One of our clients sent us this mystery data. It looked pretty random:
After running our statistical analysis software and drinking a lot of coffee and staring at the output late at night,
we figured out where the natural field boundaries are:
The fields (from left to right) turned out to be:
 | a patient ID number; | |
 | a six-digit date; | |
 | a gender code (0 = male, 1 = female); | |
 | body mass in pounds; | |
 | age in years; | |
 | six columns we’re not sure about, but they may be answers to multiple choice questions; | |
 | and in red, a pair of milligram dosages. (These data come from a clinical drug trial.) |
Because the results were published in a medical journal, we were able to match the means and variances in the article with those in the punch data, to deduce the punch field names.
;^) So we’re no longer afraid of random numbers. We may be able to help you with yours.
