The assignment was to take a set of data and perform operations on it as a data frame, per an example document.

The variables with the initial data were:

Name <- c("Jeb", "Donald", "Ted", "Marco", "Carly", "Hillary", "Berine")
ABC <- c(4, 62, 51, 21, 2, 14, 15)
CBS <- c(12, 75, 43, 19, 1, 21, 19)

I tried turning that into a matrix…

candidates.m <- cbind(Name, ABC, CBS)

…and got a matrix full of strings.

     Name      ABC  CBS
[1,] "Jeb"     "4"  "12"
[2,] "Donald"  "62" "75"
[3,] "Ted"     "51" "43"
[4,] "Marco"   "21" "19"
[5,] "Carly"   "2"  "1"
[6,] "Hillary" "14" "21"
[7,] "Berine"  "15" "19"

So I made a data frame from the data.

candidates.df <- data.frame(Name, ABC, CBS)

The resulting data frame:

     Name ABC CBS
1     Jeb   4  12
2  Donald  62  75
3     Ted  51  43
4   Marco  21  19
5   Carly   2   1
6 Hillary  14  21
7  Berine  15  19

I tried to run the mean command on the data frame, but it didn’t work out in RStudio the way it did in the example. The example showed an error for the text column and then means for the numeric columns, but all I got was the error.

> mean(candidates.df)
[1] NA
Warning message:
In mean.default(candidates.df) :
  argument is not numeric or logical: returning NA

I got the same error when specifying the two numeric columns.

mean(candidates.df[,2:3])

But I could get the mean of a specific column.

> mean(candidates.df[,2])
[1] 24.14286

To get the means of the columns I had to use colMeans on the numeric columns specifically – using colMeans on the full dataframe resulted in an error that “‘x’ must be numeric”.

colMeans(candidates.df[2:3])

Result:

     ABC      CBS
24.14286 27.14286

I could also use rowMeans to get means by row.

> rowMeans(candidates.df[2:3])
[1]  8.0 68.5 47.0 20.0  1.5 17.5 17.0

Using as.matrix(candidates.df) gave the same matrix as I listed at the top of this post.

Data frames are interesting as a means of storing mixed data in a matrix-like structure. It has its caveats when the columns have different modes, but those can be worked around with the right code.

In module 2, the class was asked to test and evaluate R code. We were given two commands:

assignment2 <- c(16, 18, 14, 22, 27, 17, 19, 17, 17, 22, 20, 22)
myMean <- function(assignment2) { return(sum(assignment)/length(someData)) }

The first line creates a dataset labeled assignment2, and the second line creates a function named myMean. Judging by the code, myMean is meant to calculate the mean of the dataset sent to it as an argument.

Unfortunately, the code doesn’t work as written, giving an object not found error. The problem lies in the variables used in the body of the function – neither assignment nor someData have been given values. If the goal is to find the mean of the dataset passed to the function as an argument, then the names of the variables in the function body should match that of the function’s argument.

The function works, then, if it’s rewritten as:

myMean <- function(assignment2) { return(sum(assignment2)/length(assignment2)) }

For clarity it might make sense to change the argument of the function to assignment instead of assignment2. That way there wouldn’t be a potential for confusion between the global variable and the internal variable processed by the function. Because of the scope difference, however, the identical names don’t create a problem for the result – passing a different dataset to the function causes it to return the mean of that dataset, not the dataset contained in the global assignment2 variable.

This category of posts will relate to my coursework in LIS 4930, Introduction to R Programming.

Project code will be accessible in the GitHub repository at <https://github.com/jered0/lis4930-rpackage>.