consolidate columns that may or may not have overlapping data, using R

Question

Let's say you have a data set that looks like this:

          Vietnam    Gulf War     Iraq War
veteran1    1           0            0
veteran2    0           1            0
veteran3    0           0            1
veteran4    0           1            1  # <---- Note this row

You want to consolidate these columns without affecting other columns in the dataframe like so:

          Service  
veteran1    1                  
veteran2    2                     
veteran3    3                            
veteran4    2  # <---- Note this row

Where

• 1 = Vietnam, 2 = Gulf War, 3 = Iraq War
• If a veteran has served 2 or more it should pick only one (as is the case with veteran4 where it picked their left-most column)
• there are many other columns in the dataframe, and they shouldn't be affected by any of this

Questions:

How would you do this in R?

(Note: if it's easier to do in some other free open source program, please feel free to share which program and how you would do it. This is a massive dataset: 3 million rows, the American Community Survey.)

Answer 1

查看您的数据，似乎是一个简单的问题：

如果越南＆gt; 0，然后使用1，否则如果海湾战争＆gt; 0然后2，否则如果伊拉克＆gt; 0然后是3，否则为0

vietnam = c(1, 0, 0,0) 
gulfwar = c(0,1,0,1)
iraq = c(0,0,1,1)
df = data.frame(vietnam, gulfwar, iraq) 
df$service <- ifelse(df$vietnam > 0,1,ifelse(df$gulfwar>0,2,ifelse(df$iraq>0,3,0)))
df

结果：

       vietnam gulfwar iraq service
  1       1       0    0       1
  2       0       1    0       2
  3       0       0    1       3
  4       0       1    1       2

Answer 2

可能有点复杂（关于其他解决方案），但这是使用df$service <- apply(df, 1, function(x) which(x == 1)[1] ) df vietnam gulfwar iraq service 1 1 0 0 1 2 0 1 0 2 3 0 0 1 3 4 0 1 1 2 的一种方法：

X