Message-ID: <85D3E337-3E39-4542-A819-12BC8E20ACE6@ed.ac.uk>
Date: 2014-04-08T21:26:33Z
From: iwhite@staffmail.ed.ac.uk
Subject: Including Nested Factor in Repeated Measures Logistic Regression
In-Reply-To: <CAGmtyttK2uH2J3mUQYpF=+T2sCTQRuRLOjpBgsZVn3owzTkb_A@mail.gmail.com>
How many trial types are there? How many blocks? Does each subject do each combination of trial type
and block several times?
On 8 Apr 2014, at 19:39, Maxwell Elliott <maxwe128 at umn.edu> wrote:
> I have still been unable to solve this issue. Any help related to dealing
> with nested variables in lmer regression would be uch appreciated.
>
> -Max Elliott
>
>
> On Tue, Apr 1, 2014 at 2:14 PM, Maxwell Elliott <maxwe128 at umn.edu> wrote:
>
>> Hello,
>>
>> I am attempting to model accuracy in a task that contains multiple trial
>> types, multiple blocks and many subjects. It is a repeated measure design
>> because each subject completes 160 trials. My current formula for the
>> regression is this:
>>
>> Accuracy ~ TrialType * Block + (1 | subjectNumber)
>>
>> I am now trying to include a symptom measure in the design. This measure
>> will be the one continuous value for each subject. So the symptom measure
>> is nested within subject. What is the syntax for including this symptom
>> measure in the design. I want to know if the symptom measure contributes to
>> model fit and if higher values increase or decrease the likelihood of
>> answering the the trial correctly. I know I cannot do this:
>> Accuracy ~ TrialType * Block * Symptom + (1 | subjectNumber)
>> , because symptom is nested within subject but I am not sure what else to
>> do.
>>
>> Thanks for the help,
>>
>> Max Elliott
>>
>
>
>
> --
>
> -Max
>
> [[alternative HTML version deleted]]
>
> _______________________________________________
> R-sig-mixed-models at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-sig-mixed-models
>
--
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.