Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: re: SSC Activity, November 2009


From   Roy Wada <roywada@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: re: SSC Activity, November 2009
Date   Sat, 5 Dec 2009 13:24:25 -0800

> Some doubt was voiced in an earlier posting regarding the legitimacy of the download
> statistics. As it happens, we have been analysing these statistics for suspicious
> activity for several months.

I am going to add the following because I don't like the idea of
someone given a license to keep on doing this:

I am not optimistic about checking for IP addresses. They can be
reassigned automatically with each visit. I have not done it myself,
but this is not terribly difficult. Addresses can be rented by
thousands. Kit could re-analyze the data at the disaggregated level,
check the revisit frequency over the year and compare what they
downloaded to the average IP behavior, but this would be tiresome and
will not always work because it assumes that the same list of
addresses have been used.

It's interesting that this has gone on this long without anyone saying
anything. If you are suspecting the worst, you might be right.  Just
about everything else have been done at promoting a program, so why
not cross the last line. Manipuation of the ssc ranking is not
especially surprising given that virtually all public rankings (e.g.
the bestseller list, best college, etc) have been manipulated.

The massive downloads started about 15 months ago when I chided a few
people for not leaving other people alone. It started out that way but
it apparently developed into another way of promoting a program. Did
this person go away or is he still aroud? If I were doing this, I
would create noise both externally (download someone else's programs)
and internally (create excuses for sudden popularity).

I did a simple trend analysis about six months ago, using publicly
available informaiton. Trends are very difficult to fake. The analysis
tells me that outreg, outreg2, xml_tab, and tabout could not have been
manipulated to any important degree. It also tells me Kit Baum has
been accurate in his reporting because it shows up on my graphs.

I am updating it right now. I should have this ready by tomorrow.

Roy
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index