Our [ @andrea_e_martin] paper is OUT‼️

We present our path model of science. And how computational modelling forces us to confront our intuitions that remain unexamined — over and above stewardship of experimental practice (e.g., preregistration).

🧵1/n

https://doi.org/10.1177/1745691620970585
Firstly, just a quick link to the thread from the preprint, which is useful and which also demonstrates how nicely the journal (Perspectives on Psychological Science, @PsychScience) redid our figures:

https://twitter.com/o_guest/status/1230144174751698946

2/n
Also a link to the original tweet that started it ALL, at least with respect to the pizza problem: https://twitter.com/o_guest/status/1186141920239730689

If only I knew which 1 in a 100k tweets of mine are good! I guess that's why I have @andrea_e_martin who was like "yes, this must be in our paper!" 🤣

3/n
Notwithstanding of course, just as many thanks and much credit should go to @richarddmorey for inviting me/us to write this and to uh, myself [LMAO — sorry, could not resist] for knowing @andrea_e_martin would be the perfect the co-author [for this theme and in general TBH].

4/n
Before I jump in with some points from the paper...

In case useful, here are some talks [in 60, 45, and 15 min versions] on this paper: https://www.youtube.com/playlist?list=PLYg0wn9U8Q7-VHAb5Kff46bty9S2PI8qO

For ease of access, the 15 minute version is here:

5/n
Now, I'll outline some MAIN points of the paper, which might be like 2-3 papers in one. So you're getting quite a bang for your buck here, OK!? Shhhhhh. Don't question the process. ✌️🏻

It's what happens, I guess, when one stews ideas for a decade. Who knows! 🌈

Anyway...

6/n
Firstly, we explain what a computational model is, how it is built, and how formalization is FORCED at the steps along the way. We illustrate how specifying a model naturally results in better specified theories and therefore in better #openscience.

https://doi.org/10.1177/1745691620970585
7/n
Next, we present our path model of science and how to improve the relationship between theory and data. Stepping through theory, specification, & implementation is required before an interpretation can have explanatory force in relation to a theory.

https://doi.org/10.1177/1745691620970585
8/n
Our model allows us to pinpoint where in the path questionable research occurs, such as p-hacking — something many struggle with because not everything that is post hoc is "bad". Theory is (also) post hoc!

BONUS: Doing it "right" upsets Nazis.

https://doi.org/10.1177/1745691620970585
9/n
Finally, computational modelling addresses the lacking theory building that underlies the so-called replication crisis in, e.g., social psychology — and the whole field in general too!

Modelling can be achieved by anyone with benefit to all.

https://doi.org/10.1177/1745691620970585

10/n
I dunno what else I might add, but PLEASE FEEL FREE TO ASK ME ANYTHING!!! 💜🌼✌️🏻

Including here: https://twitter.com/o_guest/status/1322193887725785089

11/n
Also be nice... if you like these ideas, cite me and @andrea_e_martin.

You know who you are... From my/our end, it's not personal.

I just wanna be cited and want you to not be doing academic misconduct. Or something.

https://twitter.com/o_guest/status/1322242744551657472

12/n
You can follow @o_guest.
Tip: mention @twtextapp on a Twitter thread with the keyword “unroll” to get a link to it.

Latest Threads Unrolled:

By continuing to use the site, you are consenting to the use of cookies as explained in our Cookie Policy to improve your experience.