[Rockhounds] Inferring meaning where there is none by geologists

[J Bryan Kramer]

Ideally journals should make researcher register their hypothesis
before starting. This would prevent F stat games.

BK

“The problem isn’t that Johnny can’t read. The problem isn’t even that
Johnny can’t think. The problem is that Johnny doesn’t know what
thinking is; he confuses it with feeling.”

Thomas Sowell

J Bryan Krämer       North Florida, USA
photos at: http://pbase.com/photoburner

On Sat, Jun 1, 2019 at 9:20 PM Peter Richards <rpr at heidelberg.edu> wrote:
>
> The secret of research is knowing when to stop gathering data…. Of course, it may not be good research.
>
> Pete
> ___________________________
> R. Peter Richards, Ph.D.
> rpr at heidelberg.edu
> Morphological Crystallographer
>
> > On Jun 1, 2019, at 8:04 PM, Tim Fisher <nospam at orerockon.com> wrote:
> >
> > Welcome to the wonderful world of statistics. We deal with that constantly
> > and yes it's pretty tough to put 1000 hours or so into an analysis only to
> > fail to support your own hypothesis. And the journals make it worse, they
> > don't want to publish "our hypothesis wasn't proven by the evidence and it's
> > probably random chance" papers. They don't get cited as often which isn't
> > good for business. We've run into that a couple times and the paper winds up
> > in the "grey literature" realm which in turn no one wants to cite because it
> > wasn't published in a refereed journal. We'll be starting an analysis next
> > month that I have a feeling will nullify a prior paper we published 10 years
> > ago, or at least significantly change or weaken our inferences. Many times a
> > little extra data pushes the models past their breaking point.
> >
> > Tim Fisher
> > Http://OreRockOn.com
> > Email nospam at orerockon.com
> >
> > -----Original Message-----
> > From: Rockhounds [mailto:rockhounds-bounces at rockhounds.drizzle.com] On
> > Behalf Of Paul
> > Sent: Saturday, June 01, 2019 2:09 PM
> > To: Rockhound List
> > Subject: [Rockhounds] Inferring meaning where there is none by geologists
> >
> > Perceived connections: Inferring meaning where there is none
> > https://www.geolsoc.org.uk/Geoscientist/Archive/Dec-2018/Feature-3
> >
> > Meaningless data are tough words to swallow. John Armitage and Tom Coulthard
> > argue that Earth scientists must face up to the fact that some observations
> > might be an aggregation of seemingly random events, where there is no cause
> > and effect.
> >
> > Papers
> >
> > Armitage, J. & Coulthard, T., Perceived connections: Inferring meaning where
> > there is none. Geoscientist 28 (11), 18-21, 2018
> > https://doi.org/10.1144/geosci2018-030
> > https://www.geolsoc.org.uk/~/media/shared/documents/Geoscientist/2018/Decemb
> > er%202018/F3_Dec%202018.pdf?la=en
> >
> > Falk, R. and Konold, C., 1997. Making sense of randomness: Implicit encoding
> > as a basis for judgment. Psychological Review, 104(2), p.301-318
> > https://www.srri.umass.edu/publications/falk-1997msr/
> > https://www.srri.umass.edu/sites/srri/files/FalkKonold1997/index.pdf
> >
> > A related editorial well worth finding and reading is:
> >
> > Wright, V.P., 2019. Memes, False News, and the Death of Empiricism.
> > Journal of Sedimentary Research, 89(4), pp.310-311.
> > https://pubs.geoscienceworld.org/sepm/jsedres/article-abstract/89/4/310/5700
> > 50
> >
> > Yours,
> >
> > Paul H.
> >
> >
> >
> > _______________________________________________
> > Rockhounds mailing list
> > Subscription Services:
> > http://rockhounds.drizzle.com/mailman/listinfo/rockhounds_rockhounds.drizzle
> > .com
> > List Usage Policy: http://Tomaszewski.net/Kreigh/Rockhounds/Rockhounds.shtml
> >
> >
> > _______________________________________________
> > Rockhounds mailing list
> > Subscription Services:  http://rockhounds.drizzle.com/mailman/listinfo/rockhounds_rockhounds.drizzle.com
> > List Usage Policy: http://Tomaszewski.net/Kreigh/Rockhounds/Rockhounds.shtml
>
> _______________________________________________
> Rockhounds mailing list
> Subscription Services:  http://rockhounds.drizzle.com/mailman/listinfo/rockhounds_rockhounds.drizzle.com
> List Usage Policy: http://Tomaszewski.net/Kreigh/Rockhounds/Rockhounds.shtml