Typical Problem Steps (on a Mac)

Return to Topics page

This page, written first in August 2016, is meant to serve as a model for how we will be doing most of our work in this class. Note that some of the screen images have been altered to highlight certain features.

In general, we will
  1. be given some data in a table
  2. need to generate that data in a RStudio session
  3. be give some questions about the data or requests to produce some graphs of the data
  4. need to find a way to have R produce answers to those questions or to produce the requested graphs.
In the Fall 2016 term, students will have been given a USB drive that starts out with copies of my special files for students to use in the course. Furthermore, we will do all of our work on those USB drives. It will be most convenient if, for each problem that we want to do we create a separate folder on the UBS drive to hold all of our work for that problem. This page will step through doing this on a Mac.

We are asked to find the mean, median and mode of the data.

We plug in our USB drive. After a moment or two we will find the icon on the desktop. We double click on that to open the Finder window shwn in Figure 0.

Figure 0

The files on the USB drive hold various function that have been given to you.

Now, to do this demonstration, we want a new folder. To create the new folder we click on the File menu item in the Finder menu as shown in Figure 1.

Figure 1

This creates the new folder shown in Figure 2.

Figure 2

The proposed title of the new folder, namely, untitled folder, should be replaced by something more meaningful. We will use demo_one which we just type into the folder name space as shown in Figure 3.

Figure 3

Our next step is to create a file in that new folder. The file should have an .R extension. Also, we want the file to contain a valid R statement. Rather that go through a lot of work to do this, we have a model of such a file on the USB drive. We can just copy that file to our new directory.

Figure 4 shows where we right click on the name of the file, model.R, to get the options shown in the window on the lower left. Click on the Copy option.

Figure 4

Now that we have a copy of model.R we want to place that copy into the demo_one folder. Therefore, still in Figure 4, we locate the demo_one folder icon, , and double click on it to open the screen shown in Figure 5.

Figure 5

We just created this folder so there is nothing in it. Point to the empty part of the screen and right click there to open the window shown in Figure 6.

Figure 6

In the window shown in Figure 6 click on the Paste option. That places the copy of the file into the folder.

Figure 7

Now, with this file, model.R, in our directory we can double click on it to open a new RStudio session shown in Figure 8.

Figure 8

As we see in Figure 8, the editor window is open and it shows the contents of our model.R file. There are no variables defined in the Environment pane. Only the splash screen is in the Console pane.

We will change the model.R edit screen to hold the following lines
#This is a demonstration
#Load the required functions
source("../gnrnd4.R")
source("../mode.R")
#generate the data
gnrnd4(1343409604,4000538)
#just for fun we will display all the values
L1
##do the required computations
mean(L1)
median(L1)
Mode(L1)
Things to note:
  1. The use of upper and lower case letters is important.
  2. Once R finds a # character R considers that character and anything after it on that line to be a comment. We should use comments to explain what we are doing.
  3. You can just type the lines into the editor pane or you could copy the lines from this page and just paste them into the editor pane.
  4. The editor is just that. Nothing happens just because the lines ae in the editor pane.
  5. The editor colors different components in different ways. This helps us to identify some mistakes should we make them.
  6. Just becasue the lines are in the editor pane it does not mean that the lines have been saved to the file. That takes a separate action.


Figure 9

In Figure 9 we see that the new lines have been placed into the editor pane. Now would be a good time to save this work to the model.R file. To do this we need only click on the icon.

Once the lines have been saved we can look at the Files tab in the lower right pane of the RStudio window. In Figure 10 we see that the model.R is there, that it has been updated with a new date and time stamp, and that its size is shown as more than just a few characters.

Figure 10

Back in the editor pane, we can highlight our first command, source("../gnrnd4.R") so that we can perform it. This command will look at the directory (folder) above our current directory to find a file called gnrnd4.R and the command will run the commands that are in that file. The gnrnd4.R file contains R commands that I wrote to accomplish certain tasks. We will make repeated use of those commands.

Once the source("../gnrnd4.R") command is highlighted we click on the icon to actually run the command.

Figure 11

Figure 12 shows the result in the Console pane. We see that the command has appeared here and that it has been performed. [The fact that there is no error message means that it performed without an error.]

Figure 12

Figure 13 shows the Environment pane. Here we see that running the commands in the gnrnd4.R file caused two ne variables to be created and given some strange values, and it caused two new functions to be created, one is called gn_get_urand and the other is called gnrnd4.

Figure 13

While we are at it, we might as well perform the second command, source("../mode.R"). Again, in the editor pane we highlight the command, as shown in Figure 14 and then we press the icon.

Figure 14

Figure 15 shows the updated Console pane.

Figure 15

Figure 16 shws the updated Environment pane. There we see that a new function, Mode, has been defined.

Figure 16

We return to the Editor pane and highlight the next command, gnrnd4(1343409604,4000538). That command came from the original problem statement at the start of this page. There we were given a table of data. At the bottom of the table we were told that we could generate the same data by using the gnrnd4 function along with two key values, 1343409604 and 4000538. To do that we create the command gnrnd4(1343409604,4000538) and then we use the icon to run it.

Figure 17

Figure 18 shows the update of the Console pane. Performing this command produced some output and a final "Done" indication.

Figure 18

The real result of running the command is relected in the update of the Environment ane. This is shown in Figure 19. First note that there are some new variables. Second note that the value of the variable gn_urands has changed from what it was back in Figure 16. Third, note that we new variable L1 holds numeric values and it holds 97 such values. Also, the first five of those values are identical to the values that we in our original table of values.

Figure 19

We can get R to display all of the values by using the command L1. In the Editor pane we have highlighted the comment and that command. Then we click on the icon to run those two lines.

Figure 20

In Figure 21 we see that the comment has appeared as is. That is followed by all 97 values in the variable L1. Each line of the listing of values starts with a number in square brackets. That number tells us the position within the variable L1 of the first value displayed in that output line. Thus, 62.8 is in position 1, 48.5 is in position 13, 51.3 is in position 25, and so on.

Figure 21

We can return to the Editor pane and highlight the remaining lines. [Just as a note, mean and median are functions that are part of the R language. Mode, on the other hand, is not a part of the language. I had to write special commands to get R to be able to find the mode of a set of data. We have loaded Mode into our workspace so we can now use it.]

Figure 22

Then click on icon to run those lines.

Figure 23

The output from the mean(L1) and median(L1) commands is clear. The mean value is 54.19072 and the median value is 54.6. The output from the Mode(L1) command tells us that there were two values, 55.3 and 55.7, that each appeared 3 times and that no other value appeared that oftern or more often.

We have finished our assignment. We are ready to close RStudio. We use the q()command and R responds by asking if we should save the workspace. We say y and press the Enter key to end the session.

Figure 24

Back in the Finder window, shown in Figure 25, we note that we still only see the model.R file. The two "hidden" files do not appear.

Figure 25

if we start RStudio from this folder, by double clicking the model.R file, our workspace, inclucing the variables and the functions that we have loaded, will be right back where they were when we finished in Figure 24. This is shown in Figure 26.

Figure 26

If we then take a loser look at the contents of the Files tab in the lower right pane, shown in Figure 27, we actually do get to see the two "hiddlen" files.

Figure 27

having seen that we would again close the RStudio session since we are all done with our work.

Return to Topics page
©Roger M. Palay     Saline, MI 48176     August, 2016