University of Minnesota, Twin Cities     School of Statistics     Stat 3011     Rweb     Textbook (Wild and Seber)

Stat 3011 (Geyer) In-Class Examples (Chapter 2)

General Instructions

Dot Plots (Strip Plots)

Exercises for Section 2.3.1 (p. 49), Question 1.

Comments

• The square brackets select a subset of the values of the variable unemploy, those for which the variable group has the value "eec". This is documented on the R on-line help for square brackets and relational operators. To see that these are indeed the data you are to plot, look at the data file unemploy.txt.
• R does not seem to have any way to put the labels on the plot requested by this exercise.

Stem-and-Leaf Plots

Example 2.3.2 (p. 53).

Comments

• The additional argument scale=3 to the stem function is necessary so that R will draw the same stem-and-leaf plot as in the textbook. This is documented on the R on-line help for stem.

  There is no way to tell exactly what any particular value of the scale argument does. You just have to experiment until you get a plot that you like.

Histograms

Figure 2.3.8 (p. 56).

Comments

• The square brackets select a subset of the values of the variable length, those for which the variable gender has the value "female". This is just like the dot plot example. To see that these are indeed the data you are to plot, look at the data file coyote.txt.
• The additional argument right=FALSE to the hist function is necessary so that R will draw the same histogram as in the textbook. This is documented on the R on-line help for hist.

  There is no reason to prefer the histogram you get with right=TRUE over the default behavior right=FALSE. Thus there is no reason to use this argument if you don't want to. We only used it to match the illustration in the textbook.

Numerical Summaries

Section 2.4 (p. 61 ff.).

Comments

• summary calculates the "five number summary" (actually six numbers), described on pp. 61-69 of Wild and Seber. Some of the individual numbers in the "five number summary" are calculated by the following four lines
• mean calculates the mean.
• median calculates the median.
• There is no R function that calculates just quartiles. However, the quantile function calculates arbitrary quantiles (p. 244 ff. in Wild and Seber).
  • quantile(x, 0.25) calculates the lower quartile (also called the 0.25 quantile or the 25th percentile) of the variable x.
  • quantile(x, 0.75) calculates the lower quartile (also called the 0.75 quantile or the 75th percentile) of the variable x.
• sd calculates the standard deviation.
• IQR calculates the interquartile range (IQR).

Data Entry

Exercises for Section 2.4.2 (p. 69), Question 1(a)

Comments

• Note that R disagrees with the answer in the textbook about the upper quartile (technically, R is right and the textbook is wrong).
• The first command x <- c(1, 2, 2, 5, 8, 8, 11, 12, 14) inputs the data for this problem (we just type it in rather than read it from some file on the web). The symbol <- is the R assignment operator. The function c "collects" a bunch of numbers into one data vector.
• The summary command is explained in the numerical summaries example.
• Note that any variable name will do. Call it what you like (any word with any numbers or letters, upper or lower case, starting with a letter), for example fred.

  fred <- c(1, 2, 2, 5, 8, 8, 11, 12, 14)
  summary(fred)
  

  Does exactly the same thing as the example.

  Note that upper and lower case are different, that is, Fred with a capital "F" is a different variable.

Bar Plots (Bar Graphs)

Figure 2.5.1 (p. 56).

Comments

• The first command names(freq) <- nstrata associates the names of the strata with the frequencies. The symbol <- is the R assignment operator.
• The barplot command draws the barplot. This is documented on the R on-line help for barplot.