Computer readable data files come on the CD ROM which comes with Moore and McCabe. A somewhat more complete set of files is available from the text's web site in the form of zip archive files for Windows and stuffit archive files for Macintosh. Unfortunately, none of the files available completely match the most convenient way to read data in MacAnova.

I retrieved the complete set of files from the text's web site and have edited them for use in the class. Each file can be read by readdata() which automatically creates named MacAnova variables from information in the file.

The edited files are self documenting in that I have add the page number where the example, exercise or table appears and given a brief description of the variables. This information is printed when a data set is read by readdaata() in MacAnova. You should always read this information carefully when using a data sets. See below for examples of how they can be read in MacAnova.

You can download the all the data sets in compressed archive files, IPS4Data.sit, for Macintosh and IPS4Data.zip for Windows.

Or if you prefer you can download individual files from this page.

The data files follow a simple naming scheme:

• Names of the form egM_N.txt (eg01_005.txt, for example) contain data associated with Example M.N in Moore and McCabe Chapter M.
• Names of the form exM_n.txt (ex01_021.txt, for example) contain data associated with Exercise M.N at the end of Moore and McCabe Chapter M.
• Names of the form ttM_N.txt (ta01_012.txt, for example) contain data associated with Table M.N in Moore and McCabe Chapter M.

Some data sets appear in more than one file. This can occur when an exercise uses data in a numbered table or different exercises use the same data.

Many but not all of the original data sets include a column of case numbers which serve little or any purpose. In most cases these have been retained in the edited versions, but in a few cases I deleted them. Often they are read as variable id

In other cases, the original files had data for different groups in different columns. I generally reformatted these so that all the values for a variable occur in one column with an additional column indicated the group number. See an example below of how to extract the data for just one group for separate analysis.

Examples of using these data files in MacAnova

Here is what file ex02_121.txt looks like.

# File ex02_121.txt from publisher's web site
# Data for Exercise 2.121, p. 216 of IPS4
# Col. 1: len = length of beam made from wood flakes (inches)
# Col. 2: strength = strength of beam (lbs/in^2)
len strength
  5   446   
  6   371   
  7   334   
  8   296   
  9   249   
 10   254   
 11   244   
 12   246   
 13   239   
 14   234   

Here we use readdata() (on a Macintosh) to read the columns in the data files in to MacAnova vectors len and strength and compute some simple descript statistics and make a stemplot of strength.

Cmd> readdata("")
# File ex02_121.txt from publisher's web site
# Data for Exercise 2.121, p. 216 of IPS4
# Col. 1: len = length of beam made from wood flakes (inches)
# Col. 2: strength = strength of beam (lbs/in^2)
Read from file "TP1:Stat5021:Data:Ch02:ex02_121.txt"
Column 1 saved as REAL vector len 
Column 2 saved as REAL vector strength 

Cmd> list(len,strength) # see what we've got
len             REAL   10   
strength        REAL   10   

Cmd> describe(hconcat(len,strength),mean:T,median:T,stddev:T,iqr:T)
component: median
(1)         9.5       251.5
component: mean
(1)         9.5       291.3
component: stddev
(1)      3.0277      71.195
component: iqr
(1)           5          90

Cmd> stemleaf(strength)
    5    2*|33444
    5    2.|59
    3    3*|3
    2    3.|7
    1    4*|4

         1*|1 represents 110  Leaf digit unit = 10

Here is an example of a data file where two groups have been combined into one. In the book, data for women and men are in two tables. In the original data file from the CD ROM, there was a single column of speeds (which are the same for both men and women) and separate columns for men's and women's stride length. In the edited file there is a column for gender which translates F and M to 1 and 2 respectively, a column for speed and a column for stride length.

Cmd> readdata("")
# File ex02_120.txt from publisher's web site
# Data for Exercise 2.120, p. 215 of IPS4
# Col. 1: gender (factor, 1 = F, 2 = M)
# Col. 2: speed = running speed (ft/sec)
# Col. 3: stride = stride rate (steps/sec)
Read from file "TP1:Stat5021:Data:Ch02:ex02_120.txt"
Column 1 saved as factor gender 
Column 2 saved as REAL vector speed 
Column 3 saved as REAL vector stride 

Cmd> gender
           F           F           F           F           F
           F           F           M           M           M
           M           M           M           M
           1           1           1           1           1
           1           1           2           2           2
           2           2           2           2

Cmd> # F's and M's are case labels; actual values are 1 or 2

Cmd> speed_male <- speed[gender==2] #select men's speed

Cmd> speed_female <- speed[gender==1] #select women's speed

The usage speed[gender==2] selects all values of variable speed for which variable gender is 2, that is just the men's data. This is an example of the use of subscripts in MacAnova.

The left pointing arrow <- is the "assignment operator". It creates a variable with the name on its left side and fills it with whatever is on its right side.

Cmd> speed_male # here are the speeds for men
(1)       15.86       16.88        17.5       18.62       19.97
(6)       21.06       22.11

Cmd> speed_female # same as speed_male
(1)       15.86       16.88        17.5       18.62       19.97
(6)       21.06       22.11

Cmd> stride_male <- stride[gender==2] #select men's stride

Cmd> stride_fem <- stride[gender==1] #select women's stride

Cmd> stride_male # mens stride lengths
(1)        2.92        2.98        3.03        3.11        3.22
(6)        3.31        3.41

Cmd> describe(stride_male, mean:T,median:T) #mean and median for men
component: median
(1)        3.11
component: mean
(1)        3.14

Cmd> stride_fem # women's stride lengths
(1)        3.05        3.12        3.17        3.25        3.36
(6)        3.46        3.55

Cmd> describe(stride_fem, mean:T,median:T) #mean and median for women
component: median
(1)        3.25
component: mean
(1)        3.28

Another way to use a data set is to copy it to the clipboard and then use clipreaddata() to read it. Click on ex02_121.txt, choose Select All on the Edit menu, and then Copy on the Edit menu. Then switch over to the MacAnova menu and type clipreaddata().

Cmd> clipreaddata() # data set was copied to clipboard
# File ex02_121.txt from publisher's web site
# Data for Exercise 2.121, p. 216 of IPS4
# Col. 1: len = length of beam made from wood flakes (inches)
# Col. 2: strength = strength of beam (lbs/in^2)
Column 1 saved as REAL vector len 
Column 2 saved as REAL vector strength 

Cmd> plot(len,strength,title:"Strength vs length for Exercise 2.121 data",\
xlab:"Beam length",ylab:"Beam Strength")

Individual Files

These are grouped by chapter: Chapter 1, Chapter 2, Chapter 4, Chapter 6, Chapter 7, Chapter 8, Chapter 9, Chapter 10, Chapter 11, Chapter 12, Chapter 13 and Appendix.