I have the following question that I am trying to solve with R:

"For each year, first calculate the mean observed value for each country (to allow for settings where countries may have more than 1 value per year, note that this is true in this data set). Then rank countries by increasing MMR for each year. Calculate the mean ranking across all years, extract the mean ranking for 10 countries with the lowest ranking across all years, and print the resulting table."

This is what I have so far:

dput(mmr)
tib2 <- mmr %>% 
  group_by(country, year) %>% 
  summarise(mean = mean(mmr)) %>% 
  arrange(mean) %>% 
  group_by(country)
tib2

My output is so close to where I need it to be, I just need to make each country have only one row (that has the mean ranking for each country).

Here is the result:

Output

Thank you!

CodePudding user response：

Not sure without the data, but does this work?

tib2 <- mmr %>% 
  group_by(country, year) %>% 
  summarise(mean1 = mean(mmr)) %>% 
  ungroup() %>% 
  group_by(year) %>% 
  mutate(rank1 = rank(mean1)) %>% 
  ungroup() %>% 
  group_by(country) %>% 
  summarise(rank = mean(rank1))%>% 
  ungroup() %>% 
  arrange(rank) %>% 
  slice_head(n=10)

CodePudding user response：

Just repeat the same analysis, but instead of grouping by (country, year), just group by country:

tib2 <- mmr %>% 
  group_by(country, year) %>% 
  summarise(mean_mmr = mean(mmr)) %>% 
  arrange(mean) %>% 
  group_by(country) %>%
  summarise(mean_mmr = mean(mean_mmr)) %>% 
  arrange(mean_mmr) %>% 
  ungroup() %>%
  slice_min(n=10)
  
tib2