[R-sig-ME] Multilevel logistic regression guessing parameter

Fri May 12 15:51:58 CEST 2017

Hi Dominik,

I guess it depends on what you theoretically expect participants to do. From what you wrote earlier, it seemed that it was possible that participants could have a less than 50% chance of getting the question correct, despite you trying to guard against that by deleting certain stimuli, so I wondered whether the predictions from your model were actually reflective of how participants were answering. It seems potentially dangerous to delete the conditions you would expect some wrong applications of the learnt rule, rather than include those and model them explicitly.

Best
Conor

________________________________
From: Dominik Ćepulić <dcepulic at gmail.com>
Sent: Friday, May 12, 2017 1:19 PM
To: Conor Michael Goold
Cc: Paul Buerkner; r-sig-mixed-models at r-project.org
Subject: Re: [R-sig-ME] Multilevel logistic regression guessing parameter

Dear Conor and Jonathan, thanks for your replies!

Maybe I am wrong but I believe that there is a difference when applying logistic regression  when you ask an open question, without any answers given, and the answer may be right or wrong (so dependent variable is dichotomous) and
when you ask a question but offer two answers among which one of them is correct. That has to be taken into account somehow when doing logistic regression, according to my viewpoint.

Am I missing something?

On Fri, May 12, 2017 at 1:12 PM, Conor Michael Goold <conor.goold at nmbu.no<mailto:conor.goold at nmbu.no>> wrote:
Hi Dominik,

I meant to write equation 4 in appendix 1 of the hyperlinked paper.

Conor

________________________________________
From: R-sig-mixed-models <r-sig-mixed-models-bounces at r-project.org<mailto:r-sig-mixed-models-bounces at r-project.org>> on behalf of Conor Michael Goold <conor.goold at nmbu.no<mailto:conor.goold at nmbu.no>>
Sent: Friday, May 12, 2017 11:32 AM
To: Paul Buerkner; Dominik Ćepulić
Cc: r-sig-mixed-models at r-project.org<mailto:r-sig-mixed-models at r-project.org>
Subject: Re: [R-sig-ME] Multilevel logistic regression guessing parameter

Hi Dominik,

I may have misunderstood your problem, but I don't understand why you want to constrain the probability of success to be 0.5 at the lowest. If participants are choosing their answers completely randomly, their average probability of choosing the correct response (i.e. a 1 if the responses are coded 0 = incorrect and 1 = correct) across tasks may be around 0.5, but it seems completely plausible that the average probability of choosing the correct response could be between 0 and 1, and this propensity for a correct answer could vary between participants. This seems like a normal application of logistic regression. Sorry if I am missing something!

If you have reason to believe that participants do just guess sometimes, which may result in some 'outlying' data points (i.e. correct or incorrect responses where we may not expect them), as others have said, this can be included in a Bayesian model. John Kruschke has an example in his book Doing Bayesian Data Analysis (using JAGS) and also in this paper (see equation 3 in appendix 4): http://journal.sjdm.org/14/14721a/jdm14721a.html

Best regards
Conor
________________________________________
From: R-sig-mixed-models <r-sig-mixed-models-bounces at r-project.org<mailto:r-sig-mixed-models-bounces at r-project.org>> on behalf of Paul Buerkner <paul.buerkner at gmail.com<mailto:paul.buerkner at gmail.com>>
Sent: Friday, May 12, 2017 11:04 AM
To: Dominik Ćepulić
Cc: r-sig-mixed-models at r-project.org<mailto:r-sig-mixed-models at r-project.org>
Subject: Re: [R-sig-ME] Multilevel logistic regression guessing parameter

Hi Dominik,

I mean that brms uses Stan (http://mc-stan.org/) for the model fitting, but
you don't need to worry about that. I am confident that brms will allow you
to fit the model you have in mind.

Best,
Paul

2017-05-12 10:55 GMT+02:00 Dominik Ćepulić <dcepulic at gmail.com<mailto:dcepulic at gmail.com>>:

> Dear everybody, thank you for your ideas and messages!
>
> First, Philipp, yes, you are right. We have a simple two-choice
> recognition task. Participants were learning some stimuli,  and after some
> the recognition phase started. Always one stimuli per screen, and they have
> to say whether it is one of the learnt ones or not. B is therefore coded as
> response1 and response2 and afterwards coded in correct/incorrect.
> The problem that might have appeared is that some distractors may have
> been very similar to some well learnt items, and were simultaneously paired
> with a poorly learnt target. That might produce the effect of correctness
> below 0.5 We searched for such tasks and deleted them from further analysis.
>
> My problem is that when I try to plot probability functions (x - predictor
> variable, y - Accuracy from 0 to 1) for domains, they go below 0.5 which
> doesn´t make sense, as this was a two-choice task. Their lower asymptote
> should be on 0.5 not on 0. That´s why I am asking.
>
> @Paul: Thanks for recommendation, but what do you mean by "Stan under the
> hood"? I basically need a typical multilevel logistic regression (with
> random effects for 2 crossed levels) but with lower asymptote being 0.5 and
> not 0.
>
> I will take a look at the functions!
>
> Best,
> Dominik
>
> On Fri, May 12, 2017 at 9:36 AM, Paul Buerkner <paul.buerkner at gmail.com<mailto:paul.buerkner at gmail.com>>
> wrote:
>
>> Hi Dominik,
>>
>> in addition to what Jake said, you can do this with the brms package
>> (using Stan under the hood). After installing brms, you can learn how to
>> fit such models in the "brms_nonlinear" vignette: Type
>> vignette("brms_nonlinear") in R.
>>
>> Best,
>> Paul
>>
>> 2017-05-11 13:00 GMT+02:00 Dominik Ćepulić <dcepulic at gmail.com<mailto:dcepulic at gmail.com>>:
>>
>>> I  have a following situation:
>>>
>>> I want to predict variable B (which is dichotomous) from variable A
>>> (continous) controlling for random effects on the level of a) Subjects;
>>> b)
>>> Tasks.
>>>
>>> A -> B (1)
>>>
>>> The problem is that when I use model to predict the values of B from A,
>>> values below probability of 0.5 get predicted, and in my case that
>>> doesn´t
>>> make sense, because, if you guess at random, the probability of correct
>>> answer on B would be 0.5.
>>>
>>> I want to know how I can constrain the model (1) in lme4 so that it
>>> doesn´t
>>> predict values lower than 0.5 in variable B.
>>>
>>> Thank you,
>>>
>>> Dominik!
>>>
>>>         [[alternative HTML version deleted]]
>>>
>>> _______________________________________________
>>> R-sig-mixed-models at r-project.org<mailto:R-sig-mixed-models at r-project.org> mailing list
>>> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
>>
>>
>>
>

        [[alternative HTML version deleted]]

_______________________________________________
R-sig-mixed-models at r-project.org<mailto:R-sig-mixed-models at r-project.org> mailing list
https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models

_______________________________________________
R-sig-mixed-models at r-project.org<mailto:R-sig-mixed-models at r-project.org> mailing list
https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models

	[[alternative HTML version deleted]]