post-submission cleaning

.gitignore

@@ -0,0 +1,4 @@
+.Rproj.user
+.Rhistory
+.RData
+.Ruserdata

R/R.Rproj

@@ -0,0 +1,13 @@
+Version: 1.0
+
+RestoreWorkspace: Default
+SaveWorkspace: Default
+AlwaysSaveHistory: Default
+
+EnableCodeIndexing: Yes
+UseSpacesForTab: Yes
+NumSpacesForTab: 2
+Encoding: UTF-8
+
+RnwWeave: Sweave
+LaTeX: pdfLaTeX

R/contribRDDAnalysis.R

@@ -45,9 +45,15 @@ mrg_actions_data <- windowed_data[which(windowed_data$observation_type == "mrg")
 all_actions_data$logged_count <- log(all_actions_data$count)
 all_actions_data$log1p_count <- log1p(all_actions_data$count)
 #EDA
+sd(all_actions_data$count)
+grouped_averages <- aggregate(all_actions_data$count, list(all_actions_data$upstream_vcs_link), mean)
+quantile(grouped_averages$x)
+quantile(all_actions_data$before_auth_new)
+quantile(all_actions_data$after_auth_new)
+
 range(all_actions_data$log1p_count) # 0.000000 6.745236
 mean(all_actions_data$log1p_count) # 1.200043
-var(all_actions_data$log1p_count) # 1.753764
+sd(all_actions_data$log1p_count) 
 median(all_actions_data$log1p_count) # 0.6931472
 # now for merge
 mrg_actions_data$logged_count <- log(mrg_actions_data$count)
@@ -58,12 +64,12 @@ library(optimx)
 library(lattice)
 #model
 print("fitting model")
-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 <- 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("0711_contrib_all.rda")
 summary(all_gmodel)
-saveRDS(all_gmodel, "0711_contrib_all_01.rda")
+#saveRDS(all_gmodel, "0711_contrib_all_01.rda")
 
 all_residuals <- residuals(all_gmodel)
 qqnorm(all_residuals)

R/contrib_docChar_outcomes.R

@@ -1,5 +1,7 @@
 #libraries
-library(stringr)
+library(readr)
+library(tidyverse)
+library(plyr)
 contrib_df <- read_csv("../final_data/deb_contrib_did.csv")
 contrib_pop_df <- read_csv("../final_data/deb_contrib_pop_change.csv")
 contrib_readability_df <- read_csv('../text_analysis/dwo_readability_contributing.csv')
@@ -68,10 +70,12 @@ contrib_total_df <- contrib_pop_df |>
 contrib_total_df <- contrib_total_df|>
   join(summed_data, by="upstream_vcs_link")
 #outcome variable that is number of commits by number of new contributors
-contrib_total_df$commit_by_contrib = contrib_total_df$summed_count * contrib_total_df$after_contrib_new
+contrib_total_df$commit_by_contrib = contrib_total_df$summed_count + contrib_total_df$after_contrib_new * 2
 contrib_total_df$logged_outcome = log1p(contrib_total_df$commit_by_contrib)
+contrib_total_df$logged_contribs = log1p(contrib_total_df$after_contrib_new)
+contrib_total_df$logged_commits = log1p(contrib_total_df$summed_count)
 # test regressions
 library(MASS)
-lm1 <- glm.nb(summed_count ~ reading_time + linsear_write_formula + flesch_reading_ease + mcalpine_eflaw + word_count, data = contrib_total_df)
+lm1 <- glm.nb(logged_contribs ~ reading_time + linsear_write_formula + flesch_reading_ease + mcalpine_eflaw + word_count, data = contrib_total_df)
 qqnorm(residuals(lm1))
 summary(lm1)

R/contrib_topic_outcomes.R

@@ -2,9 +2,12 @@
 library(stringr)
 library(plyr)
 contrib_topics_df <- read_csv("../text_analysis/contrib_file_topic_distributions.csv")
+colMeans(subset(contrib_topics_df, select = -filename))
 contrib_df <- read_csv("../final_data/deb_contrib_did.csv")
 contrib_pop_df <- read_csv("../final_data/deb_contrib_pop_change.csv")
 
+median(contrib_df$age_in_days)
+
 #get the contribution count 
 #some preprocessing and expansion 
 col_order <- c("upstream_vcs_link", "age_in_days", "first_commit", "first_commit_dt", "event_gap", "event_date",  "event_hash", "before_all_ct", "after_all_ct", "before_mrg_ct", "after_mrg_ct", "before_auth_new", "after_auth_new", "before_commit_new",  "after_commit_new")
@@ -73,8 +76,19 @@ contrib_total_df <- contrib_total_df|>
 #outcome variable that is number of commits by number of new contributors
 contrib_total_df$commit_by_contrib = contrib_total_df$summed_count * contrib_total_df$after_contrib_new
 contrib_total_df$logged_outcome = log1p(contrib_total_df$commit_by_contrib)
+contrib_total_df$logged_contrib = log1p(contrib_total_df$after_contrib_new)
+contrib_total_df$logged_commits = log1p(contrib_total_df$summed_count)
 #running regressions 
 library(MASS)
-lm1 <- glm.nb(commit_by_contrib ~ t2+  t3, data = contrib_total_df)
+contrib_ <- glm.nb(logged_contrib ~ t0 + t1 + t2 + t3, data = contrib_total_df)
+commits_ <- glm.nb(logged_commits ~ t0 + t1 + t2 + t3, data = contrib_total_df)
 qqnorm(residuals(lm1))
-summary(lm1)
+summary(contrib_)
+summary(commits_)
+texreg(list(contrib_, commits_), stars=NULL, digits=3, use.packages=FALSE, 
+       custom.model.names=c( 'Contributions','Commits'),
+       custom.coef.names=c('(Intercept)', 'Topic 1', 'Topic 2', 'Topic 3'), 
+       table=FALSE, ci.force = TRUE)
+
+saveRDS(commits_, "0731_topic_commitoutcome_contrib.rda")
+saveRDS(contrib_, "0731_topic_contriboutcome_contrib.rda")

R/delta_analysis.R

@@ -1,10 +1,22 @@
 library(tidyverse)
 contrib_df <- read_csv("../final_data/deb_contrib_did.csv")
 readme_df <- read_csv("../final_data/deb_readme_did.csv")
+median(readme_df$age_in_days)
+sd(readme_df$age_in_days)
+
+median(contrib_df$age_in_days)
+sd(contrib_df$age_in_days)
+
 contrib_df <- contrib_df |>
   filter(event_gap >= 0)
 readme_df <- readme_df |>
   filter(event_gap >= 0)
+
+quantile(readme_df$age_in_days)
+quantile(contrib_df$age_in_days)
+
+quantile(readme_df$event_gap)
+quantile(contrib_df$event_gap)
   
 hist(readme_df$event_gap)
 median(readme_df$event_gap)

R/documentReadabilityAnalysis.R

@@ -10,6 +10,8 @@ try(setwd(dirname(rstudioapi::getActiveDocumentContext()$path)))
 readme_df <- read_csv("../text_analysis/dwo_readability_readme.csv")
 contributing_df <- read_csv("../text_analysis/dwo_readability_contributing.csv")
 head(readme_df)
+quantile(readme_df$reading_time)
+quantile(contributing_df$reading_time)
 aggregate(readme_df[, 3:10], list(readme_df$subdir), median)
 aggregate(contributing_df[, 3:10], list(contributing_df$subdir), median)
 #getting basic stats for the readme readability 
@@ -19,45 +21,61 @@ median(readme_df$mcalpine_eflaw)
 median(readme_df$reading_time)
 # establishing the color scheme 
 subdirColors <-
-  setNames( c('firebrick1', 'forestgreen', 'cornflowerblue')
+  setNames( c('#31449c', '#4a7c85', '#c5db68')
             , levels(contributing_df$subdir)  )
+readmeSubdirColors <-
+  setNames( c('#4a7c85', '#c5db68')
+            , levels(readme_df$subdir)  )
 
 #plotting linsear scoring
 readme_linsear_plot <- ggplot(readme_df, aes(x=linsear_write_formula, group=as.factor(subdir))) + 
-  geom_density(aes(color = as.factor(subdir), fill=as.factor(subdir)), alpha=0.2, position="identity") +
+  geom_density(aes(color = as.factor(subdir), fill=as.factor(subdir)), alpha=0.35, position="identity") +
   xlim(-30, 30) +
   theme_bw()
 #plotting readme reading ease
 readme_reading_ease <- ggplot(readme_df, aes(x=flesch_reading_ease, group=as.factor(subdir))) + 
-  geom_density(aes(color = as.factor(subdir), fill=as.factor(subdir)), alpha=0.2, position="identity") +
-  scale_color_manual(values = subdirColors) + 
+  geom_density(aes(fill=as.factor(subdir)), position="fill") +
+  scale_fill_manual(values = readmeSubdirColors) + 
   xlim(-5, 90) +
-  ylab("readme density") +
+  labs(x= "Flesch Reading Ease", y= "README Density")+ 
   guides(fill="none", color="none")+
   theme_bw() 
 readme_reading_ease 
 #plotting readme reading time 
 readme_reading_time_plot <- ggplot(readme_df, aes(x=reading_time, group=as.factor(subdir))) + 
-  geom_density(aes(color = as.factor(subdir), fill=as.factor(subdir)), alpha=0.2, position="identity") +
-  scale_color_manual(values = subdirColors) + 
+  geom_density(aes(fill=as.factor(subdir)), position="fill") +
+  scale_fill_manual(values = readmeSubdirColors) + 
   xlim(-5, 90) +
-  ylab("readme density") +
+  labs(x= "Reading Time (s)", y= NULL)+ 
+  guides(fill="none", color="none")+
+  theme_bw() 
+readme_reading_time_plot
+readme_reading_time_no_group <- ggplot(readme_df, aes(x=reading_time)) + 
+  geom_histogram(fill='forestgreen') +
+  xlim(-5, 190) +
+  ylab("Count of README Files") +
+  xlab("Reading Time (s)") +
+  ggtitle("Reading Time for README files from FLOSS Projects (n=2280)")+
   guides(fill="none", color="none")+
   theme_bw() 
 
+readme_reading_time_no_group 
+
 readme_linsear_plot <- ggplot(readme_df, aes(x=linsear_write_formula, group=as.factor(subdir))) + 
-  geom_density(aes(color = as.factor(subdir), fill=as.factor(subdir)), alpha=0.2, position="identity") +
-  scale_color_manual(values = subdirColors) + 
-  xlim(-5, 30) +
-  ylab("readme density") +
+  geom_density(aes(fill=as.factor(subdir)), alpha=0.35, position="identity") +
+  scale_fill_manual(values = readmeSubdirColors) + 
+  xlim(-5, 90) +
+  labs(x= "Linsear Write Score", y= NULL)+ 
   guides(fill="none", color="none")+
   theme_bw() 
+readme_linsear_plot
 readme_mcalpine_eflaw <- ggplot(readme_df, aes(x=mcalpine_eflaw, group=as.factor(subdir))) + 
-  scale_color_manual(values = subdirColors) + 
-  geom_density(aes(color = as.factor(subdir), fill=as.factor(subdir)), alpha=0.2, position="identity") +
-  xlim(-5, 60) +
+  geom_density(aes(fill=as.factor(subdir)), alpha=0.35, position="identity") +
+  scale_fill_manual(values = readmeSubdirColors) + 
+  xlim(-5, 90) +
+  labs(x= "McAlpine EFLAW", y= NULL)+ 
   guides(fill="none", color="none")+
-  theme_bw()
+  theme_bw() 
 #theme(axis.title.y=element_blank())
 #plot of reading_ease 
 #readme_df <- readme_df |>
@@ -72,33 +90,64 @@ median(contributing_df$reading_time)
 median(contributing_df$linsear_write_formula)
 # plotting contributing linsear writing formula
 contributing_linsear_plot <- ggplot(contributing_df, aes(x=linsear_write_formula, group=as.factor(subdir))) + 
-  scale_color_manual(values = subdirColors) + 
-  geom_density(aes(color = as.factor(subdir), fill=as.factor(subdir)), alpha=0.2, position="identity") +
-  xlim(-5, 30) +
-  guides(fill="none", color="none")+
-  theme_bw()
-# plotting contributing reading time 
-contributing_reading_time_plot <- ggplot(contributing_df, aes(x=reading_time, group=as.factor(subdir))) + 
-  scale_color_manual(values = subdirColors) + 
-  geom_density(aes(color = as.factor(subdir), fill=as.factor(subdir)), alpha=0.2, position="identity") +
-  xlim(-5, 90) +
-  ylab("contributing density") +
-  guides(fill="none", color="none")+
-  theme_bw()
-# plotting contributing mcalpine eflaw
-contributing_mcalpine_eflaw <- ggplot(contributing_df, aes(x=mcalpine_eflaw, group=as.factor(subdir))) + 
-  scale_color_manual(values = subdirColors) + 
-  geom_density(aes(color = as.factor(subdir), fill=as.factor(subdir)), alpha=0.2, position="identity") +
-  xlim(-5, 60) +
-  guides(fill="none", color="none")+
-  theme_bw()
-# plotting contributing reading ease
-contributing_reading_ease <- ggplot(contributing_df, aes(x=flesch_reading_ease, group=as.factor(subdir))) + 
-  geom_density(aes(color = as.factor(subdir), fill=as.factor(subdir)), alpha=0.2, position="identity") +
-  scale_color_manual(values = subdirColors) + 
-  ylab("contributing density") +
+  geom_density(aes(fill=as.factor(subdir)), alpha=0.35, position="identity") +
+  scale_fill_manual(values = subdirColors, labels=c('CI < 0', '0 in CI', '0 < CI')) + 
+  labs(x= NULL, y= NULL, fill="RE Grouping")+ 
   xlim(-5, 90) +
   theme_bw() + 
-  theme(legend.position = "top") 
+  guides(fill="none", color="none")
+# plotting contributing reading time 
+contributing_reading_time_plot <- ggplot(contributing_df, aes(x=reading_time, group=as.factor(subdir))) + 
+  scale_fill_manual(values = subdirColors, labels=c('CI < 0', '0 in CI', '0 < CI')) + 
+  geom_density(aes(fill=as.factor(subdir)), position="fill") +
+  xlim(-5, 90) +
+  labs(x= NULL, y= NULL, fill="RE Grouping")+ 
+  theme_bw() +
+  theme(legend.position = "inside", 
+        legend.position.inside = c(.93, .93),
+        legend.justification = c("right", "top"),
+        legend.direction = "horizontal",
+        legend.margin = margin(6, 6, 6, 6))
+contributing_reading_time_plot
+# plotting contributing mcalpine eflaw
+contributing_mcalpine_eflaw <- ggplot(contributing_df, aes(x=mcalpine_eflaw, group=as.factor(subdir))) + 
+  scale_fill_manual(values = subdirColors, labels=c('CI < 0', '0 in CI', '0 < CI')) + 
+  geom_density(aes(fill=as.factor(subdir)), alpha=0.35, position="identity") +
+  xlim(-5, 90) +
+  labs(x= NULL, y= NULL, fill="RE Grouping")+ 
+  theme_bw() +
+  theme(legend.position = "inside", 
+        legend.position.inside = c(.93, .93),
+        legend.justification = c("right", "top"),
+        legend.direction = "vertical",
+        legend.margin = margin(6, 6, 6, 6))
+# plotting contributing reading ease
+contributing_reading_ease <- ggplot(contributing_df, aes(x=flesch_reading_ease, group=as.factor(subdir))) + 
+  geom_density(aes(fill=as.factor(subdir)), position="fill") +
+  scale_fill_manual(values = subdirColors, labels=c('CI < 0', '0 in CI', '0 < CI')) + 
+  labs(x= NULL, y="CONTRIBUTING Density", fill="RE Grouping")+ 
+  xlim(-5, 90) +
+  theme_bw() + 
+  guides(fill="none", color="none")
 contributing_reading_ease 
-grid.arrange(contributing_reading_ease, contributing_reading_time_plot,contributing_linsear_plot, contributing_mcalpine_eflaw, readme_reading_ease, readme_reading_time_plot, readme_linsear_plot, readme_mcalpine_eflaw, nrow = 2)
+grid.arrange(contributing_reading_ease, contributing_reading_time_plot, readme_reading_ease, readme_reading_time_plot, nrow = 2)
+
+doctypeColors <-
+  setNames( c('#5da2d8', '#c7756a')
+            , c("CONTRIBUTING", "README"))
+readme_df$type = "README"
+contributing_df$type = "CONTRIBUTING"
+all_df = rbind(readme_df, contributing_df)
+length_plot_all <- ggplot(all_df, aes(x=word_count, group=as.factor(type))) + 
+  geom_density(aes(fill = as.factor(type)), color = NA, alpha=0.6, position="identity")+
+  scale_fill_manual(values = doctypeColors) + 
+  xlim(-10, 500) +
+  labs(
+    x = "Word Count",
+    y = "Density Across Documents",
+    fill="Document Type"
+  ) +
+  theme_bw() + 
+  theme(legend.position = "top") 
+length_plot_all
+#grid.arrange(contributing_reading_time_plot, readme_reading_time_plot, nrow = 1)

R/gam_plot_documents.R

@@ -50,21 +50,52 @@ windowed_readme_data <- expanded_readme_data |>
 
 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$count))
+all_actions_contrib_data$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_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)
+library(scales)
+expm1_trans <- trans_new(
+  name = 'expm1',
+  transform = function(x) expm1(x),
+  inverse = function(x) log1p(x)
+)
+
+doctypeColors <-
+  setNames( c('#5da2d8', '#c7756a')
+            , c("CONTRIBUTING", "README"))
+
 time_plot <- all_actions_data |>
-  ggplot(aes(x=week_offset, y=log1p_count, color=factor(document_type))) + 
+  ggplot(aes(x=week_offset, y=log1p_count, color=factor(document_type))) +
+  labs(x="Weekly Offset", y="Commit Count", color="Document Type") +
+  scale_color_manual(values = doctypeColors) + 
   geom_smooth() +
   geom_vline(xintercept = 0)+
   theme_bw() +
   theme(legend.position = "top")
 time_plot
+#code to change the axes
+
+#scale_y_continuous(breaks = c(0, 0.5, 1.0, 1.5), 
+#                   labels = round(c(expm1(0), expm1(0.5), expm1(1.0), expm1(1.5)), 1)) +
+
 #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)
+
+#all_actions_contrib_data$log1p_count <- log1p(all_actions_contrib_data$count)
+#contrib_time_plot <- all_actions_contrib_data |>
+#  ggplot(aes(x=week_offset, y=log1p_count)) + 
+#  geom_smooth(color=forestgreen) +
+#  geom_vline(xintercept = 0)+
+#  annotate("text", x=3, y=1, label="CONTRIBUTING.md Publication", angle=0)+
+#  theme_bw() +
+#  ylab("Log Transformed Count of Contributions") +
+#  xlab("Offset Weeks") +
+#  theme(legend.position = "top")
+#contrib_time_plot
+

R/model_presentation.R

@@ -11,10 +11,10 @@ texreg(list(readme_rdd, contrib_rdd), stars=NULL, digits=3, use.packages=FALSE,
        table=FALSE, ci.force = TRUE)
 
 readme_groupings <- read.csv('../final_data/deb_readme_interaction_groupings.csv')
-contrib_groupings <- read.csv('../final_data/0711_contrib_inter_groupings.csv')
+contrib_groupings <- read.csv('../final_data/deb_contrib_interaction_groupings.csv')
 
 subdirColors <-
-  setNames( c('firebrick1', 'forestgreen', 'cornflowerblue')
+  setNames( c('#31449c', '#4a7c85', '#c5db68')
             , c(0,1,2)  )
 
 readme_g <- readme_groupings |> 
@@ -22,16 +22,18 @@ readme_g <- readme_groupings |>
   geom_linerange(aes(ymin= conf.low, ymax= conf.high)) +
   scale_color_manual(values = subdirColors) + 
   guides(fill="none", color="none")+
-  theme_bw()
+  theme_bw() + 
+  labs(x="RE Coefficient Rank", y="RE Coefficient Estimate", color="Estimate Grouping")
 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) + 
+  scale_color_manual(values = subdirColors, labels=c('CI < 0', '0 in CI', '0 < CI')) + 
   theme_bw() +
-  theme(legend.position = "top") 
+  theme(legend.position = "top")  +
+  labs(x="RE Coefficient Rank", y="RE Coefficient Estimate", color="Estimate Grouping")
 contrib_g
 
 library(gridExtra)

R/readmeRDDAnalysis.R

@@ -47,6 +47,10 @@ mrg_actions_data <- windowed_data[which(windowed_data$observation_type == "mrg")
 all_actions_data$logged_count <- log(all_actions_data$count)
 all_actions_data$log1p_count <- log1p(all_actions_data$count)
 range(all_actions_data$log1p_count)
+
+grouped_averages <- aggregate(all_actions_data$count, list(all_actions_data$upstream_vcs_link), mean)
+quantile(grouped_averages$x)
+
 # 3 rdd in lmer analysis
 # rdd: https://rpubs.com/phle/r_tutorial_regression_discontinuity_design
 # lmer: https://www.youtube.com/watch?v=LzAwEKrn2Mc
@@ -57,14 +61,16 @@ library(lattice)
 #some more EDA to go between Poisson and neg binomial
 var(all_actions_data$log1p_count) # 1.125429
 mean (all_actions_data$log1p_count) # 0.6426873
+sd(all_actions_data$log1p_count)
 median(all_actions_data$log1p_count) #0
 var(all_actions_data$count) # 268.4449
 mean (all_actions_data$count) # 3.757298
+sd (all_actions_data$count)
 median(all_actions_data$count) # 0
 print("fitting model")
 #all_log1p_gmodel <- glmer.nb(log1p_count ~ D * week_offset+ scaled_project_age + scaled_event_gap + (D * week_offset | upstream_vcs_link), data=all_actions_data, nAGQ=1,  control=glmerControl(optimizer="bobyqa",
 #                         optCtrl=list(maxfun=1e5)))
-all_log1p_gmodel <- readRDS("final_models/0624_readme_all_rdd.rda")
+#all_log1p_gmodel <- readRDS("final_models/0624_readme_all_rdd.rda")
 summary(all_log1p_gmodel)
 print("model fit")
 #I grouped the ranef D effects on 0624

R/readme_docChar_outcomes.R

@@ -1,5 +1,6 @@
 #libraries
 library(stringr)
+library(tidyverse)
 readme_df <- read_csv("../final_data/deb_readme_did.csv")
 readme_pop_df <- read_csv("../final_data/deb_readme_pop_change.csv")
 readme_readability_df <- read_csv('../text_analysis/dwo_readability_readme.csv')
@@ -94,9 +95,11 @@ readme_total_df <- readme_total_df|>
 readme_total_df$commit_by_contrib = NA
 readme_total_df$commit_by_contrib = readme_total_df$summed_count * (readme_total_df$after_contrib_new + 1)
 readme_total_df$logged_outcome = log1p(readme_total_df$commit_by_contrib)
+readme_total_df$logged_contribs = log1p(readme_total_df$after_contrib_new)
+readme_total_df$logged_commits= log1p(readme_total_df$summed_count)
 # test regressions
 library(MASS)
-lm1 <- glm.nb(logged_outcome~ reading_time + linsear_write_formula + flesch_reading_ease + mcalpine_eflaw + word_count, data = readme_total_df)
+lm1 <- glm.nb(logged_commits~ word_count, data = readme_total_df)
 qqnorm(residuals(lm1))
 summary(lm1)
 

R/readme_topic_outcomes.R

@@ -1,7 +1,8 @@
 
 library(stringr)
-library(plyr)
+library(tidyverse)
 readme_topics_df <- read_csv("../text_analysis/readme_file_topic_distributions.csv")
+colMeans(subset(readme_topics_df, select = -filename))
 readme_df <- read_csv("../final_data/deb_readme_did.csv")
 readme_pop_df <- read_csv("../final_data/deb_readme_pop_change.csv")
 
@@ -72,9 +73,21 @@ readme_total_df <- readme_total_df|>
 
 #outcome variable that is number of commits by number of new readmeutors
 readme_total_df$commit_by_contrib = readme_total_df$summed_count *readme_total_df$after_contrib_new
-readme_total_df$logged_outcome = log1p(readme_total_df$commit_by_readme)
+readme_total_df$logged_outcome = log1p(readme_total_df$commit_by_contrib)
+readme_total_df$logged_contrib = log1p(readme_total_df$after_contrib_new)
+readme_total_df$logged_commits = log1p(readme_total_df$summed_count)
 #running regressions 
 library(MASS)
-lm1 <- glm.nb(commit_by_contrib ~ t0+t1+t2+t7+t3 +t4 + t5, data = readme_total_df)
+lm1 <- glm.nb(logged_contrib~ t0+t1+t2+t7+t3 +t6 + t5, data = readme_total_df)
 qqnorm(residuals(lm1))
 summary(lm1)
+#saveRDS(lm1, "0725_topic_contriboutcome_readme.rda")
+contrib_ <- glm.nb(logged_contrib~ t0+t1+t2+t3+ t5 +t6 +t7, data = readme_total_df)
+commit_ <- glm.nb(logged_commits~ t0+t1+t2+t3+ t5 +t6 +t7, data = readme_total_df)
+
+library(texreg)
+
+texreg(list(contrib_, commit_), stars=NULL, digits=3, use.packages=FALSE, 
+       custom.model.names=c( 'Contributions','Commits'),
+       custom.coef.names=c('(Intercept)', 'Topic 1', 'Topic 2', 'Topic 3', 'Topic 4', 'Topic 6', 'Topic 7', 'Topic 8'), 
+       table=FALSE, ci.force = TRUE)

R/topic_distributions.R

@@ -0,0 +1,42 @@
+library(stringr)
+library(tidyverse)
+readme_topics_df <- read_csv("../text_analysis/readme_file_topic_distributions.csv")
+contrib_topics_df <- read_csv("../text_analysis/contrib_file_topic_distributions.csv")
+quantile(contrib_topics_df$t0)
+confint(contrib_topics_df$t0)
+sd(contrib_topics_df$t0)
+quantile(contrib_topics_df$t1)
+mean(contrib_topics_df$t1)
+sd(contrib_topics_df$t1)
+quantile(contrib_topics_df$t2)
+mean(contrib_topics_df$t2)
+sd(contrib_topics_df$t2)
+quantile(contrib_topics_df$t3)
+mean(contrib_topics_df$t3)
+sd(contrib_topics_df$t3)
+quantile(readme_topics_df$t0)
+mean(readme_topics_df$t0)
+sd(readme_topics_df$t0)
+quantile(readme_topics_df$t1)
+mean(readme_topics_df$t1)
+sd(readme_topics_df$t1)
+quantile(readme_topics_df$t2)
+mean(readme_topics_df$t2)
+sd(readme_topics_df$t2)
+quantile(readme_topics_df$t3)
+mean(readme_topics_df$t3)
+sd(readme_topics_df$t3)
+quantile(readme_topics_df$t4)
+mean(readme_topics_df$t4)
+sd(readme_topics_df$t4)
+quantile(readme_topics_df$t5)
+mean(readme_topics_df$t5)
+sd(readme_topics_df$t5)
+quantile(readme_topics_df$t6)
+mean(readme_topics_df$t6)
+sd(readme_topics_df$t6)
+quantile(readme_topics_df$t7)
+mean(readme_topics_df$t7)
+sd(readme_topics_df$t7)
+
+

main.py

@@ -103,17 +103,3 @@ if __name__ == "__main__":
 
 
 
-'''
-if largest_object[repo_path]["gha_obj"]['milestone_count'] == 0:
-    #del largest_object[repo_path]
-    #return
-    #this is to ensure that projects which don't use milestones are counted
-    largest_object[repo_path]["gha_obj"]['milestone_count'] = 0.1
-largest_object[repo_path]['new_mmt'] = compute_new_mmt(largest_object[repo_path]["perceval_obj"]['contributors'], largest_object[repo_path]["perceval_obj"]['collaborators'])
-#print('New MMT: ' + str(largest_object[repo_path]['new_mmt']))
-largest_object[repo_path]['old_mmt'] = compute_old_mmt(largest_object[repo_path]["perceval_obj"]['contributors'], largest_object[repo_path]["perceval_obj"]['collaborators'])
-#print('Old MMT: ' + str(largest_object[repo_path]['old_mmt']))
-#new mmt formality score
-largest_object[repo_path]['new_formality'] = compute_formality_score(largest_object[repo_path]['new_mmt'], largest_object[repo_path]["gha_obj"]['milestone_count'], largest_object[repo_path]["perceval_obj"]['age_of_project'])
-print(largest_object[repo_path]['new_formality'])
-'''