Closed state-o-flux closed 3 years ago
It is not clear what you exactly mean.
—— Dimitris Rizopoulos Professor of Biostatistics Erasmus University Medical Center The Netherlands
From: state-o-flux notifications@github.com Sent: Sunday, January 10, 2021 6:05:24 PM To: drizopoulos/ltm ltm@noreply.github.com Cc: Subscribed subscribed@noreply.github.com Subject: [drizopoulos/ltm] Applying scaling factor d (1.7) to apply the logistic item response model (#16)
Hi Dr.Rizopoulos,
I'm wondering if there is any way to apply the scaling factor d to the item response models in ltm. Specifically I'm working with gpcm as I have dichotomous and polytomous items.
— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHubhttps://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fdrizopoulos%2Fltm%2Fissues%2F16&data=04%7C01%7Cd.rizopoulos%40erasmusmc.nl%7Cb3ccfb6827074db645df08d8b589eca2%7C526638ba6af34b0fa532a1a511f4ac80%7C0%7C0%7C637458951268995864%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=LPLTSoMCqXqTpiefq3Rjv5iFg3LFMRKou1mVVB9kbh8%3D&reserved=0, or unsubscribehttps://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fnotifications%2Funsubscribe-auth%2FADE7TT5CMVUZ3VVIFT7TKSTSZHM5JANCNFSM4V4SCLYQ&data=04%7C01%7Cd.rizopoulos%40erasmusmc.nl%7Cb3ccfb6827074db645df08d8b589eca2%7C526638ba6af34b0fa532a1a511f4ac80%7C0%7C0%7C637458951269005819%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000&sdata=4LbYGroqijlKGqeLWUCGcZTYaPnS8%2BarhnxugDWvTM0%3D&reserved=0.
I've read that IRT models can scale their discrimination parameter by a factor of 1.7 in order to estimate the logistic item response model (as opposed to the normal ogive item response model). I'm wondering if there is a way to apply this to gpcm. I've noticed that the optim and nlminb functions both have parameters that can scale their optimization routines and wonder if it's possible to implement into the model writ large.
No, unfortunately, this will not be possible in gpcm().
From: state-o-flux notifications@github.com Sent: Tuesday, January 12, 2021 12:50 AM To: drizopoulos/ltm ltm@noreply.github.com Cc: D. Rizopoulos d.rizopoulos@erasmusmc.nl; Comment comment@noreply.github.com Subject: Re: [drizopoulos/ltm] Applying scaling factor d (1.7) to apply the logistic item response model (#16)
I've read that IRT models can scale their discrimination parameter by a factor of 1.7 in order to estimate the logistic item response model (as opposed to the normal ogive item response model). I'm wondering if there is a way to apply this to gpcm. I've noticed that the optim and nlminb functions both have parameters that can scale their optimization routines and wonder if it's possible to implement into the model writ large.
- You are receiving this because you commented. Reply to this email directly, view it on GitHubhttps://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fdrizopoulos%2Fltm%2Fissues%2F16%23issuecomment-758296861&data=04%7C01%7Cd.rizopoulos%40erasmusmc.nl%7Cefffb4f9b4d54a09e53a08d8b68b91f2%7C526638ba6af34b0fa532a1a511f4ac80%7C0%7C0%7C637460057850774250%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&sdata=daxBxWh7x8kIzGYfY7POWxJNjllsGHzlMFVUpB95jAw%3D&reserved=0, or unsubscribehttps://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fnotifications%2Funsubscribe-auth%2FADE7TT4ILSEGGYYDCZSGVYDSZOFBNANCNFSM4V4SCLYQ&data=04%7C01%7Cd.rizopoulos%40erasmusmc.nl%7Cefffb4f9b4d54a09e53a08d8b68b91f2%7C526638ba6af34b0fa532a1a511f4ac80%7C0%7C0%7C637460057850774250%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000&sdata=Kc2jFth8M8a%2Br%2B5L4vJ5QhCnOCEse3cedhDqfZ7Iv1I%3D&reserved=0.
Hi Dr.Rizopoulos,
I'm wondering if there is any way to apply the scaling factor d to the item response models in ltm. Specifically I'm working with gpcm as I have dichotomous and polytomous items.