Feeding experiments with controls.

[BEES4U]

Torula yeast suppliers.

--------------------------------------------------------------------------------

I am looking for a supplier of Torula yeast.
Hopefully one near home to keep the shipping cost down.
Please send comments.
Spray dried brewers yeast is used with a low salt, 02 %, is available.
I want to do some in-field feeding trials.

Block +TY = With Torula yeast

Block +BY = with Brewers Yeast

Block +50:50 = 50% TY + 50% BY

Block ML15 = Mann Lake 15% patties.

Block C (Minus +TY, +BY and + 1:1) = Contol

Block KJ = Kieth Jarrett. (Kieth, I wanted to know if you read this part?):D

Block P = % pollen. This could complicate the experiment.

The syrup used will be C & H granulated sugar mixed in a 2:1 by volume.

As in all experiments I need to keep the variables down to a small influence on the Raw and Final Data.--I need a very good conclusion.

I have the queen stocks to test the Breeder queen daughters.
I will try to use test blocks of at least 50 hives/block so that the experiment will produce conclusive evidence.
The reason that I want to do this research is because I have read other data reports and they did not have any controls or their data base was only 10 hives.. Now how can you say that something works or it did not work without controls or such a small data base?

Please submit your ideas.
Regards,
Ernie Lucas Apiaries
__________________
Do it right the first time to save time, material and money.

[tecumseh]

ernie writes:
As in all experiments I need to keep the variables down to a small influence on the Raw and Final Data.--I need a very good conclusion.

tecumseh writes: I don't have any idea as to the numbers of hives you plan to use per treatment? large enough numbers (samples) to accurately reflect population is typically the first flaw of many experimental designs.

there are statistical packages (at one time these were quite a challage, but now exist in pre packaged software form as to be usable/doable by almost anyone who own a computer) that will whittle out the trees from the forest. there is however nothing wrong (and significant benefit) to maintaining some degree of simplicity... or, it is in my mind beneficial to knock off the easier questions before moving on to a more complex question.

will you be feeding these options in the hive or free choice external to the hive? I would think??? this would not so much effect results (ie mean), but would have some direct impact on application cost plus (my experience would suggest) some greater variation in result in the individual treatments.

please do keep us informed as to the progress of your experiment. it sounds quite interesting and would have some obvious economic implications.

[Bodo]

There is a free stats program out there that is very powerful available at:

http://www.r-project.org/

If you need assistance with running your stats, please feel free to drop me a line.

I would also suggest that when you decide to start this study you do a little reading on experimental design. It'll be impartive to keep the variables to a minimum (location, weather, forage, genetics, etc)

[MichaelW]

Ernie,

What is/are your response variable/s? like: honey production, population, diseases, survival?

Also what is Block= % pollen?

[MichaelW]

Also, technically, your list is 'treatments' not 'blocks'
'blocks' are done to separate influences on your experiment outside your
'treatments' so those influences don't interfere with your results.
For example, its good to block your experiment per location.
Each location = one block. The thing is though you have to have
each treatment represented per block(location).

[Texas Bee]

Man that stuff is high......


http://www.iscatech.com/exec/DocLib/ISCA_Product_Brochure_-_Ball_Trap_Torula_Yeast.pdf