Promotional Supplement Testing - Number of Testers

[Viperspit]

What are your thoughts on the number of Testers a company should ask for when testing a new release product?

[TinyMan]

I like it around 4-6 usually . Too many, and there are just too many logs of one product. Too few, and the problem of flakes or incomplete logs (or if you go extremely too few, not enough input).

If you're doing different variations for a set of testing (like how MST and CL often do), then the answer changes a bit.

[Docta J]

I'd think you'd get more flakes the more testers you have. I think 5-10 is a good number.

[MattyH7688]

5-10.. to many the overall quality of loggers will most likely go down

[Viperspit]

So so far, keep it on the smaller side? what about when you want to get some statistical data for reference points? i.e. a fatburner, say 30 testers, for a 30 day period, to calculate avg. weight loss? a sampling of 5 or 6 is fairly small wouldn't you think? jsut throwing out some of my thoughts.
thanks for yours.

[ab-swing asianbabe]

I think it should just be left to the discretion of each company how many logs they want to have sponsored.

[CapitalJay]

I don't really have an opinion here on an exact number, but I think if logs are to be used for statistical data the sample size should be statistically significant. I would think a minimum of 30 data points would be needed in addition to a like-sized control group.

</engineer mode>

[Viperspit]

I think it should just be left to the discretion of each company how many logs they want to have sponsored.

Hi Liv, nice to see ya. You make a good point, but there's different schools of thought on this topic lately it seems.

I don't really have an opinion here on an exact number, but I think if logs are to be used for statistical data the sample size should be statistically significant. I would think a minimum of 30 data points would be needed in addition to a like-sized control group.

</engineer mode>

Agreed Jay, thanks. You just prompted another thought I had. Seems some are not to thrilled with larger groups Testing. I wonder if the smaller groups should be done in places like bb.com and the larger groups should be done on the companies own Website/Forum? Guess that's another topic though.

[Boots]

I'd think you'd get more flakes the more testers you have. I think 5-10 is a good number.

5-10.. to many the overall quality of loggers will most likely go down

x2 on both posts

[RobW]

Where is the RAT option

[pu12en12g]

The more the better, as long as they are all quality board vets ! (good luck )

[Viperspit]

The issue I see and this is personal opinion here, is that when you have too few Testers, I believe it increases your risk for a non quality event. This pertains especially to the case where you are trying to get new Testers and the Testers may not have been on the board that long.

Yes, you can minimize the risk factor by going with Vets as Paul suggests, which is a smart idea, but we also want to give new people a chance as well. Why is this? 2 fold, 1. People sometimes get tired of reading the same old Logs by established members. 2. Eliminate the "hand picking" factor when it comes to Testers. yes, yes, we've all seen it before.

[pu12en12g]

The issue I see and this is personal opinion here, is that when you have too few Testers, I believe it increases your risk for a non quality event. This pertains especially to the case where you are trying to get new Testers and the Testers may not have been on the board that long.

Yes, you can minimize the risk factor by going with Vets as Paul suggests, which is a smart idea, but we also want to give new people a chance as well. Why is this? 2 fold, 1. People sometimes get tired of reading the same old Logs by established members. 2. Eliminate the "hand picking" factor when it comes to Testers. yes, yes, we've all seen it before.

CLIFFS: I'll expect a package in the mail

[ScoobyDrew]

Being a Reliability Engineer, I know a good sample size is between 6 to 12 samples. However, the more data points you get, the better your distribution of results will match your whole given population. Then again, it also might depend on how much it costs per data point.

[RobW]

The issue I see and this is personal opinion here, is that when you have too few Testers, I believe it increases your risk for a non quality event. This pertains especially to the case where you are trying to get new Testers and the Testers may not have been on the board that long.

Yes, you can minimize the risk factor by going with Vets as Paul suggests, which is a smart idea, but we also want to give new people a chance as well. Why is this? 2 fold, 1. People sometimes get tired of reading the same old Logs by established members. 2. Eliminate the "hand picking" factor when it comes to Testers. yes, yes, we've all seen it before.

What would really impress me would be if some of the companies on here did "Real" tests (and this is not a dig at just ****). What I mean by "Real", is not just sending out free product to willing recipients and asking for feedback. But paying for testing in "Real" clinical conditions were the test subjects were qualified and did not know if they were getting the "Product" or a "Placibo".

This i think would be a step forward...

Just sayin...

[Viperspit]

CLIFFS: I'll expect a package in the mail

or your sending the Assasin Squad?

Being a Reliability Engineer, I know a good sample size is between 6 to 12 samples. However, the more data points you get, the better your distribution of results will match your whole given population. Then again, it also might depend on how much it costs per data point.

My former life as well. Let me guess by your location...Automotive?

What would really impress me would be if some of the companies on here did "Real" tests (and this is not a dig at just ****). What I mean by "Real", is not just sending out free product to willing recipients and asking for feedback. But paying for testing in "Real" clinical conditions were the test subjects were qualified and did not know if they were getting the "Product" or a "Placibo".

This i think would be a step forward...

Just sayin...

Valid point. I have no clue how much something like this would run, but I'm guessing it ain't cheap. You'd have to contract the whole thing out to a medical facility of sorts.

[Mickfootie]

What are your thoughts on the number of Testers a company should ask for when testing a new release product?

Depends on the different ways in which you want them to test a product.
You will need a few for testing a new product as a standalone then some to double dose, then some to stack with other products... etc...

[punksurfer024]

Depends on the different ways in which you want them to test a product.
You will need a few for testing a new product as a standalone then some to double dose, then some to stack with other products... etc...

sounds like a familiar testing thread

[fitnfirm]

I say give everybody free stuff

[fitnfirm]

CLIFFS: I'll expect a package in the mail

Im all over that !


*FnF runs to the stash closet to package up "stuff" *

[ZDub212]

I say give everybody free stuff

Sounds good to me!

[TE]

What are your thoughts on the number of Testers a company should ask for when testing a new release product?

5-10 is sufficient.

Are you checking your cup size in your av? Looks like an "a" cup to me.

[sobelik]

I think a mix between vets and newer but promising people to add up to 10 would be the best. 5 vets and 5 newer people

[Outside backer]

as long as I get selected i dont care it its 1 or 100

[Mertle]

I think a mix between vets and newer but promising people to add up to 10 would be the best. 5 vets and 5 newer people

x2

[Viperspit]

5-10 is sufficient.

Are you checking your cup size in your av? Looks like an "a" cup to me.

That was me throwin up da WestSide. changed.

I think a mix between vets and newer but promising people to add up to 10 would be the best. 5 vets and 5 newer people

I like the 50/50 split idea, but I'm beginning to think the overall number is really going to be dependent on the situation.

[terracotta]

I think a mix between vets and newer but promising people to add up to 10 would be the best. 5 vets and 5 newer people

Yes.. you need a mix. I was looking for old logs for something once, and I think I found 5 incomplete logs on the same product.

[redhawk76]

I think a mix between vets and newer but promising people to add up to 10 would be the best. 5 vets and 5 newer people

Agreed.

I like the 50/50 split idea, but I'm beginning to think the overall number is really going to be dependent on the situation.

Also agreed. If you are trying to provide data of fat lost or muscle gained, the more the better. (to an extent)

However you have to find loggers willing to provide pics and exact measurements to be able to confidently claim any of the results.

But logs that are just for feedback (like a pre-wo supp) there is no need to have 30 logs obviously

[terracotta]

If you are going to have more than 10 logs, I would say put them on the company website, and then make a product review summary thread with the final reviews.

[fitnfirm]

I just had 400 pounds of packages delivered yesterday of ALR products........... What will I ever do with all these ?