Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: RE: Probability distribution for a runs test of sequentially administered items


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: Probability distribution for a runs test of sequentially administered items
Date   Thu, 4 Mar 2010 22:48:19 -0000

I don't think the second has greater stigma or less charisma than the
first. The problem has some similarities with that tackled by 

SJ-6-4  st0044_1  . . . . . . . . . . . . Software update for nruns and
nrunsi
        (help nruns, nrunsi if installed) . . . . . . N. Smeeton and N.
J. Cox
        Q4/06   SJ 6(4):597
        introduces commands nruns and nrunsi for analyzing
        sequences stored in a variable or supplied on the fly

SJ-3-3  st0044  Do-it-yourself shuffling & the number of runs under
randomness
        . . . . . . . . . . . . . . . . . . . . . . . N. Smeeton and N.
J. Cox
        Q3/03   SJ 3(3):270--277                                 (no
commands)
        explains estimation of the probability of an event under
        randomness; illustrates a simple method of random shuffling;
        and shows derivation of conditional probability distribution
        of the number of runs

so you might find that a useful start. Both paper and code are freely
available. 

Nick 
n.j.cox@durham.ac.uk 


-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu
[mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Barth Riley
Sent: 04 March 2010 22:43
To: statalist@hsphsun2.harvard.edu
Subject: st: Probability distribution for a runs test of sequentially
administered items

Dear Statalist

I am exploring the possibility of using a Runs test to assess whether a
set
of sequentially administered item responses (i.e., as in a computerized
adaptive test) is "in process" (with alternating correct/incorrect
responses
towards the end of the test) or out of process (reflecting aberrant
responding, such as guessing or specialized knowledge). To do this, I
would
like to apply a probability distribution for the run length (number of
consecutive pass/yes or fail/no responses) one would expect at a given
point
during the testing given that the respondent were answering the items as
expected (in process). I was wondering if such a probability
distribution
already exists (and if so, what is it?) or if I would have to derive it
via
simulation.

*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index