A number of people suggested that the culminating event of
the Democratic National Convention, Barack Obama's speech
formally accepting the party nomination, should be a global
event. For example, on Fri, 29 Aug 2008, Anna Zaigraeva wrote:
Hello! I see you guys analyzed some data
from June 3rd, when Obama secured the nomination. Are you going to look at
the time on August 28th when he was speaking live in Denver? The speech was
broadcast to something like 38 million households, according to the
Nielsen ratings, so more people have watched him speak than have watched the
Olympic Games opening! Much less diffuse than securing the nomination - the
window would only be as long as the speech.
We did indeed plan a formal analysis, as part of a
developing series of replications of political events that,
while US-centric, nevertheless interest huge numbers of
people around the world.
I did not note the exact time Obama's speech began, nor
could I find authoritative timing via Google, but it
was certainly well after the appointed time (8 pm in
Denver, Colorado). The GCP event was set for a two hour period that
would definitely include all of the Obama address, as well as a bit of
the prelude and introduction and probably at least half an
hour of aftermath.
The result is a positive trend, with
Chisquare 7446.76 on 7200 df, for p = 0.021 and Z = 2.039.
The graph shows a fairly consistent trend, beginning at 8
pm, local time in Denver, before the speech began and
continuing for about an hour post speech.
It is important to keep in mind that we have only a tiny statistical
effect, so that it is always hard to distinguish signal from
noise. This means that every "success" might be largely
driven by chance, and every "null" might include a real
signal overwhelmed by noise. In the long run, a real effect can
be identified only by patiently accumulating replications of similar analyses.
|