updating with revised plots and data

analysis_data/scratch.R

@@ -2,24 +2,76 @@ library(tidyverse)
 main_csv <-"~/analysis_data/120725_unified.csv"
 main_df <- read.csv(main_csv, header = TRUE) 
 
-author_closer <- main_df |>
-  filter(comment_type == "task_description") |>
-  filter(author_closer == "True") 
-table(author_closer$isAuthorWMF)
+dsl_csv <-"~/dsl/120725_DSL_frame.csv"
+dsl_df <- read.csv(dsl_csv, header = TRUE) 
+
+needs_triage <- dsl_df |>
+  filter(week_index >= -4) |>
+  filter(priority=="Needs Triage") |>
+  mutate(ttr_weeks = TTR/168) |>
+  group_by(source, isAuthorWMF)|>
+  summarize(
+    mean_ttr_weeks = mean(ttr_weeks),
+    sd_ttr_weeks = sd(ttr_weeks)
+  )
+  
+changes<- dsl_df |>
+  filter(priority == "Needs Triage" | 
+           priority == "Unbreak Now!" |
+           priority == "High") |>
+  mutate(
+    period = case_when(
+      week_index >= -4 & week_index <= 4   ~ "8 weeks after announcement",
+      week_index >= -13 & week_index <= -5 ~ "8 weeks before deployment announcement",
+      TRUE ~ NA
+    )
+  ) %>%
+  filter(!is.na(period)) |>
+  mutate(ttr_weeks = TTR/168) |>
+  group_by(source, period, priority) %>%
+  summarise(
+    count = n(),
+    mean_ttr_weeks = mean(ttr_weeks, na.rm = TRUE),
+    sd_ttr_weeks = sd(ttr_weeks, na.rm = TRUE),
+  )
 
 
-new_authors_summary <- main_df |>
+# new contributors 
+first_task <- main_df |>
   filter(comment_type == "task_description") |>
   group_by(source, AuthorPHID) |>
   summarise(
     task_count = n(),
-    first_task = min(week_index)
-  ) |>
-  group_by(first_task, source) |>
-  summarise(
-    new_authors_count = n()
+    first_task_week = min(week_index)
   ) 
-ggplot(new_authors_summary, aes(x = first_task, y = new_authors_count)) +
+
+tasks_flagged <- main_df %>%
+  filter(comment_type == "task_description") |>
+  left_join(first_task, by = c("source", "AuthorPHID")) %>%
+  mutate(is_first_time_author = week_index == first_task_week)
+
+summary_df <- tasks_flagged %>%
+  group_by(week_index, source) %>%
+  summarize(
+    total_tasks = n(),
+    first_time_tasks = sum(is_first_time_author),
+    proportion_first_time = first_time_tasks / total_tasks
+  ) %>%
+  ungroup()
+
+period_counts <- summary_df %>%
+  mutate(
+    period = case_when(
+      week_index >= -4 & week_index <= 4   ~ "recent",
+      week_index >= -13 & week_index <= -5 ~ "prior",
+      TRUE ~ NA_character_
+    )
+  ) %>%
+  filter(!is.na(period)) %>%
+  group_by(source, period) %>%
+  summarize(period_first_time_tasks = sum(first_time_tasks), .groups = "drop")
+
+ggplot(summary_df, aes(x = week_index, y = first_time_tasks)) +
   facet_grid(source ~ ., 
              scales = "free_y",
              labeller = labeller(source = c("c1" = "VisualEditor", 

dsl/dsl.R

@@ -80,9 +80,9 @@ dev_model <- dsl(
   sample_split = 20,
   data=dsl_df
 )
-summary(dev_model)
-saveRDS(dev_model, "120725_logit_dsl.RDS")
-
+#summary(dev_model)
+#saveRDS(dev_model, "120725_logit_dsl.RDS")
+dev_model <- readRDS("dsl/120725_logit_dsl.RDS")
 library(broom)
 library(dplyr)
 tidy.dsl <- function(x, conf.int = FALSE, conf.level = 0.95, exponentiate = FALSE, ...) {
@@ -138,7 +138,7 @@ coef_df <- coef_df |>
       "WMF-affiliate Author:HTTP-deprecation"
     )))
   )
-ggplot(coef_df, aes(x = estimate, y = term)) +
+dsl_coefs <- ggplot(coef_df, aes(x = estimate, y = term)) +
   geom_point(size = 1) +
   geom_errorbar(aes(xmin = estimate - 1.96*std.error, xmax = estimate + 1.96 *std.error), height = 0.2) +
   geom_vline(xintercept = 0, linetype = "dashed", color = "red") +
@@ -146,3 +146,10 @@ ggplot(coef_df, aes(x = estimate, y = term)) +
        x = "Coefficient Estimate",
        y = "Variable") +
   theme_minimal()
+ggsave(
+  filename = "120825_dsl_coefs.png",
+  plot = dsl_coefs,
+  width = 8,    # inches
+  height = 6,   # inches
+  dpi = 600     # high resolution
+)

dsl/final_bivariate.R

@@ -93,15 +93,14 @@ c1_ttr_plot <- dsl_df |>
       size = 4) + 
     labs(x = "Weeks from Release", 
          y = "Time to Resolution (weeks)", 
-         fill = "Priority Tag",
-         title = "VisualEditor Time to Resolution by Triage Priority") +
+         fill = "Priority Tag") +
   theme(legend.position = "top")
 c1_ttr_plot
 ggsave(
   filename = "120825_c1_ttr.png",
   plot = c1_ttr_plot,
   width = 12,    # inches
-  height = 4,   # inches
+  height = 6,   # inches
   dpi = 600     # high resolution
 )
 
@@ -204,7 +203,6 @@ tasks_created <- ggplot(
   labs(
     x = "Weeks from Feature Deployment", 
     y = "Count of Tasks Created", 
-    title = "Phabricator Tasks Created by Week and Author Affiliation",
     fill = "Task Author Affiliated with WMF?"
   ) + 
   theme(legend.position = "top")
@@ -213,7 +211,7 @@ ggsave(
   filename = "120825_tasks_created.png",
   plot = tasks_created,
   width = 12,    # inches
-  height = 4,   # inches
+  height = 6,   # inches
   dpi = 600     # high resolution
 )
 

dsl/rq1_plots.R

@@ -88,8 +88,7 @@ task_status_plot <- declined_summary|>
                                             "c3" = "HTTP-deprecation"))) + 
   geom_col(position = position_dodge(width = 0.9), width = 0.8) +
   scale_fill_viridis_d(option='magma') + 
-  labs(title = "Task Status (as of February 28, 2025) by Week",
-       x = "Weeks from feature deployment",
+  labs(x = "Weeks from feature deployment",
        y = "% of items in status",
        fill = "Task Status") +
   geom_vline(xintercept = 0, linetype = "dashed", color = "black", linewidth = 0.5) +
@@ -103,7 +102,7 @@ ggsave(
   filename = "120825_tasks_status.png",
   plot = task_status_plot,
   width = 12,    # inches
-  height = 4,   # inches
+  height = 6,   # inches
   dpi = 600     # high resolution
 )