setwd(dirname(rstudioapi::getSourceEditorContext()$path))  # sets working directory
df <- read.csv(file="movies_data_clean.csv", header=TRUE)
summary(df)
summary(df$Genre)
df$Genre[1..20]
df$Genre[1:20]
setwd(dirname(rstudioapi::getSourceEditorContext()$path))  # sets working directory
df <- read.csv(file="movies_data_clean.csv", header=TRUE)
genres <- df$Genre
comedy_times <- df$Runtime[genres == "Comedy"]
comedy_times
setwd(dirname(rstudioapi::getSourceEditorContext()$path))  # sets working directory
df <- read.csv(file="movies_data_clean.csv", header=TRUE)
comedy_times <- df$Runtime[df$Genre == "Comedy"]
comedy_times
setwd(dirname(rstudioapi::getSourceEditorContext()$path))  # sets working directory
df <- read.csv(file="movies_data_clean.csv", header=TRUE)
comedy_times <- df$Runtime[df$Genre == "Comedy"]
comedy_times
length(comedy_times)
mean(comedy_times)
s <- std_dev_sample <- sd(comedy_times)
s
std_error <- s / sqrt(155)   # 155 is the sample size (N)
std_error
# 68 % CI for comedy run times
mean(comedy_times) - std_error    # lower bound
mean(comedy_times) + std_error   #upper bound
# 95% CI for comedy run times
mean(comedy_times) - 2 * std_error    # lower bound
mean(comedy_times) + 2 * std_error   #upper bound
1 - pnorm(-2) * 2
t.test(comedy_times)
setwd(dirname(rstudioapi::getSourceEditorContext()$path))  # sets working directory
df <- read.csv(file="movies_data_clean.csv", header=TRUE)
comedy_times <- df$Runtime[df$Genre == "Comedy"]
comedy_times
length(comedy_times)
mean(comedy_times)
# 68% CI for comedy run times
mean(comedy_times) - std_error    # lower bound
setwd(dirname(rstudioapi::getSourceEditorContext()$path))  # sets working directory
df <- read.csv(file="movies_data_clean.csv", header=TRUE)
comedy_times <- df$Runtime[df$Genre == "Comedy"]
comedy_times
length(comedy_times)
mean(comedy_times)
s <- std_dev_sample <- sd(comedy_times)
s
std_error <- s / sqrt(155)   # 155 is the sample size (N)
std_error
# 68% CI for comedy run times
mean(comedy_times) - std_error    # lower bound
mean(comedy_times) + std_error   #upper bound
# 95% CI for comedy run times --- 1.96 is more precise than 2
mean(comedy_times) - 2 * std_error    # lower bound
mean(comedy_times) + 2 * std_error   #upper bound
t.test(comedy_times)
drama_times <- df$Runtime[df$Genre == "Drama"]
drama_times
length(drama_times)
mean(drama_times)
s <- std_dev_sample <- sd(drama_times)
s
std_error <- s / sqrt(289)
std_error
# 68% CI for drama run times
mean(drama_times) - std_error    # lower bound
mean(drama_times) + std_error   #upper bound
# 95% CI for comedy run times --- 1.96 is more precise than 2
mean(drama_times) - 2 * std_error    # lower bound
mean(drama_times) + 2 * std_error   #upper bound
t.test(drama_times)
t.test(comedy_times, drama_times)
