Reporte de Practica de R

ANOVA de dos vías

library(tidyverse)

## ── Attaching packages ─────────────────────────────────────── tidyverse 1.3.2 ──
## ✔ ggplot2 3.4.0      ✔ purrr   1.0.0 
## ✔ tibble  3.1.8      ✔ dplyr   1.0.10
## ✔ tidyr   1.2.1      ✔ stringr 1.5.0 
## ✔ readr   2.1.3      ✔ forcats 0.5.2 
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag()    masks stats::lag()

library(datarium)

jobsatisfaction

Analisis exploratorio de datos (EDA)

library(dlookr)

## 
## Attaching package: 'dlookr'

## The following object is masked from 'package:tidyr':
## 
##     extract

## The following object is masked from 'package:base':
## 
##     transform

library(gtsummary)

jobsatisfaction %>% group_by(gender, education_level) %>% describe(score)

Comprobar la normalidad (el supuesto de normal)

jobsatisfaction %>% group_by(gender, education_level) %>%  normality(score)

jobsatisfaction %>% group_by(gender, education_level) %>%  plot_normality(score)

calculo de varianza

PRUEBA DE LEVENE

Cuando p < 0.05 = NO SE ASUME IGUALDAD EN VARIANZAS

Cuando p > 0.05 = SE AS IGUALDAD EN VARIANZAS

La prueba de levene fue no significativa (p = 0.06 ) Podemos asumir que la homeneidad es igual

jobsatisfaction %>% rstatix::levene_test(score ~ gender*education_level)

Visualización de los datos con gráficas

BOXPLOT

library(ggpubr)

bxp <- ggboxplot(jobsatisfaction, x="gender", y="score",
                 fill= "education_level" , palette = "Paired")

bxp

Paletas de colores

valores atípicos (outliers)

con función: identify_outliers

jobsatisfaction %>% group_by(gender, education_level) %>% rstatix::identify_outliers(score)

ANOVA de 2 Vías

la función para anova es: aov o también: anova_test

res.aov<-jobsatisfaction %>% rstatix::anova_test(score ~ gender*education_level)

res.aov

Si se encontró una interacción estadísticamente significativa entre género y el nivel de educación para la puntuación (score) de satisfacción laboral (p = 0.002)

ANÁLISIS POST-HOC

Se trabajar con el paquete: emmeans

#install.packages("emmeans")
library(emmeans)
library(rstatix)

## 
## Attaching package: 'rstatix'

## The following object is masked from 'package:stats':
## 
##     filter

library(gt)

cm <-jobsatisfaction %>% group_by(gender) %>% emmeans_test(score ~ education_level, p.adjust.method =)

jobsatisfaction %>% group_by(gender) %>% emmeans_test(score ~ education_level) %>%  gt()

gender	term	.y.	group1	group2	df	statistic	p	p.adj	p.adj.signif
female	education_level	score	school	college	52	-2.935226	4.952503e-03	1.485751e-02	*
female	education_level	score	school	university	52	-10.834318	6.074300e-15	1.822290e-14	****
female	education_level	score	college	university	52	-7.899092	1.838428e-10	5.515284e-10	****
male	education_level	score	school	college	52	-3.072573	3.374111e-03	1.012233e-02	*
male	education_level	score	school	university	52	-15.295045	6.865425e-21	2.059628e-20	****
male	education_level	score	college	university	52	-12.142652	8.421634e-17	2.526490e-16	****

Visualización de valores p del análisis post hoc en la grafica boxplot