>> But 100K is a decent sample
> That 100K might be a decent sample if we had total medical histories...
That 100K is an extravagantly large sample.
I don't think all that personal info is needed to answer the question, as we
need only survey those beekeepers who meet "qualifying" criteria, such as 8+
years working bees hands-on, regular stings, no history of other major
illness and so on.
100K is such an overkill sample, we can cut it by 1/3 without even trying -
to represent a total US population of 300 million, if we wanted a 0.5% error
rate, and a 99% confidence level, a sample size of just over 66,000 is
required.
And the sample size need not be anywhere near so large for solid acceptable
results for a survey asking beekeepers "Did you get sick with Coivd-19
despite 'regular' bee stings?"
If we leave everything above alone, and accept a higher margin of error,
16,587 respondents would yield a 1% margin of error 4,147, 2%, and only 644
sampled, 5%.
Tweaking the acceptable confidence level, 38,410 takes us to 95%. 27,053
takes us to 90%.
All of the above scenarios are "publishable" in any legit referred journal
one might choose, so this question CAN be answered with a manageable number
of people.
I think Bee-L hovers somewhere around 1000 subscribers (Aaron?). Don't know
what fraction are in the US, but with a sample size of 1037, we can have a
99% confidence level and a 4% margin of error in representing that same
300000000 people, and while some opine so often, one wonders when they have
time to open a hive, the group as a whole clearly does get stung often.
And we really should wait a bit, as the experience of the major cities is
only now starting to be repeated in the rural areas and smaller cities.
***********************************************
The BEE-L mailing list is powered by L-Soft's renowned
LISTSERV(R) list management software. For more information, go to:
http://www.lsoft.com/LISTSERV-powered.html
|