In matt92253/Sta486Finalmtf83: Math 125 data analysis

library(tidyverse)
library(emmeans)
load("~/Sta486Finalmtf83/data/mat125data.rda")

# data becomes more normal when we drop test scores that are zero
# also working with actual tests not pre, practice, honors code or learning aids 
mat125data <- filter( mat125data, score != 0 &
                                  practice_test == 0 & 
                                  learning_aid == 0 & 
                                  honors_code == 0 & 
                                  pre_test == 0)

# create precovid data frame, fall of 2019
f19 <- mat125data %>% filter( year == 19 & season == "fall")
f19$covid <- "fall 2019, before covid"

# create post  covid data frame, fall of 2021
f21 <- mat125data %>% filter( year == 21 & season == "fall")
f21$covid <- "fall 2021, after covid"

# bind the two data frames together for visualization
pre_post <- rbind( f19, f21)

# order the factors
pre_post <- pre_post %>% 
  mutate( covid = fct_relevel( covid,
                               "fall 2019, before covid",
                               "fall 2021, after covid"))

# plot
ggplot( pre_post, aes( x = module_final, y = score, fill = covid)) +
  geom_boxplot() + 
  labs( title = "Pre and Post Covid In Person Module Test Scores")

knitr::kable(anova( lm( score ~ covid, pre_post )))

pre_post_model <- lm( score ~  covid, pre_post )

pre_post_model %>% summary()

plot(pre_post_model, which = 1:2 )

matt92253/Sta486Finalmtf83 documentation built on Aug. 19, 2022, 11:13 p.m.