Gary Klass wrote:

> The whole idea of a statistically significant trend is bogus, I think.
>
> In this case, you probably have a situation where you've added 20 or 30
> people to the staff over five years.
> A significance test would allow you to test the hypothesis that you have
> added nobody.
> Does that make sense?

I had assumed that the economists wanted to distinguish an "actual"
increase from a randomly fluctuating but otherwise stable staff level.

But to do that, they would need to have a sense of how staff levels
would fluctuate randomly, which is a very different null from sampling
variation.  I suppose they could simulate it -- take an agency with N
people, each of which has an L% chance of leaving, and in which
vacancies are filled with probability R of replacement every year.  Run
it over five years, and iterate the runs 10,000 or 100,000 times.  Then
compare actual staff changes to the density of simulated staff changes.

Surely it makes more sense to look at personnel authorizations, if they
can.  If personnel authorizations went up, then they did.  With certainty.

--
James S. Coleman Battista
Dept. of Political Science, Univ. of North Texas
[log in to unmask] (940)565-4960

**********************************************************
             Political Methodology E-Mail List
        Editor: Karen Long Jusko <[log in to unmask]>
**********************************************************
        Send messages to [log in to unmask]
  To join the list, cancel your subscription, or modify
           your subscription settings visit:

          http://polmeth.wustl.edu/polmeth.php

**********************************************************