But publishing is essentially a form of pre-registration. Once published, it's a permanent record of the model's particular instantiation. No HARKing possible.
-
-
Replying to @bradpwyble @IrisVanRooij and
HARking occurs during model building. E.g. people change their hypotheses about model mechanisms and do incremental adjustments until they're happy with the model fit to the data.
1 reply 0 retweets 2 likes -
Replying to @GunnarBlohm @IrisVanRooij and
That's just model building. All models are built based on data. How else would you do it?
2 replies 0 retweets 4 likes -
Replying to @bradpwyble @IrisVanRooij and
How is this different from trying 5 different analysis methods for experimental data and submitting the best result only?
3 replies 0 retweets 1 like -
Replying to @GunnarBlohm @IrisVanRooij and
Because building the model isn't the final step, it's the first step. The real work then comes in future publications as you test the pre-published predictions.
1 reply 0 retweets 2 likes -
Replying to @bradpwyble @IrisVanRooij and
I see your point. So you'd advise against pre-registration of modeling projects?
1 reply 0 retweets 2 likes -
Replying to @GunnarBlohm @IrisVanRooij and
I guess my view is that publishing a paper already accomplishes pre-registration, so we don't really need a new mechanism. For faster turn-around do a registration on OSF. Upload the model code, hit registration, and that version of the model becomes date-stamped and immutable.
1 reply 0 retweets 5 likes -
Replying to @bradpwyble @IrisVanRooij and
Yes I understand your point re experimental hypothesis. But building a model that describes a phenomenon is essentially making hypotheses about specific mechanisms that could achieve this description. And thus my question: should this be pre-registered and if so how
1 reply 0 retweets 1 like -
Replying to @GunnarBlohm @IrisVanRooij and
I'm a little confused. This is accomplished by publishing a paper that describes the model, including code, parameters. Then it becomes clear that you had already designed the model on date X and all further modifications would be on a different version of the model.
1 reply 0 retweets 0 likes -
Replying to @bradpwyble @GunnarBlohm and
Though we may have different definitions of pre-registration, or model building.
1 reply 0 retweets 1 like
Woah! I'm in the middle of coding and I "wake up" to 45 notiffffffs! Will come back soon to this as I do not want to lose my working memory. But just quickly: for me modelling and prereg make no sense. We have crossval and other tools/methods for model eval. Back soon though. 
-
-
Replying to @o_guest @bradpwyble and
Love your description of this as "waking up" - that's exactly how I feel when I stop after coding for a while!
0 replies 0 retweets 3 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.