library(dplyr)
df<-read.csv('~/Research/kkex_repo/power_data_110923.csv')
df1 <- read.csv('/data_ext/users/kcz1100/kaylea_dissertation/collabnetXS/raw_data/inst_all_packages_full_results.csv')

hist(df$age/365) #there's a big bump at 9 years, why?
hist(df$contributors) #skewed
hist(log(df$contributors)) #better
hist(df$collaborators) #skewed
hist(log(df$collaborators)) #better
hist(df$milestones)

#one of many ways to generate a dichotomous variable
df$uses_milestones <- case_when(df$milestones != 0 ~ TRUE, 
                                .default = FALSE)

table(df$uses_milestones)

#playing around
cor.test(df$contributors, as.numeric(df$uses_milestones))
cor.test(df$collaborators, as.numeric(df$uses_milestones))
cor.test(df$contributors / df$collaborators, df$age / 365)

t.test(df$age)
#95 percent confidence interval:
# 2793.638 3066.417
t.test(df$contributors)
#95 percent confidence interval:
#  27.6519 154.6866
t.test(df$collaborators)
#95 percent confidence interval:
# 50.01884 96.77090

t.test(df1$up.fac.mean)
#95 percent confidence interval:
# -0.1961401 -0.1647757
df$mmt <- (df$contributors + (2 * df$collaborators)) / (df$contributors + df$collaborators)
t.test(df$mmt)
# 95 percent confidence interval:
# 1.610638 1.684438
#