Welcome! Log In Create A New Profile

Advanced

Experimental Design and Statistics.

Posted by Old Spice 
This forum is currently read only. You can not log in or make any changes. This is a temporary situation.
Experimental Design and Statistics.
August 04, 2020 01:15AM
I see both of these mentioned here and there on the Towards 0.1 Micron Forums and other places on the internet. But I don't often see detailed discussion of them. I also don't think that resources available for them are very accessible for the bulk of the community interested in knives. I.E they aren't cracking open textbooks on regression analysis, ANOVA, model selection, or any other advanced topic on statistics. While that probably isn't necessary I think basic literacy in math and statistics is needed. Especially to get the full value from the literature posted in this sub-forum.

We should have a place we can point people to, where they can find a detailed explanation of how an experiment is designed (or how to design one) to produce quality data, and how it is analyzed to produce a meaningful interpretation.

We can also just talk about general experience with the topic.

I will be looking for articles to post in here as examples with some explanation of what they did and why other than the obvious.
Re: Experimental Design and Statistics.
August 04, 2020 01:28AM
It might be useful to simply take a post/video, etc. which claims a conclusion and break it down :

[www.cliffstamp.com]

That forum was created in part to do that for literature, there is no reason why that can't be extended to critical analysis of work in general.
Re: Experimental Design and Statistics.
August 04, 2020 02:29AM
Cliff, do you mean to suggest we use the “Bro, Do You Even Science?” forum more broadly to encompass what Old Spice is suggesting?

Personally I think it would be pretty helpful or at least interesting to have a specific thread, as Old Spice suggested. Just to sort of consolidate the topic a little bit by having a single reference source for it.
Re: Experimental Design and Statistics.
August 04, 2020 12:20PM
That is possible, however one long thread is difficult to read/parse and reference.

I created a forum, we could post individual threads there on reviews, and have a pinned thread on method/summaries, links to useful information, etc. .
Re: Experimental Design and Statistics.
August 04, 2020 04:00PM
Alrighty Cliff, fair enough. thumbs up