I have been following these debates, and I am not sure what is meant by "modeler". What is a "modeler." Is it someone that fits custom models ?
-
-
Replying to @wdonald_1985 @zerdeve and
I sometimes fit models that take weeks of tinkering to even get the model to converge. Is this modeling ?
1 reply 0 retweets 0 likes -
Replying to @wdonald_1985 @EJWagenmakers and
I don’t know if the definition matters all that much for my comment. You could replace it with researchers or just people and my question would be the same. But I think we all do different kinds of modeling so there isn’t a specific mold if that’s what you meant to ask.
1 reply 0 retweets 2 likes -
Replying to @zerdeve @wdonald_1985 and1 reply 0 retweets 3 likes
-
Indeed. Before coming to psychology I was in biology. When someone was a "modeler" this almost NEVER meant they used *statistical* models. Rather, they were mathematical models to, for example, study animal response and survival from environmental change. Basically, to understand
1 reply 0 retweets 1 like -
Replying to @wdonald_1985 @o_guest and
the logic of a system and to provide some interesting ways to think of things for empirical researchers. On the other hand, now that I am in psychology, the "modeling" I do is all based on statistical models and, IMO, can almost never be used to draw strong conclusions. Rather,
1 reply 0 retweets 1 like -
Replying to @wdonald_1985 @o_guest and
I literally "trail blaze" a path from the data to the finished model, all in an effort to first get a model that converges and to also highlight what the MODEL can be used for in general. There are so (so) many data dependent decisions along the way, that I would conclude nothing
1 reply 0 retweets 1 like -
Replying to @wdonald_1985 @o_guest and
from them, which is to EJs point. However, the intention is never meant to be inferential but to find some data to show case a MODEL that the hope is others can use in their own research.
1 reply 0 retweets 1 like -
Replying to @wdonald_1985 @o_guest and
So for this kind of modeling the data is often intentionally selected (and combed over) with the goal of highlighting what the MODEL can be used for. However, I do think pre-reg would be helpful here as applied researchers often like to interpret the findings.
0 replies 0 retweets 2 likes
Yup.
-
-
Replying to @o_guest @VandekerckhoveJ and
Berna D. Retweeted Berna D.
Mechanistic models are what comes to my mind at first too but it’s also good to know modeling can vary quite a bit. This post gets to that distinction.https://twitter.com/zerdeve/status/1065651263667286017?s=21 …
Berna D. added,
Berna D. @zerdeveReplying to @IrisVanRooijThis is a nice blog post about mechanistic vs. phenomenological models with great response thread (see Florian Hartig’s lengthy response for example). It’s from ecology but very informative nonetheless. https://theartofmodelling.wordpress.com/2012/02/19/mechanistic-models-what-is-the-value-of-understanding/ …0 replies 0 retweets 4 likesThanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.