edaddicot + tpqp08.edaddicot + tpqp09.edaddicot + tpqp10.edaddicot + tpqp11.edaddicot + tpqp12.edaddicot + tpqp13.edaddicot + tpqp14.edaddicot + tpqp15.edaddicot + tpqp16.edaddicot
int.sex_LTm =~ tpqm01.sexo + tpqm02.sexo + tpqm03.sexo + tpqm04.sexo + tpqm05.sexo + tpqm06.sexo + tpqm07.sexo + tpqm08.sexo + tpqm09.sexo + tpqm10.sexo + tpqm11.sexo + tpqm12.sexo + tpqm13.sexo + tpqm14.sexo + tpqm15.sexo + tpqm16.sexo
int.age_LTm =~ tpqm01.edaddicot + tpqm02.edaddicot + tpqm03.edaddicot + tpqm04.edaddicot + tpqm05.edaddicot + tpqm06.edaddicot + tpqm07.edaddicot + tpqm08.edaddicot + tpqm09.edaddicot + tpqm10.edaddicot + tpqm11.edaddicot + tpqm12.edaddicot + tpqm13.edaddicot + tpqm14.edaddicot + tpqm15.edaddicot + tpqm16.edaddicot
# REGRESIONES
SIQS ~ LTp + LTm + sexo + edaddicot + int.sex_LTp + int.age_LTp + int.sex_LTm + int.age_LTm
'
ajuste_LT <- datos_mod %>%
sem(modelo, ., estimator="ULS")
ajuste_LT %>%
summary(fit.measures = TRUE, standardized = TRUE)## lavaan 0.6.16 ended normally after 204 iterations
##
## Estimator ULS
## Optimization method NLMINB
## Number of model parameters 239
##
## Number of observations 656
##
## Model Test User Model:
##
## Test statistic 3149.643
## Degrees of freedom 5539
## P-value (Unknown) NA
##
## Model Test Baseline Model:
##
## Test statistic 125127.995
## Degrees of freedom 5670
## P-value NA
##
## User Model versus Baseline Model:
##
## Comparative Fit Index (CFI) 1.000
## Tucker-Lewis Index (TLI) 1.020
##
## Root Mean Square Error of Approximation:
##
## RMSEA 0.000
## 90 Percent confidence interval - lower 0.000
## 90 Percent confidence interval - upper 0.000
## P-value H_0: RMSEA <= 0.050 1.000
## P-value H_0: RMSEA >= 0.080 0.000
##
## Standardized Root Mean Square Residual:
##
## SRMR 0.052
##
## Parameter Estimates:
##
## Standard errors Standard
## Information Expected
## Information saturated (h1) model Unstructured
##
## Latent Variables:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## it =~
## siqs01 1.000 0.945 0.859
## siqs02 0.929 0.025 36.905 0.000 0.878 0.701
## siqs03 1.136 0.029 39.164 0.000 1.074 0.887
## cc =~
## siqs04 1.000 0.933 0.714
## siqs05 1.060 0.034 31.383 0.000 0.989 0.728
## siqs06 0.958 0.031 30.650 0.000 0.894 0.754
## ia =~
## siqs07 1.000 0.840 0.873
## siqs08 1.025 0.029 34.884 0.000 0.862 0.884
## siqs09 0.930 0.028 33.620 0.000 0.782 0.711
## SIQS =~
## it 1.000 0.864 0.864
## cc 0.718 0.024 30.102 0.000 0.629 0.629
## ia 0.923 0.029 31.899 0.000 0.897 0.897
## LTp =~
## tpqp02 1.000 0.677 0.667
## tpqp05 1.295 0.024 53.053 0.000 0.877 0.798
## tpqp12 1.621 0.029 56.661 0.000 1.098 0.807
## tpqp16 1.132 0.022 50.437 0.000 0.766 0.743
## tpqp01 1.126 0.022 50.335 0.000 0.763 0.699
## tpqp09 1.427 0.026 54.727 0.000 0.966 0.816
## tpqp11 1.399 0.026 54.397 0.000 0.947 0.775
## tpqp14 1.215 0.023 51.849 0.000 0.823 0.847
## tpqp03 0.885 0.020 44.927 0.000 0.599 0.548
## tpqp07 1.247 0.024 52.346 0.000 0.844 0.706
## tpqp10 1.221 0.024 51.956 0.000 0.827 0.833
## tpqp13 1.529 0.027 55.812 0.000 1.035 0.814
## tpqp04 1.243 0.024 52.295 0.000 0.842 0.760
## tpqp06 1.229 0.024 52.074 0.000 0.832 0.811
## tpqp08 1.313 0.025 53.295 0.000 0.889 0.801
## tpqp15 1.469 0.027 55.193 0.000 0.995 0.813
## LTm =~
## tpqm02 1.000 0.633 0.636
## tpqm05 1.009 0.024 41.954 0.000 0.638 0.719
## tpqm12 1.398 0.029 47.564 0.000 0.884 0.793
## tpqm16 0.768 0.021 36.271 0.000 0.486 0.709
## tpqm01 0.903 0.023 39.709 0.000 0.571 0.631
## tpqm09 1.211 0.027 45.299 0.000 0.766 0.788
## tpqm11 1.290 0.028 46.332 0.000 0.816 0.785
## tpqm14 1.136 0.026 44.170 0.000 0.718 0.848
## tpqm03 0.822 0.022 37.741 0.000 0.520 0.518
## tpqm07 1.017 0.024 42.102 0.000 0.643 0.613
## tpqm10 1.087 0.025 43.378 0.000 0.688 0.830
## tpqm13 1.433 0.030 47.920 0.000 0.906 0.785
## tpqm04 0.905 0.023 39.752 0.000 0.572 0.694
## tpqm06 1.181 0.026 44.861 0.000 0.747 0.809
## tpqm08 1.069 0.025 43.069 0.000 0.677 0.743
## tpqm15 1.365 0.029 47.208 0.000 0.863 0.789
## int.sex_LTp =~
## tpqp01.sexo 1.000 0.380 0.730
## tpqp02.sexo 0.890 0.064 13.887 0.000 0.338 0.681
## tpqp03.sexo 0.654 0.057 11.533 0.000 0.248 0.456
## tpqp04.sexo 1.067 0.070 15.161 0.000 0.405 0.761
## tpqp05.sexo 1.150 0.074 15.638 0.000 0.436 0.834
## tpqp06.sexo 1.106 0.072 15.395 0.000 0.420 0.841
## tpqp07.sexo 1.032 0.069 14.934 0.000 0.392 0.666
## tpqp08.sexo 1.139 0.073 15.580 0.000 0.432 0.796
## tpqp09.sexo 1.250 0.077 16.141 0.000 0.474 0.828
## tpqp10.sexo 1.058 0.070 15.105 0.000 0.402 0.833
## tpqp11.sexo 1.225 0.076 16.022 0.000 0.465 0.777
## tpqp12.sexo 1.348 0.081 16.562 0.000 0.512 0.771
## tpqp13.sexo 1.293 0.079 16.336 0.000 0.491 0.805
## tpqp14.sexo 1.049 0.070 15.045 0.000 0.398 0.838
## tpqp15.sexo 1.241 0.077 16.102 0.000 0.471 0.793
## tpqp16.sexo 0.941 0.066 14.294 0.000 0.357 0.709
## int.age_LTp =~
## tpqp01.edaddct 1.000 0.385 0.712
## tpqp02.edaddct 0.878 0.071 12.371 0.000 0.338 0.666
## tpqp03.edaddct 0.787 0.068 11.647 0.000 0.303 0.559
## tpqp04.edaddct 1.093 0.080 13.712 0.000 0.421 0.761
## tpqp05.edaddct 1.128 0.081 13.886 0.000 0.435 0.799
## tpqp06.edaddct 1.093 0.080 13.709 0.000 0.421 0.821
## tpqp07.edaddct 1.121 0.081 13.852 0.000 0.432 0.719
## tpqp08.edaddct 1.146 0.082 13.975 0.000 0.442 0.804
## tpqp09.edaddct 1.237 0.086 14.376 0.000 0.477 0.813
## tpqp10.edaddct 1.073 0.079 13.605 0.000 0.414 0.838
## tpqp11.edaddct 1.185 0.084 14.153 0.000 0.457 0.754
## tpqp12.edaddct 1.425 0.095 15.038 0.000 0.549 0.805
## tpqp13.edaddct 1.339 0.091 14.760 0.000 0.516 0.813
## tpqp14.edaddct 1.071 0.079 13.594 0.000 0.413 0.851
## tpqp15.edaddct 1.269 0.088 14.505 0.000 0.489 0.803
## tpqp16.edaddct 1.015 0.076 13.280 0.000 0.391 0.756
## int.sex_LTm =~
## tpqm01.sexo 1.000 0.260 0.590
## tpqm02.sexo 1.186 0.105 11.328 0.000 0.309 0.639
## tpqm03.sexo 0.760 0.084 9.016 0.000 0.198 0.396
## tpqm04.sexo 1.077 0.099 10.872 0.000 0.280 0.705
## tpqm05.sexo 1.341 0.113 11.856 0.000 0.349 0.827
## tpqm06.sexo 1.352 0.114 11.891 0.000 0.352 0.786
## tpqm07.sexo 1.003 0.095 10.514 0.000 0.261 0.497
## tpqm08.sexo 1.140 0.102 11.142 0.000 0.296 0.659
## tpqm09.sexo 1.517 0.123 12.330 0.000 0.395 0.859
## tpqm10.sexo 1.307 0.111 11.752 0.000 0.340 0.848
## tpqm11.sexo 1.554 0.125 12.414 0.000 0.404 0.814
## tpqm12.sexo 1.623 0.129 12.561 0.000 0.422 0.785
## tpqm13.sexo 1.665 0.132 12.644 0.000 0.433 0.791
## tpqm14.sexo 1.261 0.109 11.597 0.000 0.328 0.796
## tpqm15.sexo 1.545 0.125 12.394 0.000 0.402 0.761
## tpqm16.sexo 0.901 0.091 9.952 0.000 0.234 0.705
## int.age_LTm =~
## tpqm01.edaddct 1.000 0.284 0.637
## tpqm02.edaddct 1.114 0.118 9.428 0.000 0.316 0.633
## tpqm03.edaddct 0.926 0.107 8.648 0.000 0.263 0.524
## tpqm04.edaddct 1.011 0.112 9.029 0.000 0.287 0.692
## tpqm05.edaddct 1.077 0.116 9.292 0.000 0.306 0.697
## tpqm06.edaddct 1.324 0.132 10.063 0.000 0.376 0.812
## tpqm07.edaddct 1.163 0.121 9.595 0.000 0.330 0.629
## tpqm08.edaddct 1.188 0.123 9.675 0.000 0.337 0.750
## tpqm09.edaddct 1.314 0.131 10.037 0.000 0.373 0.774
## tpqm10.edaddct 1.216 0.125 9.762 0.000 0.345 0.831
## tpqm11.edaddct 1.413 0.138 10.277 0.000 0.401 0.773
## tpqm12.edaddct 1.567 0.148 10.583 0.000 0.444 0.800
## tpqm13.edaddct 1.597 0.150 10.636 0.000 0.453 0.788
## tpqm14.edaddct 1.281 0.129 9.949 0.000 0.363 0.855
## tpqm15.edaddct 1.508 0.144 10.475 0.000 0.428 0.785
## tpqm16.edaddct 0.862 0.104 8.327 0.000 0.245 0.704
##
## Regressions:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## SIQS ~
## LTp 0.112 0.015 7.612 0.000 0.093 0.093
## LTm 0.412 0.023 18.092 0.000 0.319 0.319
## sexo 0.340 0.088 3.889 0.000 0.417 0.208
## edaddicot -0.184 0.087 -2.103 0.035 -0.225 -0.112
## int.sex_LTp -0.106 0.052 -2.028 0.043 -0.049 -0.049
## int.age_LTp -0.240 0.054 -4.475 0.000 -0.113 -0.113
## int.sex_LTm 0.036 0.090 0.398 0.691 0.011 0.011
## int.age_LTm 0.098 0.076 1.287 0.198 0.034 0.034
##
## Covariances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## LTp ~~
## LTm 0.273 0.006 43.751 0.000 0.638 0.638
## int.sex_LTp -0.064 0.004 -17.882 0.000 -0.249 -0.249
## int.age_LTp 0.004 0.002 2.469 0.014 0.016 0.016
## int.sex_LTm -0.046 0.003 -13.377 0.000 -0.263 -0.263
## int.age_LTm -0.001 0.002 -0.562 0.574 -0.004 -0.004
## LTm ~~
## int.sex_LTp -0.062 0.004 -16.947 0.000 -0.259 -0.259
## int.age_LTp -0.001 0.002 -0.604 0.546 -0.005 -0.005
## int.sex_LTm -0.065 0.005 -13.711 0.000 -0.395 -0.395
## int.age_LTm 0.001 0.002 0.392 0.695 0.004 0.004
## int.sex_LTp ~~
## int.age_LTp -0.006 0.002 -3.023 0.003 -0.042 -0.042
## int.sex_LTm 0.061 0.005 11.659 0.000 0.616 0.616
## int.age_LTm 0.005 0.002 2.819 0.005 0.048 0.048
## int.age_LTp ~~
## int.sex_LTm 0.005 0.002 2.909 0.004 0.050 0.050
## int.age_LTm 0.070 0.007 10.358 0.000 0.636 0.636
## int.sex_LTm ~~
## int.age_LTm 0.000 0.002 0.171 0.864 0.004 0.004
## sexo ~~
## edaddicot 0.023 0.039 0.600 0.549 0.023 0.094
##
## Variances:
## Estimate Std.Err z-value P(>|z|) Std.lv Std.all
## .siqs01 0.319 0.052 6.094 0.000 0.319 0.263
## .siqs02 0.797 0.050 15.908 0.000 0.797 0.508
## .siqs03 0.314 0.058 5.440 0.000 0.314 0.214
## .siqs04 0.834 0.055 15.038 0.000 0.834 0.490
## .siqs05 0.868 0.058 14.931 0.000 0.868 0.470
## .siqs06 0.605 0.054 11.238 0.000 0.605 0.431
## .siqs07 0.220 0.051 4.308 0.000 0.220 0.238
## .siqs08 0.208 0.052 3.987 0.000 0.208 0.219
## .siqs09 0.598 0.049 12.199 0.000 0.598 0.495
## .tpqp02 0.573 0.041 13.863 0.000 0.573 0.556
## .tpqp05 0.439 0.043 10.227 0.000 0.439 0.363
## .tpqp12 0.647 0.045 14.310 0.000 0.647 0.349
## .tpqp16 0.478 0.042 11.379 0.000 0.478 0.449
## .tpqp01 0.607 0.042 14.467 0.000 0.607 0.511
## .tpqp09 0.468 0.044 10.692 0.000 0.468 0.334
## .tpqp11 0.597 0.044 13.706 0.000 0.597 0.400
## .tpqp14 0.266 0.042 6.270 0.000 0.266 0.282
## .tpqp03 0.834 0.041 20.431 0.000 0.834 0.699
## .tpqp07 0.717 0.043 16.831 0.000 0.717 0.502
## .tpqp10 0.302 0.042 7.115 0.000 0.302 0.306
## .tpqp13 0.547 0.044 12.289 0.000 0.547 0.338
## .tpqp04 0.518 0.043 12.157 0.000 0.518 0.422
## .tpqp06 0.361 0.043 8.497 0.000 0.361 0.343
## .tpqp08 0.440 0.043 10.224 0.000 0.440 0.358
## .tpqp15 0.507 0.044 11.515 0.000 0.507 0.339
## .tpqm02 0.590 0.041 14.272 0.000 0.590 0.596
## .tpqm05 0.380 0.041 9.179 0.000 0.380 0.482
## .tpqm12 0.460 0.044 10.555 0.000 0.460 0.371
## .tpqm16 0.234 0.040 5.786 0.000 0.234 0.497
## .tpqm01 0.492 0.041 12.029 0.000 0.492 0.601
## .tpqm09 0.358 0.042 8.435 0.000 0.358 0.379
## .tpqm11 0.414 0.043 9.647 0.000 0.414 0.383
## .tpqm14 0.201 0.042 4.782 0.000 0.201 0.280
## .tpqm03 0.739 0.041 18.209 0.000 0.739 0.732
## .tpqm07 0.688 0.041 16.602 0.000 0.688 0.624
## .tpqm10 0.214 0.042 5.115 0.000 0.214 0.311
## .tpqm13 0.512 0.044 11.668 0.000 0.512 0.384
## .tpqm04 0.353 0.041 8.636 0.000 0.353 0.519
## .tpqm06 0.295 0.042 6.973 0.000 0.295 0.346
## .tpqm08 0.372 0.042 8.914 0.000 0.372 0.448
## .tpqm15 0.451 0.043 10.402 0.000 0.451 0.377
## .tpqp01.sexo 0.126 0.041 3.039 0.002 0.126 0.466
## .tpqp02.sexo 0.132 0.041 3.223 0.001 0.132 0.536
## .tpqp03.sexo 0.234 0.040 5.848 0.000 0.234 0.792
## .tpqp04.sexo 0.119 0.042 2.850 0.004 0.119 0.420
## .tpqp05.sexo 0.083 0.042 1.971 0.049 0.083 0.304
## .tpqp06.sexo 0.073 0.042 1.739 0.082 0.073 0.293
## .tpqp07.sexo 0.193 0.042 4.637 0.000 0.193 0.557
## .tpqp08.sexo 0.108 0.042 2.569 0.010 0.108 0.367
## .tpqp09.sexo 0.103 0.043 2.407 0.016 0.103 0.314
## .tpqp10.sexo 0.071 0.042 1.704 0.088 0.071 0.306
## .tpqp11.sexo 0.142 0.043 3.331 0.001 0.142 0.397
## .tpqp12.sexo 0.178 0.043 4.103 0.000 0.178 0.405
## .tpqp13.sexo 0.130 0.043 3.028 0.002 0.130 0.351
## .tpqp14.sexo 0.067 0.042 1.611 0.107 0.067 0.298
## .tpqp15.sexo 0.131 0.043 3.065 0.002 0.131 0.371
## .tpqp16.sexo 0.127 0.041 3.074 0.002 0.127 0.498
## .tpqp01.edaddct 0.144 0.042 3.427 0.001 0.144 0.493
## .tpqp02.edaddct 0.144 0.041 3.470 0.001 0.144 0.557
## .tpqp03.edaddct 0.203 0.041 4.954 0.000 0.203 0.688
## .tpqp04.edaddct 0.129 0.043 3.018 0.003 0.129 0.421
## .tpqp05.edaddct 0.107 0.043 2.486 0.013 0.107 0.361
## .tpqp06.edaddct 0.086 0.043 2.006 0.045 0.086 0.326
## .tpqp07.edaddct 0.175 0.043 4.065 0.000 0.175 0.483
## .tpqp08.edaddct 0.106 0.043 2.468 0.014 0.106 0.353
## .tpqp09.edaddct 0.116 0.044 2.651 0.008 0.116 0.338
## .tpqp10.edaddct 0.072 0.043 1.695 0.090 0.072 0.297
## .tpqp11.edaddct 0.159 0.043 3.651 0.000 0.159 0.432
## .tpqp12.edaddct 0.164 0.045 3.604 0.000 0.164 0.352
## .tpqp13.edaddct 0.137 0.045 3.061 0.002 0.137 0.339
## .tpqp14.edaddct 0.065 0.043 1.518 0.129 0.065 0.275
## .tpqp15.edaddct 0.132 0.044 2.993 0.003 0.132 0.355
## .tpqp16.edaddct 0.115 0.042 2.725 0.006 0.115 0.429
## .tpqm01.sexo 0.127 0.040 3.155 0.002 0.127 0.652
## .tpqm02.sexo 0.138 0.041 3.392 0.001 0.138 0.591
## .tpqm03.sexo 0.209 0.040 5.279 0.000 0.209 0.843
## .tpqm04.sexo 0.079 0.040 1.968 0.049 0.079 0.503
## .tpqm05.sexo 0.056 0.041 1.366 0.172 0.056 0.316
## .tpqm06.sexo 0.077 0.041 1.864 0.062 0.077 0.382
## .tpqm07.sexo 0.207 0.040 5.166 0.000 0.207 0.753
## .tpqm08.sexo 0.115 0.040 2.829 0.005 0.115 0.566
## .tpqm09.sexo 0.055 0.042 1.325 0.185 0.055 0.262
## .tpqm10.sexo 0.045 0.041 1.106 0.269 0.045 0.281
## .tpqm11.sexo 0.083 0.042 1.996 0.046 0.083 0.338
## .tpqm12.sexo 0.111 0.042 2.635 0.008 0.111 0.383
## .tpqm13.sexo 0.112 0.042 2.658 0.008 0.112 0.374
## .tpqm14.sexo 0.062 0.041 1.526 0.127 0.062 0.367
## .tpqm15.sexo 0.117 0.042 2.811 0.005 0.117 0.421
## .tpqm16.sexo 0.055 0.040 1.388 0.165 0.055 0.503
## .tpqm01.edaddct 0.118 0.041 2.878 0.004 0.118 0.595
## .tpqm02.edaddct 0.149 0.042 3.596 0.000 0.149 0.599
## .tpqm03.edaddct 0.183 0.041 4.477 0.000 0.183 0.726
## .tpqm04.edaddct 0.089 0.041 2.172 0.030 0.089 0.521
## .tpqm05.edaddct 0.099 0.041 2.382 0.017 0.099 0.514
## .tpqm06.edaddct 0.073 0.043 1.708 0.088 0.073 0.340
## .tpqm07.edaddct 0.166 0.042 3.970 0.000 0.166 0.604
## .tpqm08.edaddct 0.088 0.042 2.110 0.035 0.088 0.438
## .tpqm09.edaddct 0.093 0.043 2.178 0.029 0.093 0.400
## .tpqm10.edaddct 0.053 0.042 1.271 0.204 0.053 0.310
## .tpqm11.edaddct 0.108 0.043 2.516 0.012 0.108 0.403
## .tpqm12.edaddct 0.111 0.044 2.521 0.012 0.111 0.360
## .tpqm13.edaddct 0.125 0.044 2.827 0.005 0.125 0.379
## .tpqm14.edaddct 0.049 0.042 1.149 0.251 0.049 0.269
## .tpqm15.edaddct 0.114 0.044 2.602 0.009 0.114 0.383
## .tpqm16.edaddct 0.061 0.041 1.499 0.134 0.061 0.504
## .it 0.227 0.029 7.800 0.000 0.254 0.254
## .cc 0.526 0.032 16.590 0.000 0.604 0.604
## .ia 0.138 0.029 4.808 0.000 0.196 0.196
## .SIQS 0.522 0.026 19.787 0.000 0.783 0.783
## LTp 0.459 0.013 34.000 0.000 1.000 1.000
## LTm 0.400 0.014 29.632 0.000 1.000 1.000
## int.sex_LTp 0.144 0.014 10.446 0.000 1.000 1.000
## int.age_LTp 0.149 0.016 9.414 0.000 1.000 1.000
## int.sex_LTm 0.068 0.009 7.364 0.000 1.000 1.000
## int.age_LTm 0.081 0.013 6.378 0.000 1.000 1.000
## sexo 0.248 0.039 6.358 0.000 0.248 1.000
## edaddicot 0.249 0.039 6.368 0.000 0.249 1.000semPaths(# Argumentos globales
ajuste_LT, what="diagram", whatLabels="std",layout="tree3", rotation = 2, width=50, height=35,exoVar = F,
# Etiquetas
curvePivot=T, curvature = .5
)