To stat or not to stat, that is the question...
posted on
Aug 11, 2007 11:50AM
Whether 'tis nobler in the mind to suffer
The stats and interpretations of outrageous analysis,
Or to take arms against a sea of numbers,
and by opposing end them? To die: to sleep;
Or to put to sleep all, and to end
The heart-ache and the thousand brainy shocks
That we endow on the masses.
Geoff,
I would have to admit that I have dived into the analysis
about as far as I wish to go at this time.
In the "suspect" P-Values, I would defer to Dr. Sorgente's knowledge.
Regardless, what we both have brought to the forefront is the interpretation
of the stats that there is a significant performance of Agrastim in treating
Necrotic Enteritis.
We differ on the interpretation of the stats of the boiler chicken study.
And, no, I was not using the Feed Conversion Ratio as kg to produce kg
but was using the difference of groups' FCR to determine a factor to use
in the increase/decrease of the FCR.
Both stat pages are useful to initially show to prospective customers the
potential of Agrastim. More studies are underway both by BioAgra and by
potential customers.
These two pages of stats are just the beginning of many studies
with rigorous protocols to produce objective results.
What we both need do, if we are presented in the future with more stats,
is to review the stats in further depth before posting, to avoid
misinterpretation with resulting concern/panic from our readers.
And acknowledge that there is far more testing/studies being performed
with Agrastim; with continued data collection and analysis than what we
have presented to us.
And, I would be bold to say, that I believe that we both agree that current
stats do indicate that Agrastim is a viable alternative to antibiotics, thus
enabling the industry serving the human food chain to remove antibiotics
while still averting the much larger losses to their production that would result
if nothing was administered to the animals.
I do interpret that Agrastim administered as a normal part of the feeding
process would also generate fiscal savings to the industry both in the
health of the animal and in its ability to gain weight using less food.
I acknowledge that Geoff differs from my interpretation for this application.
That is the bottom line.
Now, this is not quite an apology but more empathy to those who have called
or e-mailed me regarding the mental overloads or enhanced sleep while
reading my/our previous posts filled with numbers and attempts at statistical
interpretations. I'll try to not compose such posts very often.
Please do not use my previous posts as a primer guide to statistical analysis.
With imprecise numbers used mixed with precise numbers, I can easily see
holes to drive through, especially if I can slip pounds in for kg. And, as Geoff
illustrated in his previous post, there are disagreements of interpretation,
especially when one makes assumptions regarding another's use of numbers.
I quote a saying we learned in day 1 of Stats 101:
"There are liars,
There are Damn liars,
and there are statisticians."
I do like to see statistics and I can be informed by them. I am pleased that
BioAgra has posted some stats for all of us to see. I interpret them as very
bullish for the need for Agrastim, resulting in significant sales.
And I admit that Geoff's challenge pushed me into delving into statistical
analysis such as I have not gone in several decades. I guess that can be
interpreted as a good thing.
Now, off to WallyMart to get a new bottle of aspirin.
Kent
Oh, and I avoided posting last night while watching a boxing match and
I have had Nasa TV on while composing this post and am amazed at how
beautiful the HD cameras on the astronuts helmets appear on my TV.
Incredible progress.