A test prep company, *Acme 4 Sure Pass, Unlimted*, set up example tests
for two types of tests, Test A and Test B. To evaluate how well the targeted
potential customers will excel in these prep tests, the company had 200
evaluators in each of the six regions (a total of 1200 evaluators) tried on
the tests. Each evaluator was asked to worked on both tests with
10 similar repetitions (20 points each) on each test. However, each evaluator
may end up not finished some repetitions.
A data set in plain text testEval.txt, accessible
by SAS FILENAME URL statement, contain the outcomes from this study.
It is a combined raw score book from
6 regions identified by column 1 (regions A through F).
Column 2 contains evaluators' IDs (1 through 200 in each region).
The next 10 columns are the respective outcomes from the 10 repetitions
of Test A. The last 10 columns are the outcomes from the 10 repetitions
of Test B.
Your WIN will set a random seed so that the raw score data of a randomly
selected region (from regions A to E) will be assigned to you to produce
a summary report. The summary report for region F was carried out and
presented in finalExamEgOutput.txt for
your reference.

A SAS data set named ** raw** will
first be created for the region
you are working on. This data set has columns named

**Evaluator, A1,
A2, A3, A4, A5, A6, A7, A8, A9, A10, B1, B2, B3, B4, B5, B6, B7, B8, B9,**
and **B10**

. This data set is then split into two data sets,
`testA`

`testB`

`testA`

**Evaluator**

, **A1**

through
**A10**

, and an additional column named
**a_ok**

which gives the number of repetitions of test A each
evaluator had actually completed.
Moreover, only observations in which 6 or more repetitions were completed
will be kept in `testA`

`testB`

**Evaluator**

, **B1**

through
**B10**

and **b_ok**

.
Each of these two new data sets will be updated with two additional
variables added: variable

gives the mean score of
the 10 repetitions, variable **xbar**

is then created
based on the values of variable **Category**

:
**xbar**

value ofFinally, these two data sets (only variablesxbarCategory------------- ------------- [19,20] Very High [17,19) Somewhat High [15.5,17) Moderate [0,15.5) Low

**Evaluator**

,
**xbar**

and **Category**

in each data set will be kept) are concatenated and interleaved to form the
final data set named `combined`

**test**

is created
to indicate test types.
The final summary report will be produced using PROC REPORT on the
** combined** data set.

- syntax errors;
- logic errors but not syntax;
- errors due to wrong calculations or inadequate codes.

jung-chao.wang@wmich.edu