Rob Forsyth
2020-Jun-07 10:59 UTC
[R] Using item difficulties from a fitted Partial Credit Model to predict person abilities in an extended dataset?
I am using the eRm package to examine the properties of a clinical rating scale using a Partial Credit Model (PCM). I understand how to extract the person ability estimates (thetas) from a simple fitted PCM but I have a dataset with repeated observations over time (~1200 observations of the instrument in ~250 individuals). So as not to violate assumptions of conditional independence I've fitted the PCM to single observations drawn at random from each subject. This works but I would now like to use the item diffculty estimates from the fitted PCM to generate person-ability estimates for the remaining ~950 observations (at other timepoints) not used to fit the model; and I can't see from the eRm package documentation how to do this? Advice very much appreciated Rob
Bert Gunter
2020-Jun-07 15:15 UTC
[R] Using item difficulties from a fitted Partial Credit Model to predict person abilities in an extended dataset?
Such package/methodology specific questions may well go unanswered here. They are essentially offtopic anyway: this list is about general R programming questions and cannot be expected to support the ~ 20000 packages now in the ecosystem. I suggest that you contact the package maintainer (?maintainer) for help or to find out what support resources may be available. Bert Gunter "The trouble with having an open mind is that people keep coming along and sticking things into it." -- Opus (aka Berkeley Breathed in his "Bloom County" comic strip ) On Sun, Jun 7, 2020 at 3:59 AM Rob Forsyth <rob.forsyth at newcastle.ac.uk> wrote:> I am using the eRm package to examine the properties of a clinical rating > scale using a Partial Credit Model (PCM). I understand how to extract the > person ability estimates (thetas) from a simple fitted PCM but I have a > dataset with repeated observations over time (~1200 observations of the > instrument in ~250 individuals). So as not to violate assumptions of > conditional independence I've fitted the PCM to single observations drawn > at random from each subject. This works but I would now like to use the > item diffculty estimates from the fitted PCM to generate person-ability > estimates for the remaining ~950 observations (at other timepoints) not > used to fit the model; and I can't see from the eRm package documentation > how to do this? > > Advice very much appreciated > Rob > ______________________________________________ > R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide > http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code. >[[alternative HTML version deleted]]
Rob Forsyth
2020-Jun-07 15:31 UTC
[R] Using item difficulties from a fitted Partial Credit Model to predict person abilities in an extended dataset?
OK thanks for the guidance Rob> On 7 Jun 2020, at 16:15, Bert Gunter <bgunter.4567 at gmail.com> wrote: > > ? External sender. Take care when opening links or attachments. Do not provide your login details. > Such package/methodology specific questions may well go unanswered here. They are essentially offtopic anyway: this list is about general R programming questions and cannot be expected to support the ~ 20000 packages now in the ecosystem. I suggest that you contact the package maintainer (?maintainer) for help or to find out what support resources may be available. > > Bert Gunter > > "The trouble with having an open mind is that people keep coming along and sticking things into it." > -- Opus (aka Berkeley Breathed in his "Bloom County" comic strip ) > > > On Sun, Jun 7, 2020 at 3:59 AM Rob Forsyth <rob.forsyth at newcastle.ac.uk> wrote: > I am using the eRm package to examine the properties of a clinical rating scale using a Partial Credit Model (PCM). I understand how to extract the person ability estimates (thetas) from a simple fitted PCM but I have a dataset with repeated observations over time (~1200 observations of the instrument in ~250 individuals). So as not to violate assumptions of conditional independence I've fitted the PCM to single observations drawn at random from each subject. This works but I would now like to use the item diffculty estimates from the fitted PCM to generate person-ability estimates for the remaining ~950 observations (at other timepoints) not used to fit the model; and I can't see from the eRm package documentation how to do this? > > Advice very much appreciated > Rob > ______________________________________________ > R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see > https://stat.ethz.ch/mailman/listinfo/r-help > PLEASE do read the posting guide http://www.R-project.org/posting-guide.html > and provide commented, minimal, self-contained, reproducible code.