Quality of Proof in Discussion

General Discussion of Diary Posts and Questions on Beekeeping Matters
Post Reply
Allen Dick
Site Admin
Posts: 1824
Joined: February 25th, 2003, 10:09 pm
Location: Swalwell, Alberta
Contact:

Quality of Proof in Discussion

Unread post by Allen Dick »

At risk of beating this to death, let us all remember back to grade school and all the various characters we knew all too well; the bullies, the cheats, the deranged, the narcissist, the slow one, the compulsive liar, the jock, the Barbie...

Back then, in our innocence, they were transparent to us but they all grew up and most learned to conceal their iniquities behind a facade of politeness and sometimes professionalism. They are still out there and in their core they have not changed much.

Many went to university, and some cheated their way through and even went on to to grad school where they had to do lab work and contribute to published studies. Some even got tenure.

Even those of us who know only a little about science and bees realize, for example that the "Harvard" study "proving" that neonics cause CCD was entirely bogus in spite of the fact it was treated with respect and cited in many articles.

In many cases, volume of publication, not quality, matter, and there are many situations where, "I'll scratch you back if you scratch mine" is the operative force in jurying.

Get the picture? Wall Street is not the only place where white collar deceit and self-dealing runs rampant.

I am not picking on anyone here since citing studies from only the title, the abstract, or a paraphrase in media seems to be a common practice, but if we really want to learn and share real knowledge not speculation or "win" a debate, we have to be very careful what we cite and read each study critically to see if all the information is there, and to consider whether the data was given a statistical makeover to fit the needs of the sponsor.

Stating our opinions is one thing and entirely valid. No proof is required for an opinion and none need be given.

If we do choose to try to justify our opinions when challenged, however, we can't just find titles we like and sprinkle citations like nails on the road behind us if we feel we are being too closely pursued for comfort in a discussion.
Allen Dick, RR#1 Swalwell, Alberta, Canada T0M 1Y0
51° 33'39.64"N 113°18'52.45"W
http://www.honeybeeworld.com/Allen%27s%20Beehives.kmz
Forum owner/janitor
---
Customise your experience at Honeybeeworld Forum at your User control Panel
Change the appearance and layout with your Board Preferences
Please upload your own avatar picture at Edit Avatar. It's easy!
Return to main diary page
User avatar
Colino
Forum Regular
Posts: 390
Joined: April 12th, 2014, 11:01 am
Location: Whitla Ab. (SouthEastern Alberta)
Contact:

Re: Quality of Proof in Discussion

Unread post by Colino »

Allen Dick wrote:At risk of beating this to death, let us all remember back to grade school and all the various characters we knew all too well; the bullies, the cheats, the deranged, the narcissist, the slow one, the compulsive liar, the jock, the Barbie...
Grade school? You just described my inlaws. :shock:
Merry Merry, Happy Happy and all the best in the New Year.
Colino
Narcissism is easy because it's me or I, Empathy is hard because it's they or them.-Colino
Allen Dick
Site Admin
Posts: 1824
Joined: February 25th, 2003, 10:09 pm
Location: Swalwell, Alberta
Contact:

Re: Quality of Proof in Discussion

Unread post by Allen Dick »

Hohoho. Merry Christmas!
Allen Dick, RR#1 Swalwell, Alberta, Canada T0M 1Y0
51° 33'39.64"N 113°18'52.45"W
http://www.honeybeeworld.com/Allen%27s%20Beehives.kmz
Forum owner/janitor
---
Customise your experience at Honeybeeworld Forum at your User control Panel
Change the appearance and layout with your Board Preferences
Please upload your own avatar picture at Edit Avatar. It's easy!
Return to main diary page
User avatar
TWall
Forum Regular
Posts: 36
Joined: July 6th, 2012, 11:18 am
Location: Mt Vernon, OH USA

Re: Quality of Proof in Discussion

Unread post by TWall »

Allen,

It can be really hard to sometimes separate the wheat from the chaff with research results. There can be pretty subtle nuances that can make a big difference in whether something is 'real' or not. Statistical significance is ultimately what the researchers often live and die on. That does not always translate to real world significance.

Tom
Allen Dick
Site Admin
Posts: 1824
Joined: February 25th, 2003, 10:09 pm
Location: Swalwell, Alberta
Contact:

Re: Quality of Proof in Discussion

Unread post by Allen Dick »

It can be really hard to sometimes separate the wheat from the chaff with research results. There can be pretty subtle nuances that can make a big difference in whether something is 'real' or not.
Agreed.

My main point is that I expect people to eat their own cooking, i.e. read and be prepared to supply the text of any study they cite.

I have no beefs about seeing references to studies that people have actually read and thought about, and can provide access to the text for others to dissect without difficulty, delay, or expense. I realise that some studies that I consider complete junk, others might see some value in and we all can learn from discussion.

We can hardly discuss studies intelligently, though, when we are only provided the title or an abstract and the text is unavailable to the members, except at cost.

In my experience, that sort of 'proof' is mostly rhetorical. Whether intentional or not, it amounts to a bluff.

Such bogus debate may pass on Beesource, but bluffs will almost always be called here. If the player has the cards, no problem, but if it is an empty bluff, we are spared.
Statistical significance is ultimately what the researchers often live and die on. That does not always translate to real world significance.
True, and although statistics can find patterns where none are obvious, in untrained or unscrupulous hands, stats provide a perfect smokescreen to hide failure or deceit. There is considerable latitude in the choice of methods and the selection of data, even if the data is good data to begin with, and all too often the actual data is faulty.

To me the main value in any study, no matter how 'good' or 'bad' is the method and observations. Given full disclosure in the body, I often learn things that were not intended, but useful, and even when I conclude the study failed to prove what it claims, I have learned something valuable.
Allen Dick, RR#1 Swalwell, Alberta, Canada T0M 1Y0
51° 33'39.64"N 113°18'52.45"W
http://www.honeybeeworld.com/Allen%27s%20Beehives.kmz
Forum owner/janitor
---
Customise your experience at Honeybeeworld Forum at your User control Panel
Change the appearance and layout with your Board Preferences
Please upload your own avatar picture at Edit Avatar. It's easy!
Return to main diary page
User avatar
TWall
Forum Regular
Posts: 36
Joined: July 6th, 2012, 11:18 am
Location: Mt Vernon, OH USA

Re: Quality of Proof in Discussion

Unread post by TWall »

Allen Dick wrote:Given full disclosure in the body, I often learn things that were not intended, but useful, and even when I conclude the study failed to prove what it claims, I have learned something valuable.
Allen,

I had a great example of that with my masters research. I studied nitrogen and potassium fertilizer management for cauliflower production. Some of the data I generated was N, K and dry matter accumulation for the life cycle of the crop. After a certain point in crop development there was no additional N, K or dry matter accumulation. So, any additional nutrients provide were not utilized by the crop and would be at risk of moving into groundwater.

Tom
Allen Dick
Site Admin
Posts: 1824
Joined: February 25th, 2003, 10:09 pm
Location: Swalwell, Alberta
Contact:

Re: Quality of Proof in Discussion

Unread post by Allen Dick »

Some of the data I generated was N, K and dry matter accumulation for the life cycle of the crop. After a certain point in crop development there was no additional N, K or dry matter accumulation. So, any additional nutrients provide were not utilized by the crop and would be at risk of moving into groundwater.
That is the kind of accidental discovery that often starts a whole new line of thought and can be far more important and valuable than the original goal.

When reading studies, we often find interesting things that were not intended to be the subject of the study and not even found remarkable by the authors.
Allen Dick, RR#1 Swalwell, Alberta, Canada T0M 1Y0
51° 33'39.64"N 113°18'52.45"W
http://www.honeybeeworld.com/Allen%27s%20Beehives.kmz
Forum owner/janitor
---
Customise your experience at Honeybeeworld Forum at your User control Panel
Change the appearance and layout with your Board Preferences
Please upload your own avatar picture at Edit Avatar. It's easy!
Return to main diary page
Post Reply