mgaughan/24_deb_pkg_gov
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

adding some results and analysis

Browse Source
This commit is contained in:
mjgaughan 2024-06-28 00:40:08 -04:00
parent 979dd9ccc6
commit b4a55f464e
10 changed files with 405 additions and 292 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

580
R/.Rhistory
View File

@ -1,293 +1,3 @@
contrib_readme_model <- load(file = "final_models/0623_pop_rm_contrib.rda")
contrib_readme_model <- load(file = "final_models/0623_pop_rm_contrib.rda")
contrib_readme_model <- load("final_models/0623_pop_rm_contrib.rda")
contrib_readme_model <- source("final_models/0623_pop_rm_contrib.rda")
contrib_readme_model <- readRDS("final_models/0623_pop_rm_contrib.rda")
contrib_readme_model <- readRDS("final_models/0623_pop_contrib_collab.rda")
collab_readme_model <- readRDS("final_models/0623_pop_rm_collab.rda")
texreg(list(collab_readme_model, contrib_readme_model), stars=NULL, digits=2,
custom.model.names=c( 'collab','contrib.' ),
custom.coef.names=c('(Intercept)', 'after_introduction'),
use.packages=FALSE, table=FALSE, ci.force = TRUE)
library(texreg)
texreg(list(collab_readme_model, contrib_readme_model), stars=NULL, digits=2,
custom.model.names=c( 'collab','contrib.' ),
custom.coef.names=c('(Intercept)', 'after_introduction'),
use.packages=FALSE, table=FALSE, ci.force = TRUE)
texreg(list(collab_readme_model, contrib_readme_model), stars=NULL, digits=2,
custom.model.names=c( 'collab','contrib.' ),
custom.coef.names=c('(Intercept)', 'after_introduction' 'etc'),
texreg(list(collab_readme_model, contrib_readme_model), stars=NULL, digits=2,
custom.model.names=c( 'collab','contrib.' ),
custom.coef.names=c('(Intercept)', 'after_introduction', 'etc'),
use.packages=FALSE, table=FALSE, ci.force = TRUE)
library(tidyverse)
library(plyr)
library(stringr)
try(setwd(dirname(rstudioapi::getActiveDocumentContext()$path)))
#load in data
contrib_df <- read_csv("../final_data/deb_contrib_pop_change.csv")
readme_df <- read_csv("../final_data/deb_readme_pop_change.csv")
#some expansion needs to happens for each project
expand_timeseries <- function(project_row) {
longer <- project_row |>
pivot_longer(cols = ends_with("new"),
names_to = "window",
values_to = "count") |>
unnest(count) |>
mutate(after_doc = as.numeric(str_detect(window, "after"))) |>
mutate(is_collab = as.numeric(str_detect(window, "collab")))
return(longer)
}
expanded_readme_data <- expand_timeseries(readme_df[1,])
for (i in 2:nrow(readme_df)){
expanded_readme_data <- rbind(expanded_readme_data, expand_timeseries(readme_df[i,]))
}
expanded_contrib_data <- expand_timeseries(contrib_df[1,])
for (i in 2:nrow(contrib_df)){
expanded_contrib_data <- rbind(expanded_contrib_data, expand_timeseries(contrib_df[i,]))
}
expanded_readme_data$log1pcount <- log1p(expanded_readme_data$count)
expanded_contrib_data$log1pcount <- log1p(expanded_contrib_data$count)
expanded_readme_data$logcount <- log(expanded_readme_data$count)
expanded_contrib_data$logcount <- log(expanded_contrib_data$count)
#breaking out the types of population counts
collab_pop_readme <- expanded_readme_data[which(expanded_readme_data$is_collab == 1),]
contrib_pop_readme <- expanded_readme_data[which(expanded_readme_data$is_collab == 0),]
collab_pop_contrib <- expanded_contrib_data[which(expanded_contrib_data$is_collab == 1),]
contrib_pop_contrib <- expanded_contrib_data[which(expanded_contrib_data$is_collab == 0),]
simple_collab_readme_model <- glm.nb(count ~ as.factor(after_doc), data=collab_pop_readme)
summary(simple_collab_readme_model)
anova(simple_collab_readme_model, collab_readme_model)
summary(collab_readme_model)
#load in data
full_df <- read_csv("../final_data/deb_full_data.csv")
contrib_df <- read_csv("../final_data/deb_contrib_pop_change.csv")
contrib_df <- merge(full_df, contrib_df, by="upstream_vcs_link")
View(contrib_df)
View(contrib_df)
readme_df <- read_csv("../final_data/deb_readme_pop_change.csv")
readme_df <- merge(full_df, readme_df, by="upstream_vcs_link")
# age is calculated against December 11, 2023
contrib_df |>
mutate(start_date = as.Date("2023-12-11") - age_of_project)
View(contrib_df)
# age is calculated against December 11, 2023
contrib_df <- contrib_df |>
mutate(start_date = as.Date("2023-12-11") - age_of_project)
View(contrib_df)
View(contrib_df)
View(readme_df)
readme_df <- readme_df |>
mutate(start_date = as.Date("2023-12-11") - age_of_project)
View(readme_df)
collab_readme_model_plus <- glmer.nb(log1pcount ~ as.factor(after_doc) + event_date + (after_doc| upstream_vcs_link), data=collab_pop_readme)
#some expansion needs to happens for each project
expand_timeseries <- function(project_row) {
longer <- project_row |>
pivot_longer(cols = ends_with("new"),
names_to = "window",
values_to = "count") |>
unnest(count) |>
mutate(after_doc = as.numeric(str_detect(window, "after"))) |>
mutate(is_collab = as.numeric(str_detect(window, "collab")))
return(longer)
}
expanded_readme_data <- expand_timeseries(readme_df[1,])
for (i in 2:nrow(readme_df)){
expanded_readme_data <- rbind(expanded_readme_data, expand_timeseries(readme_df[i,]))
}
expanded_contrib_data <- expand_timeseries(contrib_df[1,])
for (i in 2:nrow(contrib_df)){
expanded_contrib_data <- rbind(expanded_contrib_data, expand_timeseries(contrib_df[i,]))
}
expanded_readme_data$log1pcount <- log1p(expanded_readme_data$count)
expanded_contrib_data$log1pcount <- log1p(expanded_contrib_data$count)
expanded_readme_data$logcount <- log(expanded_readme_data$count)
expanded_contrib_data$logcount <- log(expanded_contrib_data$count)
#breaking out the types of population counts
collab_pop_readme <- expanded_readme_data[which(expanded_readme_data$is_collab == 1),]
contrib_pop_readme <- expanded_readme_data[which(expanded_readme_data$is_collab == 0),]
collab_pop_contrib <- expanded_contrib_data[which(expanded_contrib_data$is_collab == 1),]
contrib_pop_contrib <- expanded_contrib_data[which(expanded_contrib_data$is_collab == 0),]
library(tidyverse)
library(plyr)
library(stringr)
try(setwd(dirname(rstudioapi::getActiveDocumentContext()$path)))
#load in data
full_df <- read_csv("../final_data/deb_full_data.csv")
contrib_df <- read_csv("../final_data/deb_contrib_pop_change.csv")
View(contrib_df)
readme_df <- read_csv("../final_data/deb_readme_pop_change.csv")
contrib_df <- merge(full_df, contrib_df, by="upstream_vcs_link")
readme_df <- merge(full_df, readme_df, by="upstream_vcs_link")
# age is calculated against December 11, 2023
contrib_df <- contrib_df |>
mutate(start_date = as.Date("2023-12-11") - age_of_project) +
mutate(event_date_days = as.Date("2024-06-24") - event_date) +
readme_df <- readme_df |>
mutate(start_date = as.Date("2023-12-11") - age_of_project)
# age is calculated against December 11, 2023
contrib_df <- contrib_df |>
mutate(start_date = as.Date("2023-12-11") - age_of_project) +
mutate(event_date_days = as.Date("2024-06-24") - event_date)
# age is calculated against December 11, 2023
contrib_df <- contrib_df |>
mutate(start_date = as.Date("2023-12-11") - age_of_project) +
mutate(event_date_days = diff.Date(as.Date("2023-12-11"),event_date, units = "days"))
# age is calculated against December 11, 2023
contrib_df <- contrib_df |>
mutate(start_date = as.Date("2023-12-11") - age_of_project) |>
mutate(event_date_days = diff.Date(as.Date("2023-12-11"),event_date, units = "days"))
# age is calculated against December 11, 2023
contrib_df <- contrib_df |>
mutate(start_date = as.Date("2023-12-11") - age_of_project) |>
mutate(event_date_days = diff.Date(as.Date("2024-06-24"),event_date, units = "days"))
# age is calculated against December 11, 2023
contrib_df <- contrib_df |>
mutate(start_date = as.Date("2023-12-11") - age_of_project) |>
mutate(event_date_days = diff.Date(as.Date("2024-06-24"),as.Date(event_date), units = "days"))
View(contrib_df)
View(contrib_df)
# age is calculated against December 11, 2023
contrib_df <- contrib_df |>
mutate(start_date = as.Date("2023-12-11") - age_of_project) |>
mutate(event_date_days = diff.Date(as.Date("2024-06-24"),as.POSIXct(event_date, format = "%Y-%m-%d %H:%M:%S"), units = "days"))
# age is calculated against December 11, 2023
contrib_df <- contrib_df |>
mutate(start_date = as.Date("2023-12-11") - age_of_project) |>
mutate(event_date_days =
as.numeric(
difftime(as.POSIXct("2024-06-24 00:00:00", format = "%Y-%m-%d %H:%M:%S")
as.POSIXct(event_date, format = "%Y-%m-%d %H:%M:%S"),
# age is calculated against December 11, 2023
contrib_df <- contrib_df |>
mutate(start_date = as.Date("2023-12-11") - age_of_project) |>
mutate(event_date_days =
as.numeric(
difftime(as.POSIXct("2024-06-24 00:00:00", format = "%Y-%m-%d %H:%M:%S"),
as.POSIXct(event_date, format = "%Y-%m-%d %H:%M:%S"),
units = "days")))
View(contrib_df)
readme_df <- readme_df |>
mutate(start_date = as.Date("2023-12-11") - age_of_project) |>
mutate(event_date_days =
as.numeric(
difftime(as.POSIXct("2024-06-24 00:00:00", format = "%Y-%m-%d %H:%M:%S"),
as.POSIXct(event_date, format = "%Y-%m-%d %H:%M:%S"),
units = "days")))
#some expansion needs to happens for each project
expand_timeseries <- function(project_row) {
longer <- project_row |>
pivot_longer(cols = ends_with("new"),
names_to = "window",
values_to = "count") |>
unnest(count) |>
mutate(after_doc = as.numeric(str_detect(window, "after"))) |>
mutate(is_collab = as.numeric(str_detect(window, "collab")))
return(longer)
}
expanded_readme_data <- expand_timeseries(readme_df[1,])
for (i in 2:nrow(readme_df)){
expanded_readme_data <- rbind(expanded_readme_data, expand_timeseries(readme_df[i,]))
}
expanded_contrib_data <- expand_timeseries(contrib_df[1,])
for (i in 2:nrow(contrib_df)){
expanded_contrib_data <- rbind(expanded_contrib_data, expand_timeseries(contrib_df[i,]))
}
expanded_readme_data$log1pcount <- log1p(expanded_readme_data$count)
expanded_contrib_data$log1pcount <- log1p(expanded_contrib_data$count)
expanded_readme_data$logcount <- log(expanded_readme_data$count)
expanded_contrib_data$logcount <- log(expanded_contrib_data$count)
#breaking out the types of population counts
collab_pop_readme <- expanded_readme_data[which(expanded_readme_data$is_collab == 1),]
contrib_pop_readme <- expanded_readme_data[which(expanded_readme_data$is_collab == 0),]
collab_pop_contrib <- expanded_contrib_data[which(expanded_contrib_data$is_collab == 1),]
contrib_pop_contrib <- expanded_contrib_data[which(expanded_contrib_data$is_collab == 0),]
#import models
library(lme4)
library(optimx)
collab_readme_model_plus <- glmer.nb(log1pcount ~ as.factor(after_doc) + event_date_days + (after_doc| upstream_vcs_link), data=collab_pop_readme)
anova(collab_readme_model_plus, collab_readme_model)
collab_readme_model <- readRDS("final_models/0623_pop_rm_collab.rda")
anova(collab_readme_model_plus, collab_readme_model)
saveRDS(collab_readme_model, "final_models/0623_pop_rm_collab_better.rda")
summary(collab_readme_model_plus)
summary(collab_readme_model)
library(tidyverse)
#things to get:
# - delete old age column
# - normal age, in date
# - age from today in days
# - delta between first commit and document in days
#README Document updates
#loading in new ages
####RDD CSV
first_commit_df <- read_csv("../062424_did_first_commit_readme.csv")
first_commit_df_2 <- read_csv("../062424_did_first_commit_readme_2.csv")
first_commit_df <- rbind(first_commit_df, first_commit_df_2)
# need to first do an rbind with this data and the second file
# check with the head of the file/size of the file
old_rdd_readme <- read_csv("../final_data/deb_readme_did.csv")
old_rdd_readme <- merge(old_rdd_readme, first_commit_df, by="upstream_vcs_link")
new_rm_data <- old_rdd_readme |>
select(-age_of_project) |>
mutate(first_commit_dt = as.POSIXct(first_commit,
format = "%a %b %d %H:%M:%S %Y %z")) |>
mutate(age_in_days =
as.numeric(
difftime(
as.POSIXct("2024-06-24 00:00:00", format = "%Y-%m-%d %H:%M:%S"),
first_commit_dt,
units = "days"))) |>
mutate (event_gap =
as.numeric(
difftime(
as.POSIXct(event_date, format = "%Y-%m-%d %H:%M:%S"),
first_commit_dt,
units = "days")))
View(old_rdd_readme)
new_rm_data <- old_rdd_readme |>
select(-c(age_of_project)) |>
mutate(first_commit_dt = as.POSIXct(first_commit,
format = "%a %b %d %H:%M:%S %Y %z")) |>
mutate(age_in_days =
as.numeric(
difftime(
as.POSIXct("2024-06-24 00:00:00", format = "%Y-%m-%d %H:%M:%S"),
first_commit_dt,
units = "days"))) |>
mutate (event_gap =
as.numeric(
difftime(
as.POSIXct(event_date, format = "%Y-%m-%d %H:%M:%S"),
first_commit_dt,
units = "days")))
new_rm_data <- old_rdd_readme |>
mutate(first_commit_dt = as.POSIXct(first_commit,
format = "%a %b %d %H:%M:%S %Y %z")) |>
mutate(age_in_days =
as.numeric(
difftime(
as.POSIXct("2024-06-24 00:00:00", format = "%Y-%m-%d %H:%M:%S"),
first_commit_dt,
units = "days"))) |>
mutate (event_gap =
as.numeric(
difftime(
as.POSIXct(event_date, format = "%Y-%m-%d %H:%M:%S"),
first_commit_dt,
units = "days")))
View(new_rm_data)
new_rm_data <- new_rm_data |>
select(-age_of_project)
new_rm_data$age_of_project = NULL
head(new_rm_data)
write.csv(new_rm_data, file = "../final_data/deb_readme_did_updated.csv", row.names = FALSE)
old_pop_readme <- merge(old_pop_readme, first_commit_df, by="upstream_vcs_link")
####PopChange CSV
old_pop_readme <- read_csv("../final_data/deb_readme_pop_change.csv")
old_pop_readme <- merge(old_pop_readme, first_commit_df, by="upstream_vcs_link")
new_pop_data <- old_pop_readme |> new_pop_data <- old_pop_readme |>
mutate(first_commit_dt = as.POSIXct(first_commit, mutate(first_commit_dt = as.POSIXct(first_commit,
format = "%a %b %d %H:%M:%S %Y %z")) |> format = "%a %b %d %H:%M:%S %Y %z")) |>
@ -510,3 +220,293 @@ summary(contrib_contrib_model)
# optCtrl=list(maxfun=2e5)), nAGQ=0, data=all_actions_data) # optCtrl=list(maxfun=2e5)), nAGQ=0, data=all_actions_data)
all_gmodel <- readRDS("0512_contrib_all.rda") all_gmodel <- readRDS("0512_contrib_all.rda")
summary(all_gmodel) summary(all_gmodel)
#all_gmodel <- glmer.nb(log1p_count ~ D * week_offset + scaled_project_age + scaled_event_gap + (D * week_offset | upstream_vcs_link),
# control=glmerControl(optimizer="bobyqa",
# optCtrl=list(maxfun=2e5)), nAGQ=0, data=all_actions_data)
all_gmodel <- readRDS("0512_contrib_all.rda")
#identifying the quartiles of effect for D
test_condvals <- broom.mixed::tidy(all_gmodel, effects = "ran_vals", conf.int = TRUE)
test_glmer_ranef_D <- test_condvals [which(test_condvals $term == "D"),]
has_zero <- function(estimate, low, high){
return(ifelse((low < 0),ifelse((high > 0), 1, 0), 2))
}
test_glmer_ranef_D <- test_glmer_ranef_D |>
mutate(ranef_grouping = has_zero(estimate, conf.low, conf.high)) |>
mutate(rank = rank(estimate))
g <- test_glmer_ranef_D |>
ggplot(aes(x=rank, y=estimate, col = as.factor(ranef_grouping))) +
geom_linerange(aes(ymin= conf.low, ymax= conf.high)) +
theme_bw()
library(tidyverse)
g <- test_glmer_ranef_D |>
ggplot(aes(x=rank, y=estimate, col = as.factor(ranef_grouping))) +
geom_linerange(aes(ymin= conf.low, ymax= conf.high)) +
theme_bw()
g
test_glmer_ranef_D <- test_glmer_ranef_D |>
mutate(ranef_grouping = has_zero(estimate, conf.low, conf.high)) |>
mutate(rank = rank(estimate))
g <- test_glmer_ranef_D |>
ggplot(aes(x=rank, y=estimate, col = as.factor(ranef_grouping))) +
geom_linerange(aes(ymin= conf.low, ymax= conf.high)) +
theme_bw()
g
library(tidyverse)
library(plyr)
library(gridExtra)
library(ggpubr)
# script for the analysis of document readability metrics
# readability metrics will be studied controlled by their length
# gaughan@u.northwestern.edu
# loading in the data
try(setwd(dirname(rstudioapi::getActiveDocumentContext()$path)))
source("~/Desktop/git/24_deb_gov/R/documentReadabilityAnalysis.R", echo=TRUE)
source("~/Desktop/git/24_deb_gov/R/documentReadabilityAnalysis.R", echo=TRUE)
source("~/Desktop/git/24_deb_gov/R/documentReadabilityAnalysis.R", echo=TRUE)
aggregate(readme_df[, 3:10], list(readme_df$subdir), median)
readme_df <- read_csv("../text_analysis/dwo_readability_readme.csv")
aggregate(readme_df[, 3:10], list(readme_df$subdir), median)
aggregate(contributing_df[, 3:10], list(contributing_df$subdir), median)
aggregate(readme_df[, 3:10], list(readme_df$subdir), median)
readme_df <- read_csv("../text_analysis/dwo_readability_readme.csv")
contributing_df <- read_csv("../text_analysis/dwo_readability_contributing.csv")
#getting basic stats for the readme readability
median(readme_df$flesch_reading_ease)
median(readme_df$linsear_write_formula)
readme_rdd <- readRDS("final_models/0624_readme_all_rdd.rda")
contrib_rdd <- readRDS("final_models/0624_contrib_all_rdd.rda")
textreg(list(readme_rdd, contrib_rdd), stars=NULL, digits=3, use.packages=FALSE, table=FALSE, ci.force = TRUE))
textreg(list(readme_rdd, contrib_rdd), stars=NULL, digits=3, use.packages=FALSE, table=FALSE, ci.force = TRUE)
reg(list(readme_rdd, contrib_rdd), stars=NULL, digits=3, use.packages=FALSE, table=FALSE, ci.force = TRUE)
texreg(list(readme_rdd, contrib_rdd), stars=NULL, digits=3, use.packages=FALSE, table=FALSE, ci.force = TRUE)
library(texreg)
texreg(list(readme_rdd, contrib_rdd), stars=NULL, digits=3, use.packages=FALSE, table=FALSE, ci.force = TRUE)
summary(readme)
summary(readme_rdd)
texreg(list(readme_rdd, contrib_rdd), stars=NULL, digits=3, use.packages=FALSE,
custom.model.names=c( 'README','CONTRIBUTING'),
custom.coef.names=c('(Intercept)', 'Indtroduction', 'Week (Time)', 'Project Age', 'Introduction:Week', 'Event Gap'),
table=FALSE, ci.force = TRUE)
readme_groupings <- read.csv('../final_data/deb_readme_interaction_groupings.csv')
contrib_groupings <- read.csv('../final_data/deb_contrib_interaction_groupings.csv')
View(readme_groupings)
library(tidyverse)
readme_g <- readme_groupings |>
ggplot(aes(x=rank, y=estimate, col = as.factor(ranef_grouping))) +
geom_linerange(aes(ymin= conf.low, ymax= conf.high)) +
theme_bw()
readme_g
subdirColors <-
setNames( c('firebrick1', 'forestgreen', 'cornflowerblue')
, c(0,1,2) )
readme_g <- readme_groupings |>
ggplot(aes(x=rank, y=estimate, col = as.factor(ranef_grouping))) +
geom_linerange(aes(ymin= conf.low, ymax= conf.high)) +
scale_color_manual(values = subdirColors) +
theme_bw()
readme_g
contrib_groupings <- read.csv('../final_data/deb_contrib_interaction_groupings.csv')
contrib_g <- contrib_groupings |>
ggplot(aes(x=rank, y=estimate, col = as.factor(ranef_grouping))) +
geom_linerange(aes(ymin= conf.low, ymax= conf.high)) +
scale_color_manual(values = subdirColors) +
theme_bw()
contrib_g
grid.arrange(readme_g, contrib_g, nrow = 1)
library(gridExtra)
grid.arrange(readme_g, contrib_g, nrow = 1)
readme_g <- readme_groupings |>
ggplot(aes(x=rank, y=estimate, col = as.factor(ranef_grouping))) +
geom_linerange(aes(ymin= conf.low, ymax= conf.high)) +
scale_color_manual(values = subdirColors) +
guides(fill="none", color="none")+
theme_bw()
readme_g
grid.arrange(readme_g, contrib_g, nrow = 1)
grid.arrange(contrib_g, readme_g, nrow = 1)
contrib_g <- contrib_groupings |>
ggplot(aes(x=rank, y=estimate, col = as.factor(ranef_grouping))) +
geom_linerange(aes(ymin= conf.low, ymax= conf.high)) +
scale_color_manual(values = subdirColors) +
theme_bw() +
theme(legend.position = "top")
grid.arrange(contrib_g, readme_g, nrow = 1)
library(jtools)
plot_summs(readme_rdd, contrib_rdd)
?plot_summs
plot_summs(readme_rdd, contrib_rdd, plot.distributions = TRUE)
col_order <- c("upstream_vcs_link", "age_in_days", "first_commit", "first_commit_dt", "event_gap", "event_date", "event_hash", "before_all_ct", "after_all_ct", "before_mrg_ct", "after_mrg_ct", "before_auth_new", "after_auth_new", "before_commit_new", "after_commit_new")
contrib_df <- contrib_df[,col_order]
contrib_df$ct_before_all <- str_split(gsub("[][]","", contrib_df$before_all_ct), ", ")
contrib_df <- read_csv('../final_data/deb_contrib_did.csv')
col_order <- c("upstream_vcs_link", "age_in_days", "first_commit", "first_commit_dt", "event_gap", "event_date", "event_hash", "before_all_ct", "after_all_ct", "before_mrg_ct", "after_mrg_ct", "before_auth_new", "after_auth_new", "before_commit_new", "after_commit_new")
contrib_df <- contrib_df[,col_order]
contrib_df$ct_before_all <- str_split(gsub("[][]","", contrib_df$before_all_ct), ", ")
contrib_df$ct_after_all <- str_split(gsub("[][]","", contrib_df$after_all_ct), ", ")
contrib_df$ct_before_mrg <- str_split(gsub("[][]","", contrib_df$before_mrg_ct), ", ")
contrib_df$ct_after_mrg <- str_split(gsub("[][]","", contrib_df$after_mrg_ct), ", ")
drop <- c("before_all_ct", "before_mrg_ct", "after_all_ct", "after_mrg_ct")
contrib_df = contrib_df[,!(names(contrib_df) %in% drop)]
# 2 some expansion needs to happens for each project
expand_timeseries <- function(project_row) {
longer <- project_row |>
pivot_longer(cols = starts_with("ct"),
names_to = "window",
values_to = "count") |>
unnest(count)
longer$observation_type <- gsub("^.*_", "", longer$window)
longer <- ddply(longer, "observation_type", transform, week=seq(from=0, by=1, length.out=length(observation_type)))
longer$count <- as.numeric(longer$count)
#longer <- longer[which(longer$observation_type == "all"),]
return(longer)
}
expanded_data <- expand_timeseries(contrib_df[1,])
library(plyr)
contrib_df <- read_csv('../final_data/deb_contrib_did.csv')
col_order <- c("upstream_vcs_link", "age_in_days", "first_commit", "first_commit_dt", "event_gap", "event_date", "event_hash", "before_all_ct", "after_all_ct", "before_mrg_ct", "after_mrg_ct", "before_auth_new", "after_auth_new", "before_commit_new", "after_commit_new")
contrib_df <- contrib_df[,col_order]
contrib_df$ct_before_all <- str_split(gsub("[][]","", contrib_df$before_all_ct), ", ")
contrib_df$ct_after_all <- str_split(gsub("[][]","", contrib_df$after_all_ct), ", ")
contrib_df$ct_before_mrg <- str_split(gsub("[][]","", contrib_df$before_mrg_ct), ", ")
contrib_df$ct_after_mrg <- str_split(gsub("[][]","", contrib_df$after_mrg_ct), ", ")
drop <- c("before_all_ct", "before_mrg_ct", "after_all_ct", "after_mrg_ct")
contrib_df = contrib_df[,!(names(contrib_df) %in% drop)]
# 2 some expansion needs to happens for each project
expand_timeseries <- function(project_row) {
longer <- project_row |>
pivot_longer(cols = starts_with("ct"),
names_to = "window",
values_to = "count") |>
unnest(count)
longer$observation_type <- gsub("^.*_", "", longer$window)
longer <- ddply(longer, "observation_type", transform, week=seq(from=0, by=1, length.out=length(observation_type)))
longer$count <- as.numeric(longer$count)
#longer <- longer[which(longer$observation_type == "all"),]
return(longer)
}
expanded_data <- expand_timeseries(contrib_df[1,])
for (i in 2:nrow(contrib_df)){
expanded_data <- rbind(expanded_data, expand_timeseries(contrib_df[i,]))
}
View(expand_timeseries)
View(expanded_data)
window_num <- 8
windowed_data <- expanded_data |>
filter(week >= (27 - window_num) & week <= (27 + window_num)) |>
mutate(D = ifelse(week > 27, 1, 0))
windowed_data$week_offset <- windowed_data$week - 27
View(windowed_data)
time_plot <- expanded_data |>
ggplot(aes(x=week_offset, y=count))
time_plot
time_plot <- windowed_data |>
ggplot(aes(x=week_offset, y=count))
time_plot
time_plot <- windowed_data |>
ggplot(aes(x=week_offset, y=count)) +
geom_point()
time_plot
time_plot <- windowed_data |>
ggplot(aes(x=week_offset, y=median(count))) +
geom_point()
time_plot
time_plot <- windowed_data |>
ggplot(aes(x=week_offset, y=mean(count))) +
geom_point()
time_plot
time_plot <- windowed_data |>
ggplot(aes(x=week_offset, y=count)) +
geom_point()
time_plot
all_actions_data <- windowed_data[which(windowed_data$observation_type == "all"),]
all_actions_data$log1p_count <- log1p(all_actions_data$count)
time_plot <- all_actions_data |>
ggplot(aes(x=week_offset, y=log1p_count)) +
geom_point()
time_plot
time_plot <- all_actions_data |>
ggplot(aes(x=week_offset, y=log1p_count)) +
geom_smooth()+
geom_point()
time_plot
time_plot <- all_actions_data |>
ggplot(aes(x=week_offset, y=log1p_count)) +
geom_smooth()
time_plot
time_plot <- all_actions_data |>
ggplot(aes(x=week_offset, y=log1p_count)) +
geom_smooth() +
theme_bw()
time_plot
windowed_readme_data$week_offset <- windowed_readme_data$week - 27
all_actions_readme_data <- windowed_readme_data[which(windowed_readme_data$observation_type == "all"),]
source("~/Desktop/git/24_deb_gov/R/gam_plot_documents.R")
time_plot
time_plot <- all_actions_data |>
ggplot(aes(x=week_offset, y=log1p_count, color=factor(document_type))) +
geom_smooth() +
theme_bw()
time_plot
View(expanded_readme_data)
mean(all_actions_readme_data$event_gap)
mean(median$event_gap)
median(all_actions_readme_data$event_gap)
mean(all_actions_readme_data$event_gap)
mean(all_actions_contrib_data$event_gap)
median(all_actions_contrib_data$event_gap)
time_plot <- all_actions_data |>
ggplot(aes(x=week_offset, y=log1p_count, color=factor(document_type))) +
geom_smooth() +
theme_bw() +
theme(legend.position = "top")
time_plot
time_plot <- all_actions_data |>
ggplot(aes(x=week_offset, y=log1p_count, color=factor(document_type))) +
geom_smooth() +
geom_vline(x=0)
theme_bw() +
theme(legend.position = "top")
time_plot <- all_actions_data |>
ggplot(aes(x=week_offset, y=log1p_count, color=factor(document_type))) +
geom_smooth() +
geom_vline(x=0)+
theme_bw() +
theme(legend.position = "top")
time_plot
time_plot <- all_actions_data |>
ggplot(aes(x=week_offset, y=log1p_count, color=factor(document_type))) +
geom_smooth() +
geom_vline(x=0)+
theme_bw() +
theme(legend.position = "top")
time_plot <- all_actions_data |>
ggplot(aes(x=week_offset, y=log1p_count, color=factor(document_type))) +
geom_smooth() +
geom_vline(0)+
theme_bw() +
theme(legend.position = "top")
time_plot
time_plot <- all_actions_data |>
ggplot(aes(x=week_offset, y=log1p_count, color=factor(document_type))) +
geom_smooth() +
geom_vline(xintercept = 0)+
theme_bw() +
theme(legend.position = "top")
time_plot
#looking at event gap
document_event_gap <- ggplot(all_actions_data, aes(x=event_gap, group=as.factor(document_type))) +
geom_density(aes(color = as.factor(document_type), fill=as.factor(document_type)), alpha=0.2, position="identity") +
theme_bw()
document_event_gap
#looking at event gap
document_event_gap <- ggplot(all_actions_data, aes(x=scale(event_gap), group=as.factor(document_type))) +
geom_density(aes(color = as.factor(document_type), fill=as.factor(document_type)), alpha=0.2, position="identity") +
theme_bw()
document_event_gap
#looking at event gap
mean(all_actions_readme_data$event_gap)
sd(all_actions_readme_data$event_gap)
mean(all_actions_contrib_data$event_gap)
sd(all_actions_contrib_data$event_gap)
mode(all_actions_contrib_data$event_gap)
mean(all_actions_contrib_data$event_gap)

BIN
R/d_working_readability_plot.png Normal file
View File

Binary file not shown.
Side by Side

After

Width:  |  Height:  |  Size: 127 KiB

5
R/documentReadabilityAnalysis.R
View File

@ -7,10 +7,11 @@ library(ggpubr)
# gaughan@u.northwestern.edu # gaughan@u.northwestern.edu
# loading in the data # loading in the data
try(setwd(dirname(rstudioapi::getActiveDocumentContext()$path))) try(setwd(dirname(rstudioapi::getActiveDocumentContext()$path)))
readme_df <- read_csv("../text_analysis/draft_readability_readme.csv") readme_df <- read_csv("../text_analysis/dwo_readability_readme.csv")
contributing_df <- read_csv("../text_analysis/draft_readability_contributing.csv") contributing_df <- read_csv("../text_analysis/dwo_readability_contributing.csv")
head(readme_df) head(readme_df)
aggregate(readme_df[, 3:10], list(readme_df$subdir), median) aggregate(readme_df[, 3:10], list(readme_df$subdir), median)
aggregate(contributing_df[, 3:10], list(contributing_df$subdir), median)
#getting basic stats for the readme readability #getting basic stats for the readme readability
median(readme_df$flesch_reading_ease) median(readme_df$flesch_reading_ease)
median(readme_df$linsear_write_formula) median(readme_df$linsear_write_formula)

BIN
R/draft_Rdd_groupings_plot.png Normal file
View File

Binary file not shown.
Side by Side

After

Width:  |  Height:  |  Size: 46 KiB

BIN
R/draft_average_over_time_plot.png Normal file
View File

Binary file not shown.
Side by Side

After

Width:  |  Height:  |  Size: 48 KiB

BIN
R/draft_averages_plot.png Normal file
View File

Binary file not shown.
Side by Side

After

Width:  |  Height:  |  Size: 48 KiB

BIN
R/draft_rdd_ranef_plot.png Normal file
View File

Binary file not shown.
Side by Side

After

Width:  |  Height:  |  Size: 46 KiB

BIN
R/dwo_working_readability.png Normal file
View File

Binary file not shown.
Side by Side

After

Width:  |  Height:  |  Size: 159 KiB

70
R/gam_plot_documents.R Normal file
View File

@ -0,0 +1,70 @@
#trying to make a time plot showing the over-time shift
library(plyr)
contrib_df <- read_csv('../final_data/deb_contrib_did.csv')
readme_df <- read_csv("../final_data/deb_readme_did.csv")
col_order <- c("upstream_vcs_link", "age_in_days", "first_commit", "first_commit_dt", "event_gap", "event_date", "event_hash", "before_all_ct", "after_all_ct", "before_mrg_ct", "after_mrg_ct", "before_auth_new", "after_auth_new", "before_commit_new", "after_commit_new")
#first contrib
contrib_df <- contrib_df[,col_order]
contrib_df$ct_before_all <- str_split(gsub("[][]","", contrib_df$before_all_ct), ", ")
contrib_df$ct_after_all <- str_split(gsub("[][]","", contrib_df$after_all_ct), ", ")
contrib_df$ct_before_mrg <- str_split(gsub("[][]","", contrib_df$before_mrg_ct), ", ")
contrib_df$ct_after_mrg <- str_split(gsub("[][]","", contrib_df$after_mrg_ct), ", ")
drop <- c("before_all_ct", "before_mrg_ct", "after_all_ct", "after_mrg_ct")
contrib_df = contrib_df[,!(names(contrib_df) %in% drop)]
#then readme
readme_df <- readme_df[,col_order]
readme_df$ct_before_all <- str_split(gsub("[][]","", readme_df$before_all_ct), ", ")
readme_df$ct_after_all <- str_split(gsub("[][]","", readme_df$after_all_ct), ", ")
readme_df$ct_before_mrg <- str_split(gsub("[][]","", readme_df$before_mrg_ct), ", ")
readme_df$ct_after_mrg <- str_split(gsub("[][]","", readme_df$after_mrg_ct), ", ")
drop <- c("before_all_ct", "before_mrg_ct", "after_all_ct", "after_mrg_ct")
readme_df = readme_df[,!(names(readme_df) %in% drop)]
# 2 some expansion needs to happens for each project
expand_timeseries <- function(project_row) {
longer <- project_row |>
pivot_longer(cols = starts_with("ct"),
names_to = "window",
values_to = "count") |>
unnest(count)
longer$observation_type <- gsub("^.*_", "", longer$window)
longer <- ddply(longer, "observation_type", transform, week=seq(from=0, by=1, length.out=length(observation_type)))
longer$count <- as.numeric(longer$count)
#longer <- longer[which(longer$observation_type == "all"),]
return(longer)
}
expanded_contrib_data <- expand_timeseries(contrib_df[1,])
for (i in 2:nrow(contrib_df)){
expanded_contrib_data <- rbind(expanded_contrib_data, expand_timeseries(contrib_df[i,]))
}
expanded_readme_data <- expand_timeseries(readme_df[1,])
for (i in 2:nrow(readme_df)){
expanded_readme_data <- rbind(expanded_readme_data, expand_timeseries(readme_df[i,]))
}
window_num <- 8
windowed_contrib_data <- expanded_contrib_data |>
filter(week >= (27 - window_num) & week <= (27 + window_num)) |>
mutate(D = ifelse(week > 27, 1, 0))
windowed_readme_data <- expanded_readme_data |>
filter(week >= (27 - window_num) & week <= (27 + window_num)) |>
mutate(D = ifelse(week > 27, 1, 0))
windowed_contrib_data$week_offset <- windowed_contrib_data$week - 27
all_actions_contrib_data <- windowed_contrib_data[which(windowed_contrib_data$observation_type == "all"),]
all_actions_contrib_data$document_type <- rep("contributing", length(all_actions_contrib_data$count))
windowed_readme_data$week_offset <- windowed_readme_data$week - 27
all_actions_readme_data <- windowed_readme_data[which(windowed_readme_data$observation_type == "all"),]
all_actions_readme_data$document_type <- rep("readme", length(all_actions_readme_data$count))
all_actions_data <- rbind(all_actions_contrib_data, all_actions_readme_data)
all_actions_data$log1p_count <- log1p(all_actions_data$count)
time_plot <- all_actions_data |>
ggplot(aes(x=week_offset, y=log1p_count, color=factor(document_type))) +
geom_smooth() +
geom_vline(xintercept = 0)+
theme_bw() +
theme(legend.position = "top")
time_plot
#looking at event gap
mean(all_actions_readme_data$event_gap)
sd(all_actions_readme_data$event_gap)
mean(all_actions_contrib_data$event_gap)
sd(all_actions_contrib_data$event_gap)

42
R/model_presentation.R Normal file
View File

@ -0,0 +1,42 @@
library(tidyverse)
library(texreg)
readme_rdd <- readRDS("final_models/0624_readme_all_rdd.rda")
contrib_rdd <- readRDS("final_models/0624_contrib_all_rdd.rda")
summary(readme_rdd)
texreg(list(readme_rdd, contrib_rdd), stars=NULL, digits=3, use.packages=FALSE,
custom.model.names=c( 'README','CONTRIBUTING'),
custom.coef.names=c('(Intercept)', 'Indtroduction', 'Week (Time)', 'Project Age', 'Introduction:Week', 'Event Gap'),
table=FALSE, ci.force = TRUE)
readme_groupings <- read.csv('../final_data/deb_readme_interaction_groupings.csv')
contrib_groupings <- read.csv('../final_data/deb_contrib_interaction_groupings.csv')
subdirColors <-
setNames( c('firebrick1', 'forestgreen', 'cornflowerblue')
, c(0,1,2) )
readme_g <- readme_groupings |>
ggplot(aes(x=rank, y=estimate, col = as.factor(ranef_grouping))) +
geom_linerange(aes(ymin= conf.low, ymax= conf.high)) +
scale_color_manual(values = subdirColors) +
guides(fill="none", color="none")+
theme_bw()
readme_g
contrib_g <- contrib_groupings |>
ggplot(aes(x=rank, y=estimate, col = as.factor(ranef_grouping))) +
geom_linerange(aes(ymin= conf.low, ymax= conf.high)) +
scale_color_manual(values = subdirColors) +
theme_bw() +
theme(legend.position = "top")
contrib_g
library(gridExtra)
grid.arrange(contrib_g, readme_g, nrow = 1)
library(jtools)
plot_summs(readme_rdd, contrib_rdd, plot.distributions = TRUE)
?plot_summs