Clean up your data table

[VCF]

It turns out you do have some issues with your data. If you look at the str() report, you'll see something like this:

> str(Pond_Data)
'data.frame':   141 obs. of  11 variables:
 $ Date               : Factor w/ 137 levels " "," 4:15\\",..: 30 31 32 33 34 35 36 37 38 39 ...
 $ Rain.in.           : Factor w/ 49 levels ""," 4:30\\"," Algae",..: NA 21 21 21 21 NA 21 NA NA NA ...
 $ Depth.cm.          : Factor w/ 62 levels ""," Algae"," around pond",..: NA 40 39 48 42 NA 46 NA NA NA ...
 $ South              : Factor w/ 48 levels ""," Some measurements were difficult to read due to ice and snow",..: NA 34 31 33 34 NA 35 NA NA NA ...
 $ North              : Factor w/ 43 levels ""," Clear sky",..: NA 10 42 42 12 NA 4 NA NA NA ...
 $ West               : num  NA 55 54 25 25 NA 19 NA NA NA ...
 $ East               : num  NA 50 56 50 51 NA 65 NA NA NA ...
 $ Area               : Factor w/ 44 levels "","10160","14",..: NA 31 2 43 29 41 36 21 17 18 ...
 $ Temperature..F..Max: Factor w/ 43 levels ""," Warm\\","-1",..: 39 36 32 36 18 35 30 22 22 18 ...
 $ Temperature..F..Min: Factor w/ 51 levels ""," Snow/ice melting\\",..: 7 8 5 4 NA NA NA 15 15 15 ...
 $ Other.Observations : Factor w/ 30 levels ""," Algae"," Algae\\",..: 30 NA NA NA 1 1 1 1 1 1 ...

We'll get to the "Factor" thing in a bit - that's not so much an issue as a fundamental, but confusing, part of R. You now have a "data.frame" (which I'll call "DF") holding your hard-won data. DFs have a few features / rules:

• They have rows and columns. You can have as many as you like - some DFs are huge.
• Every row should have the same number of columns. Every column should have the same number of rows. If not, the data is called "ragged" and read.table will refuse to load the data (so your data isn't ragged)
• Every column must have identical "kinds" of data. There are several important kinds, the ones we will worry about are:
  • integers
  • numeric (aka "real" or "floating point")
  • character (aka "string" - any mix of letters and numbers. R really likes a flavor of character data called "factor".
  • date/time
  • There are others but we'll stick with these for now
• read.table() will try to figure out what each column is. It can "auto-guess" integer, numeric and character/factor. Character is the "Gosh, I don't know, may as well stuff this into character" guess. You have too many character (Factor) columns. You can see that you have some of your comments "sneaking into" the wrong columns. You'll need to clean those up.

Let's start by dealing with the Factor issue. We can talk about what factors are later, but for now just know that you don't want them here. You want to update your read.csv so that strings are not automatically converted to factors. Look at ?read.csv to see how to set stringsAsFactors to FALSE and then push the changes.

[VCF]

Ok, I see one problem. I told you to use \ for escapes, but that seems to have been bad advice. You should remove all slashes (eg Frozen\, Algae) and instead quote the whole value (eg "Frozen, Algae").

Per our discussion, you can also remove the $Area column - you will be using R to calculate that dynamically from your data.

[VCF]

Alright, things are improving with 706e7a4, but you still have some problems. Let's open a new issue to deal with them.