Shravan Vasishth sends this along:

Yup. Not always, though. Even though the above behavior is rewarded.

Posted by Andrew on 15 July 2018, 9:34 am

Shravan Vasishth sends this along:

Yup. Not always, though. Even though the above behavior is rewarded.

## Recent Comments

- Carlos Ungil on Thomas Basbøll will like this post (analogy between common—indeed, inevitable—mistakes in drawing, and inevitable mistakes in statistical reasoning).
- Martha (Smith) on Thomas Basbøll will like this post (analogy between common—indeed, inevitable—mistakes in drawing, and inevitable mistakes in statistical reasoning).
- Phil on Thomas Basbøll will like this post (analogy between common—indeed, inevitable—mistakes in drawing, and inevitable mistakes in statistical reasoning).
- Martha (Smith) on Thomas Basbøll will like this post (analogy between common—indeed, inevitable—mistakes in drawing, and inevitable mistakes in statistical reasoning).
- jim on Thomas Basbøll will like this post (analogy between common—indeed, inevitable—mistakes in drawing, and inevitable mistakes in statistical reasoning).
- jim on Thomas Basbøll will like this post (analogy between common—indeed, inevitable—mistakes in drawing, and inevitable mistakes in statistical reasoning).
- Tom Passin on Thomas Basbøll will like this post (analogy between common—indeed, inevitable—mistakes in drawing, and inevitable mistakes in statistical reasoning).
- Daniel Lakeland on Thomas Basbøll will like this post (analogy between common—indeed, inevitable—mistakes in drawing, and inevitable mistakes in statistical reasoning).
- Raghuveer Parthasarathy on Thomas Basbøll will like this post (analogy between common—indeed, inevitable—mistakes in drawing, and inevitable mistakes in statistical reasoning).
- Martha (Smith) on Thomas Basbøll will like this post (analogy between common—indeed, inevitable—mistakes in drawing, and inevitable mistakes in statistical reasoning).
- Andrew on Thomas Basbøll will like this post (analogy between common—indeed, inevitable—mistakes in drawing, and inevitable mistakes in statistical reasoning).
- jim on Thomas Basbøll will like this post (analogy between common—indeed, inevitable—mistakes in drawing, and inevitable mistakes in statistical reasoning).
- Daniel Lakeland on Thomas Basbøll will like this post (analogy between common—indeed, inevitable—mistakes in drawing, and inevitable mistakes in statistical reasoning).
- Martha (Smith) on Know your data, recode missing data codes
- Andrew on Thomas Basbøll will like this post (analogy between common—indeed, inevitable—mistakes in drawing, and inevitable mistakes in statistical reasoning).
- Andrew on Somethings do not seem to spread easily – the role of simulation in statistical practice and perhaps theory.
- Martha (Smith) on Somethings do not seem to spread easily – the role of simulation in statistical practice and perhaps theory.
- Martha (Smith) on Himmicanes again

## Categories

I’d say it is like this, at least hopefully…

http://www.statisticool.com/cis.jpg

Justin

Justin, this is great! I have The Course of Science on my office door, as a daily reminder when I start work. I will add this plot too. I now understand why Sabine Hossenfelder wrote in Lost in Math that physicists like the number 1 and they don’t like numbers smaller or larger than 1.

I think the joke is on you since that plot is wrong. The Newtonian value is exactly half the GR value.

https://en.wikipedia.org/wiki/Tests_of_general_relativity

Are you sure the plot is wrong? A quick google seems to indicate that gamma is a parameter that enters into the deflection equation as 0.5*(gamma + 1)…

E.g., eq. 1 here:

http://iopscience.iop.org/article/10.1088/0264-9381/32/12/124001/ampdf

Maybe the plot title is a bit misleading without the accompanying text.

You would seem to be correct. Gamma isnt “the deflection of light around the sun”, its a parameter in an approximation thats apparently zero for the Newtonian model and one in GR…

https://en.wikipedia.org/wiki/Parameterized_post-Newtonian_formalism

And a new thing to me is the amount of curvature caused by mass is supposed to depend on the speed of gravity? Ie if gravity was instantaneous there would be no curvature, according to this.

But if there were no curvature, wouldn’t there then be no gravity since that is supposedly due to the curvature? I’m going to have to assume that wikipedia page is BS and there is something more going on.

I got the graph from Statistical Sleuth, by Ramsey and Schafer from when I was in grad. school. I’d have to dig it out to see the accompanying text.

Justin

I easily found Ramsey and Schafer on my bookshelf, but didn’t find the figure until I tired “Einstein” in the index. The figure is on p. 48 (2002 edition). The accompanying text is:

“The parameter gamma, which is predicted by Einstein’s general theory of relativity to be 1 and by Newonian physics to be 0, was estimated in 1919 by British astronomers during a toal exlipes. Since then, measurements have been repeated many times, under various measurement conditions. The efforts are summarized in Display 2.13 (Date from C.M.Will, “General Relativeity at 75: How Right Was Einstin?” Science 250 (November 9, 1990): 770 -75.

Oops — sorry for any miscopying; the above is what appears before the display; I managed to submit before finishing. The discussion after the display is:

“The confidence intervals around the estimates in Display 2.13 reflect uncertainty due to measurement errors. (The actual confidence levels are not given and are not important for this illustration.) After the first relatively crude attempts to measure gamma, little improvement was made until the late 1960’s and the discovery of quasars. Measurements of light from quasar groups passing near the sun led to dramatic improvement in accuracy, as evident in the narrower intervals with later years.

“

Then there is a heading “A Note About the Cumulation of Evidence”, followed by

“This example shows that theories must withstand continual challenges from skeptical scientists. The essence of scientific theory is the ability to predict future outcomes. Experimental results are typically uncertain. So the fact that some intervals fail to include the value, gamma = 1, is not taken to disprove general relativity, but neither would it prove general relativity right if all the intervals did include gamma = 1. When a theory’s predictions are consistently denied by a series of experiments — such as the Newtonian prediction of gamma = 0 in this example — scientists agree that the theory is not adequate.

There’s a typo…the arrows from “Management Directives” to “Theoretical Misunderstandin,” “Confusion,” and “Coverup” are missing.

I believe so, “Management Directives” always have an effect – exactly opposite of that intended.

Not sure how to show that on a graph.

Looks like my work, yes

Thanks for typing all that up Martha!

I’m personally a fan of these two flowcharts myself:

https://electroncafe.files.wordpress.com/2011/05/sciencerage.png

http://phdcomics.com/comics/archive.php?comicid=761