Convoy analysis

d <- tweets_c %>%
  filter(reply_count==children) %>%
  enrich_with_analytical_categories() %>%
  inner_join(users_c,by=c("author_id"="user_id")) %>%
  mutate(followers_count_bin=floor(log2(followers_count))) %>%
  count(followers_count_bin,tweet_type) %>%
  group_by(followers_count_bin) %>% 
  mutate(tweet_type_proportion=n/sum(n)) %>%
  ungroup() %>%
  collect()

Warning: Missing values are always removed in SQL.
Use `SUM(x, na.rm = TRUE)` to silence this warning
This warning is displayed only once per session.

d %>%
  ggplot(aes(x=followers_count_bin,y=tweet_type_proportion,color=tweet_type)) +
  geom_step() +
  scale_x_continuous(breaks=seq(0,30,by=4),labels=~scales::comma(2^.)) +
  scale_y_continuous(labels=scales::percent_format(accuracy=1)) +
  theme_hsci_discrete() +
  xlab("Followers") +
  ylab("Proportion of tweets")

Warning: Removed 7 row(s) containing missing values (geom_path).

d <- tweets_c %>%
  filter(reply_count==children) %>%
  enrich_with_analytical_categories() %>%
  mutate(like_count_bin=floor(log2(like_count))) %>%
  count(like_count_bin,tweet_type) %>%
  group_by(like_count_bin) %>% 
  mutate(tweet_type_proportion=n/sum(n)) %>%
  ungroup() %>%
  collect()
d %>%
  ggplot(aes(x=like_count_bin,y=tweet_type_proportion,color=tweet_type)) +
  geom_step() +
  scale_x_continuous(breaks=seq(0,30,by=4),labels=~scales::comma(2^.)) +
  scale_y_continuous(labels=scales::percent_format(accuracy=1)) +
  theme_hsci_discrete() +
  xlab("Likes") +
  ylab("Proportion of tweets")

Warning: Removed 7 row(s) containing missing values (geom_path).

d <- tweets_c %>%
  filter(reply_count==children) %>%
  enrich_with_analytical_categories() %>%
  count(hashtags,tweet_type) %>%
  group_by(hashtags) %>% 
  mutate(tweet_type_proportion=n/sum(n)) %>%
  ungroup() %>%
  collect()
library(ggbreak)
d %>%
  ggplot(aes(x=hashtags,y=tweet_type_proportion,color=tweet_type)) +
  geom_step() +
  scale_y_continuous(labels=scales::percent_format(accuracy=1)) +
  theme_hsci_discrete() +
  xlab("Number of hashtags") +
  ylab("Proportion of tweets")

Warning: Removed 7 row(s) containing missing values (geom_path).

d <- tweets_c %>%
  filter(reply_count==children) %>%
  enrich_with_analytical_categories() %>%
  count(mentions,tweet_type) %>%
  group_by(mentions) %>% 
  mutate(tweet_type_proportion=n/sum(n)) %>%
  ungroup() %>%
  collect()
d %>%
  ggplot(aes(x=mentions,y=tweet_type_proportion,color=tweet_type)) +
  geom_step() +
  scale_y_continuous(labels=scales::percent_format(accuracy=1)) +
  theme_hsci_discrete() +
  xlab("Number of mentions") +
  ylab("Proportion of tweets")

Warning: Removed 7 row(s) containing missing values (geom_path).