将行转换为列，然后使用 R 堆叠成单列

Question

I have a dataframe of data which I need to convert to use for a panel study. I want to convert my rows into columns, then stack the columns into a single column. My current data:

STATE    1970   1971   1972   1973...2018
State-A  X      X      X      X       X
State-B  X      X      X      X       X
State-C  X      X      X      X       X

What I need:

Year   State  Data
1970    A      X
1971    A      X
...
2018    A      X
1970    B      X
1971    B      X

etc...

I have tried the stack, and melt, and reshape functions to no avail. I also tried to list my data then flatten it, which was a total disaster. Any help is much appreciated!

Answer 1

We can use tidyverse

library(dplyr)
library(tidyr)
df1 %>%
     pivot_longer(cols = -Year, values_to = 'Data') %>% 
    c
# A tibble: 12 x 3
#   Year  State Data 
#   <chr> <chr> <chr>
# 1 1970  A     X    
# 2 1971  A     X    
# 3 1972  A     X    
# 4 1973  A     X    
# 5 1970  B     X    
# 6 1971  B     X    
# 7 1972  B     X    
# 8 1973  B     X    
# 9 1970  C     X    
#10 1971  C     X    
#11 1972  C     X    
#12 1973  C     X 

Update

With the updated example, change is

df1 %>%
   pivot_longer(cols = -STATE, names_to = 'Year', values_to = 'Data')

If package version of tidyr is old, use gather

df1 %>%
    gather(name, Data, -Year) %>%
    separate(Year, into = c('other', 'State')) %>%
    select(Year = name, State, Data)

---

Or with melt

library(data.table)
melt(setDT(df1), id.var = 'Year', value.name = 'Data')[, 
    .(State = sub('.*-', '', Year), Year = variable, Data)]

data

df1 <- structure(list(Year = c("State-A", "State-B", "State-C"), `1970` = c("X", 
"X", "X"), `1971` = c("X", "X", "X"), `1972` = c("X", "X", "X"
), `1973` = c("X", "X", "X")), class = "data.frame", row.names = c(NA, 
-3L))