Supplementary Exercises 9.36 and 9.38 of IPS7e
----------------------------------------------

Data on enrollment of US high school graduates aged 16 to 24 years.

9.36
----

MTB > WOpen "H:\VHM\VHM801\Datasets\Minitab\Chapter 9\ex09_036.mtw".
Retrieving worksheet from file: 'H:\VHM\VHM801\Datasets\Minitab\Chapter
9\ex09_036.mtw'
Worksheet was saved on 24/10/2014

MTB > XTABS 'status' 'gender'; 
SUBC>   Layout 1 1;
SUBC>   Frequencies 'count';
SUBC>   Counts;
SUBC>   RowPercents;
SUBC>   ColPercents;
SUBC>   TotPercents; 
SUBC>   DMissing 'status' 'gender'.
Tabulated statistics: status, gender 

Using frequencies in count

Rows: status   Columns: gender

           g1      g2      All

s1        890     969     1859
        47.88   52.12   100.00
        18.38   17.37    17.84
        8.540   9.299   17.839

s2        340     403      743
        45.76   54.24   100.00
         7.02    7.22     7.13
        3.263   3.867    7.130

s3       2897    3321     6218
        46.59   53.41   100.00
        59.83   59.53    59.67
       27.800  31.868   59.668

s4        249     383      632
        39.40   60.60   100.00
         5.14    6.87     6.06
        2.389   3.675    6.065

s5        306     366      672
        45.54   54.46   100.00
         6.32    6.56     6.45
        2.936   3.512    6.449

s6        160     137      297
        53.87   46.13   100.00
         3.30    2.46     2.85
        1.535   1.315    2.850

All      4842    5579    10421
        46.46   53.54   100.00
       100.00  100.00   100.00
       46.464  53.536  100.000

Cell Contents:      Count
                    % of Row
                    % of Column
                    % of Total

Note that the categories for rows and columns are labeled s1-s6
and g1-g2, respectively. It is fairly easy to create a more informative
category names using the Data-Code-Text_to_Text menu.

Answers to questions:
--------------------
(a) There were 890 thousand male students in a two-year college, 
full-time program  (s1 x g1 - Count, or IPS data table).

(b) The marginal gender distribution is the totals for Men and Women.
There were 4842 thousand male students and 5579 thousand female students, 
and the proportions of male and female students were therefore 
4842/(4842+5579) = 46.46% and 53.54%.
(All x g1 and g2 - Count and % of Row)

(c) The marginal distribution of status among all students was (in %)

          s1    s2     s3    s4    s5    s6     All
All     1859   743   6218   632   672   297   10421
       17.84  7.13  59.67  6.06  6.45  2.85  100.00

(entire column for All - Count and % of Col)

note: the percentages are obtained by dividing the entry in the All column by 
the overall Total, e.g. 1859/10421=17.84%.


9.38
----
For male students:

          s1    s2     s3    s4    s5    s6     All
g1       890   340   2897   249   306   160    4842
       18.38  7.02  59.83  5.14  6.32  3.30  100.00

(entire column for g1 - Count and % of Col)

For female students:

          s1    s2     s3    s4    s5    s6     All
g2       969   403   3321   383   366   137    5579
       17.37  7.22  59.53  6.87  6.56  2.46  100.00

(entire column for g2 - Count and % of Col)

There seems to be only minor differences between the distributions
of males and females. (Note, despite the visually similar distributions, a 
chi-square test for homogeneity between gender (or independence) is strongly
significant; this is because of the large sample size, even without converting
the counts from thousands).