Creating the environment

library(tidyverse)

── Attaching packages ─────────────────────────────────────── tidyverse 1.3.1 ──

✔ ggplot2 3.3.5     ✔ purrr   0.3.4
✔ tibble  3.1.6     ✔ dplyr   1.0.8
✔ tidyr   1.2.0     ✔ stringr 1.4.0
✔ readr   2.1.2     ✔ forcats 0.5.1

── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
✖ dplyr::filter() masks stats::filter()
✖ dplyr::lag()    masks stats::lag()

Getting data

juan_data <- 
  read_csv("https://docs.google.com/spreadsheets/d/1t-dtqGcFY-QM5YxsScp4z7GDMGdQWp_cqNalOb-SYXE/export?format=csv&gid=1362921518") %>% 
  select(1:6)

New names:
Rows: 204 Columns: 12
── Column specification
──────────────────────────────────────────────────────── Delimiter: "," chr
(6): colegio, test, ...7, ...8, ...9, PROMEDIOS dbl (6): id, estra_pens,
estilo_aprendizaje, performance, PRETEST, POSTEST
ℹ Use `spec()` to retrieve the full column specification for this data. ℹ
Specify the column types or set `show_col_types = FALSE` to quiet this message.
• `` -> `...7`
• `` -> `...8`
• `` -> `...9`

Anova

juan_pretest <- 
  juan_data %>% 
  filter(test == "pretest")

summary(aov(formula = performance ~ colegio, 
    data = juan_pretest))

            Df Sum Sq Mean Sq F value Pr(>F)
colegio      2   0.17  0.0872   0.136  0.873
Residuals   99  63.44  0.6408               

No hay diferencias significativas entre las medias del desempeño de los colegios en el pretest Pr(>F) > 0.873

juan_postest <- 
  juan_data %>% 
  filter(test == "posttest")

anova_postest <- 
  aov(formula = performance ~ colegio, 
    data = juan_postest)

summary(anova_postest)

            Df Sum Sq Mean Sq F value Pr(>F)    
colegio      2  27.79  13.894    19.2  9e-08 ***
Residuals   99  71.65   0.724                   
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Hay diferencias significativas entre las medias del desempeño de los colegios en el postest. Pero no sabemos entre que colegios hay diferencias; por lo tanto, se suguiere realizar un test de comparaciones múltiples (poshoc: tukey).

Posthoc

TukeyHSD(x = anova_postest)

  Tukey multiple comparisons of means
    95% family-wise confidence level

Fit: aov(formula = performance ~ colegio, data = juan_postest)

$colegio
                           diff        lwr        upr     p adj
colegio_2-colegio_1 -1.09558824 -1.5865425 -0.6046339 0.0000020
colegio_3-colegio_1 -1.11852941 -1.6094837 -0.6275751 0.0000012
colegio_3-colegio_2 -0.02294118 -0.5138955  0.4680131 0.9932079