Is there a single function that creates as many subsets as you want?

[shaolinsnake]

Hello,

I have a large dataset that is set up like so:

Site Flow Conc.1 Conc.2 etc.
1
1
1
2
2
3
3
3
4
4
5
6
6


The other columns are also filled with data. I would like to separately analyze each site, but there are a large number of sites. Is there a function that can look at the column "sites" and create a subset of data for each site?

Thank you,

Jake

[Dason]

You can use split to put each of the subsets of the data into their own dataframe that is stored in a list.

Here's an example

# Look at mtcars, notice there are only a few values for carb
mtcars
# Split the data based on the value of carb
split(mtcars, mtcars$carb)

[shaolinsnake]

Thank you, Dason.

I have a followup question: Is there a function or series of functions that can automatically - for each site that I've just split from the main dataset - run each column against every other column?

I've done the ggplot scatterplotMatrix on my entire dataset, but I want to do it for each individual site. Does this make sense? I'm very new to R and still am not familiar with many of the basic functions.

[Dason]

lapply will apply a function to each element of a list. So you could use lapply on the output from split. You would need to write a function that would take as input a subset of interest and produce the plot you want.

dat <- split(mtcars, mtcars$carb)

# Applies "summary" to each of the subsets
lapply(dat, summary)