[Rasch] One partial-credit item or multiple dichotomous items?

Renselange rense.lange at gmail.com
Tue Mar 3 09:12:09 EST 2015


I agree with Elvind - why is the second approach being considered in the first place? 

Rense Lange

Sent from my iPhone

On Mar 2, 2015, at 15:57, Eivind Kaspersen <eivind.kaspersen at hist.no> wrote:

Hi. Wouldn't the second approach violate the local independence assumption? A score of 1 on the second item would definitely cause a score of 1 on item 1, or?

Eivind Kaspersen
HiST
________________________________________
Fra: rasch-bounces at acer.edu.au [rasch-bounces at acer.edu.au] p&#229; vegne av Stuart Luppescu [slu at ccsr.uchicago.edu]
Sendt: 2. mars 2015 22:42
Til: rasch at acer.edu.au
Emne: [Rasch] One partial-credit item or multiple dichotomous items?

Hello Fellow Raschies, I'm working on a test with some partial-credit
items. The coding is something like 0: Did not understand at all; 1:
Understood incompletely; 2: Understood completely.
I'm debating with the test developers about whether to make these
individual items scored {0, 1, 2} or two dichotomous items: item 1:
understood at least partially; item 2: understood completely. Is there
an advantage to using one or the other approach?
Also, if we go with the two-dichotomous-item approach, how should the
second item be scored if the student did not understand the question at
all? That is, does it make a difference if the two items are scored 0,
0; or 0, omitted?

TIA.
--
Stuart Luppescu
UChicagoCCSR
________________________________________
Rasch mailing list
email: Rasch at acer.edu.au
web: https://mailinglist.acer.edu.au/mailman/options/rasch/eivind.kaspersen%40hist.no
________________________________________
Rasch mailing list
email: Rasch at acer.edu.au
web: https://mailinglist.acer.edu.au/mailman/options/rasch/rense.lange%40gmail.com


More information about the Rasch mailing list