Combining multiple rows into one row with multiple columns of data R

4
votes

I have a dataset where the same effect was measured in different ways and I want to compare those measurements. My dataset looks like this:

Study     MType     ID     Insect     Mean     Sd     N
Alla      Fecundity  1      Aphid      .62      .7628  11
Alla      RGR        1      Aphid      -32.8    7.76   11
Ando      Survival   2      Bee        2.34     .67    8
Ando      RGR        2      Bee        4.56     .34    10
Ando      Fecundity  2      Bee        5.32     4.3    20

I want to combine the rows by ID number so that the MType, Mean, Sd and N for each row are preserved (although the column names need to change so the columns are distinguishable).

Hopefully, in the end it would look like:

Study ID Insect Fecundity.mean Fecundity.Sd Fecundity.N RGR.mean RGR.Sd...etc

Some difficulties:

  1. There are about 10 different MTypes
  2. each ID number has between 2 and 4 MTypes

I have messed around with reshape and with tidyr and I haven't been able to figure out how to do this with either of them. Please help!

2
rtidyrreshape2
This can definitely be done with reshape2. If you provide a reproducible example of a snippet of your data, for example using dput(head(my_data, 10)), one of us can help you! – qdread

2 Answers

6
votes

You can use reshape via base R. You want to transform your data from long to wide format according to this post: How to reshape data from long to wide format?.

If your data is in a data.frame d:

reshape(d, idvar=c("ID", "Study", "Insect"), timevar = "MType", direction="wide")

Results:

  Study ID Insect Mean.Fecundity Sd.Fecundity N.Fecundity Mean.RGR Sd.RGR N.RGR Mean.Survival Sd.Survival N.Survival
1  Alla  1  Aphid           0.62       0.7628          11   -32.80   7.76    11            NA          NA         NA
3  Ando  2    Bee           5.32       4.3000          20     4.56   0.34    10          2.34        0.67          8
4
votes

Doing this with tidyr is not obvious, because you have to first gather() and then spread(). 

library(tidyverse)
example <- tribble(
~Study, ~MType, ~ID, ~Insect, ~Mean,   ~Sd,   ~N,
"Alla", "Fecundity",  1, "Aphid", .62, .7628,  11,
"Alla", "RGR",   1, "Aphid", -32.8,  7.76, 11,
"Ando", "Survival", 2, "Bee",   2.34,   .67,  8,
"Ando", "RGR",   2, "Bee",   4.56,   .34,  10,
"Ando", "Fecundity",  2, "Bee",   5.32,   4.3,  20)

gather(example, key = "Statistic", value = "value", Mean, Sd, N) %>%
  unite(col="MType.Statistic", MType, Statistic, sep = ".") %>% 
  spread(key = MType.Statistic, value=value)
#> # A tibble: 2 x 12
#>   Study    ID Insect Fecundity.Mean Fecundity.N Fecundity.Sd RGR.Mean
#> * <chr> <dbl> <chr>           <dbl>       <dbl>        <dbl>    <dbl>
#> 1 Alla   1.00 Aphid           0.620        11.0        0.763   -32.8 
#> 2 Ando   2.00 Bee             5.32         20.0        4.30      4.56
#> # ... with 5 more variables: RGR.N <dbl>, RGR.Sd <dbl>,
#> #   Survival.Mean <dbl>, Survival.N <dbl>, Survival.Sd <dbl>