[Rasch] Negative pt-bis and fit of 1.0? How can this be?

Bond, Trevor trevor.bond at jcu.edu.au
Wed Mar 7 10:40:07 EST 2012


Dear Gregory et alii
You wrote (inter alia)

Sending this to you and not the listserv.  (whoops)  ... regardless of whether we accept the largely arbitrary range of fit so popularized in books like Bond and Fox, or the more rational fit calculated by folks like Smith.

It’s drag being popular...or least being guilty of popularising the largely arbitrary range of fit.

Still a man hears what he wants to hear
And disregards the rest (Simon & Garfunkel)
p242
The interpretation of fit statistics, more than any other aspect of Rasch modeling,
requires experience related to the particular measurement context. Then,
“[w]hen is a mean-square too large or too small? There are no hard-and-fast rules.
Particular features of a testing situation, for example, mixing item types or off-target
testing, can produce different mean-square distributions. Nevertheless, here,
as a rule of thumb, are some reasonable ranges for item mean-square fit statistics”
(Wright, Linacre, Gustafsson, & Martin-Loff, 1994; Table 12.6). Unfortunately,
many readers will scour this text merely to cite the page to support that the fit values
they have reported are within some acceptable bounds.

p243
So while a set of general guidelines like those above will be helpful for
researchers embarking on Rasch modeling, a considerable bibliography of relevant
material exists. Articles published by Smith (e.g., 1991a, 1994, 2000) provide
good starting points for a more thoroughly informed view of issues concerning fit
statistic interpretation.

Indeed, that very chapter starts:
It would be difficult to deny the claim that the most contentious issue in Rasch
measurement circles is that of fit.

I guess we could have been more circumspect, but I don’t know how. Cos, [u]nfortunately,
many readers will scour this text merely to cite the page to support that the fit values
they have reported are within some acceptable bounds.
Still a man hears what he wants to hear
And disregards the rest (Simon & Garfunkel)

Collegial regards
TGB
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://mailinglist.acer.edu.au/pipermail/rasch/attachments/20120306/bcaaeef0/attachment.html 


More information about the Rasch mailing list