R help - Jul 2024 - OFF TOPIC: Nature article on File Drawer Problem in Reserach

Again, this is off topic, not about statistics or R, but I think of
interest to many on this list. The title is:

"So you got a null result. Will anyone publish it?"

https://www.nature.com/articles/d41586-024-02383-9

Best to all,
Bert

Dear Bert,

You have made my day!! Your post is a great help and very useful in my
field.

The paper is not among the off-the-shelf research output. Some of us, who
get into unenviable conflict and disputation with some reverenced
authorities in our field, understand the weight of the article. I had not
even consumed half of it before I decided to thank you. I will quickly go
back to see how it goes. I am sure it is going to be one of my best
collections for the year!!

Thank you very much. I often follow your posts, even if the topic does not
concern me. Your tough comments always make sense to me. Today, I have
gained something vital I should have lost if I overlooked it because of the
subject heading: "
OFF TOPIC"

Please accept my warmest regards
Ogbos

On Wed, Jul 24, 2024 at 3:44?PM Bert Gunter <bgunter.4567 at gmail.com>
wrote:
> Again, this is off topic, not about statistics or R, but I think of
> interest to many on this list. The title is:
>
> "So you got a null result. Will anyone publish it?"
>
> https://www.nature.com/articles/d41586-024-02383-9
>
> Best to all,
> Bert
>
> ______________________________________________
> R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>
	[[alternative HTML version deleted]]

Bert,

Although the ?ticle was interesting, I have to wonder how much publishing there
has been in formal journals related to R, especially recently, that is of a
research variety.

I am thinking of an example and wonder if we picked something that is often
re-implemented by many parties such as ways of doing graphics or making and
manipulating things like variants of data.frames. If someone came up with some
new design, such as tibbles or data.table and hypothesized they would be better
in some ways, then that could be the basis for doing some serious testing and
perhaps publishing results. Sometimes it could just be a comparison of all kinds
of cases and a discussion of when one or the other might be better, but other
times, the hypothesis might be determined in advance to be looking for a
specific outcome and if wrong, publishing it fairly would let people know that
your guess was wrong!

I once had a chance to get an Erd?s number of two as my adviser later published
several times with Paul Erd?s, but I ended up proving the opposite of my
hypothesis, which was not really worthy of publishing! LOL!

I am curious where people go to see research papers in various aspects of
Computer Science, or in ones dedicated to specific languages or systems. Is
there the same publish or perish aspect in academia as for some older sciences
or is the field different in many ways and perhaps often seen as a helper in
other disciplines so you can publish elsewhere?

Some areas of the field like aspects of AI, might still be considered quite
active but perhaps other areas are seen as less worthy of much further analysis
and refinement. Unless actively being changed, where do programming languages
like R fit in?


-----Original Message-----
From: R-help <r-help-bounces at r-project.org> On Behalf Of Bert Gunter
Sent: Wednesday, July 24, 2024 10:44 AM
To: R-help <R-help at r-project.org>
Subject: [R] OFF TOPIC: Nature article on File Drawer Problem in Reserach

Again, this is off topic, not about statistics or R, but I think of
interest to many on this list. The title is:

"So you got a null result. Will anyone publish it?"

https://www.nature.com/articles/d41586-024-02383-9

Best to all,
Bert

______________________________________________
R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

Here is one response: https://doi.org/10.1093/jisesa/iew092
Or paraphrased: yes.

Regards,
Tim

-----Original Message-----
From: R-help <r-help-bounces at r-project.org> On Behalf Of Bert Gunter
Sent: Wednesday, July 24, 2024 10:44 AM
To: R-help <R-help at r-project.org>
Subject: [R] OFF TOPIC: Nature article on File Drawer Problem in Reserach

[External Email]

Again, this is off topic, not about statistics or R, but I think of interest to
many on this list. The title is:

"So you got a null result. Will anyone publish it?"

https://www.nature.com/articles/d41586-024-02383-9

Best to all,
Bert

______________________________________________
R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.r-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

I know you didn't want to stimulate discussion, but the problem is not
confined to publication.  "Adverse reaction to medication" monitoring
programs are plagued by a similarly massive under-reporting problem:
adverse reactions are seldom reported unless they are particularly bad
or surprising.  (The Ministry of Health in my country estimates 90% of
cases are never reported.)  Remembering to check for possible bias
from unreported cases is a human problem for analysts.  Which, if any,
R packages have proven useful to detecting the existence of a
systematic under-reporting problem might well be an appropriate topic
for this list.

On Thu, 25 Jul 2024 at 02:44, Bert Gunter <bgunter.4567 at gmail.com>
wrote:
>
> Again, this is off topic, not about statistics or R, but I think of
> interest to many on this list. The title is:
>
> "So you got a null result. Will anyone publish it?"
>
> https://www.nature.com/articles/d41586-024-02383-9
>
> Best to all,
> Bert
>
> ______________________________________________
> R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.