[R] rbind with different columns

Antje niederlein-rstat at yahoo.de
Tue Oct 20 15:57:36 CEST 2009


Hello there,

with the following dummy code I'd like to give you an idea how my data 
looks like:

df1 <- data.frame(a = rnorm(10), b = rnorm(10))
df2 <- data.frame(a = rnorm(10), c = rnorm(10))
df3 <- data.frame(a = rnorm(10), b = rnorm(10), c = rnorm(10))

myList <- list(df1, df2, df3) # (myList is representing the data 
structure I have to handle, it's read from single files with lapply)

In every list entry is a data.frame but the columns are only partially 
the same. If I have exactly the same columns, I could do the following 
command to combine my data:

do.call("rbind", myList)

but of course it does not work with differnt column names. Is there any 
easy way to retrieve a combined table like this:

              a          b          c
1  -0.54586587 -0.3607873         NA
2   1.10876842  1.1439414         NA
3   0.57357988 -1.2117743         NA
4  -1.40975759 -1.2390305         NA
5   0.03371093 -1.8860748         NA
6  -1.27682961  0.9990840         NA
7  -1.78715858  0.8400642         NA
8  -0.22663310  1.5224016         NA
9   0.45703787  0.0599217         NA
10 -1.21984635  1.1991689         NA
11 -2.58848301         NA  0.2394272
12  0.71155177         NA -0.7107332
13 -2.16440676         NA -0.1744845
14  1.33043121         NA  0.5951272
15  1.51034297         NA  0.1956956
16  1.00844947         NA  0.6726101
17  0.78693840         NA  1.2189904
18  0.68622170         NA  1.2230500
19 -1.09376863         NA  0.4267472
20  2.23647873         NA  0.7328574
21 -0.38144792  0.1532647  1.4824618
22  0.27078024 -0.4264737  0.1317450
23  1.10812086  1.2550117  0.1677935
24  0.14881701 -0.2928157 -1.4081529
25 -1.00635045 -0.7885968 -0.3502532
26  0.32024094  0.4681016 -1.5477557
27  0.82974710  0.2345186 -0.6572728
28  0.49608133  1.7463265  0.6493405
29 -0.33022738  1.9510503 -1.7930093
30 -0.62615365  0.7330671 -0.4032405

The only thing I can think about is checking the names of each list 
entry and adding NA-columns before combining them.
Is there any other way to do so?


Antje




More information about the R-help mailing list