Read the 2020 data into R.

x <- read.csv("oe2_2020_feedback_anon.csv")

Convert timestamp strings to date format.

x$Timestamp <- dmy_hms(x$Timestamp)

Convert character strings to factors.

x[,3:13] <- lapply(x[,3:13], as.factor)

Convert scores to numeric.

x$Score <- word(x$Score) %>% as.numeric

Subset data to include only the first semester.

x <- x %>% filter(Timestamp < as.Date("2020-08-01"))

Plot heatmap of 2020 discussion leaders’ scores given by participants in the first semester.

x %>% ggplot(aes(x=Who.was.your.discussion.leader.,
                 y = Who.are.YOU.)) +
  geom_tile(aes(fill=Score)) +
  labs(title = "Discussion Leaders' Feedback Scores - 2020 Spring",
       x="participant",
       y = "discussion leader") +
  theme(axis.text.x = element_text(angle=90, hjust=1)) +
  scale_fill_gradient(low = "purple", high = "green")

Comment 1: Each square represents the average of two scores that one participant gave to a discussion leader in one semester.

Comment 2: The participants assessments were mostly positive and discussion leaders scores were generally consistent.

Read the 2021 data into R and convert data types.

y <- read.csv("oe2_2021_feedback_anon.csv")
y$Timestamp <- ymd_hms(y$Timestamp)
y[,3:13] <- lapply(y[,3:13], as.factor)

Plot heatmap of 2021 discussion leaders’ scores given by participants in the first semester.

y %>% ggplot(aes(x=Who.was.your.discussion.leader.,
                 y = Who.are.YOU.)) +
  geom_tile(aes(fill=Score)) +
  labs(title = "Discussion Leaders' Feedback Scores - 2021 Spring", 
  x="participant",
       y = "discussion leader") +
  theme(axis.text.x = element_text(angle=90, hjust=1)) +
  scale_fill_gradient(low = "purple", high = "green")

Comment 2: The participants assessments were mostly negative.

Create a variable for minutes past nine o’clock.

y$hour <- hour(y$Timestamp)
y$minute <- minute(y$Timestamp)
y$timing <- (y$hour - 9)*60 + y$minute

Plot a point graph of the students timing for April 29th.

y %>% filter(Timestamp <= as_date("2021-04-30") &
               Timestamp >= as_date("2021-04-28")) %>%
  filter(timing<100) %>%
  filter(timing>0) %>%
  ggplot(aes(x=Who.are.YOU., y=timing)) +
  geom_point() +
  labs(title = "Participation Times - April 29th",
       x = "participant") +
  theme(axis.text.x = element_text(angle=90, hjust=1))

Comment: Kana and Takayuki had been absent from class yet they still “participated” by providing feedback to the discussion leaders via their Google forms.

Decision: Plot the point graph and include the absentee students’ names in the whole class feedback report so that they may be discouraged from doing so again.

A few weeks later, compare “How was today?” answers.

wk4 <- c(1,1,3,6,5)
wk8 <- c(0,0,1,8,11)
df = rbind(wk4, wk8)
colnames(df) <- c("terrible", "not_so_good", "average", "good", "great")

Run a chi squared test to compare discussion 4 with discussion 8.

chisq.test(df)
## Warning in chisq.test(df): Chi-squared approximation may be incorrect
## 
##  Pearson's Chi-squared test
## 
## data:  df
## X-squared = 5.1549, df = 4, p-value = 0.2718

Comment: There was no difference between the two lessons.