Klystron wrote:

In sum, I believe that the small change in licensing numbers does not
rise to the level of statistical significance. Given the magnitude of
the unknowable quantities described above, we probably cannot tell
whether the population of live, active hams has grown or shrunken. Is
there a statistician in the house? I would like to see an estimate of
the margin of uncertainty of those numbers (plus or minus x percent),
given the various unknown factors.

I don't claim to be a statistician but it is an easy bet that there are
sufficient variables to place the current result within the error of the
measurement technique. This is especially true when you add that magic
word "active" to your requirement.

The most reliable statistic is the number of new hams being licensed.
Even for those, it's not possible to say if they're "active" (no matter
what your definition might be for that term). But if 27,000 people
invested the effort to obtain a license during the last year, at least
we know that about 1 in 10,000 people cared enough to study for and pass
the exam.

73, Steve KB9X