Increase in Precision when Estimating Parameters in Computer Assisted Testing Using Response Time

Increase in Precision when Estimating Parameters in Computer Assisted Testing Using Response Time A model is given in this article that associates response time to each item with its relative difficulty for the person examined. The properties of the model are presented with respect to the quantity of information, while deducing the parameter estimators. Likewise, an empirical study to prove the feasibility and plausibility of the model is carried out, feasibility being understood as the existence of adequate estimators for the parameters of the model and plausibility the adjustment of the model proposed to the items in a spatial aptitude test used to demonstrate this (“Rotation of solid figures”). For this, the model is tested with simulated data, comparing the behaviour of the estimates carried out on the model proposed with the estimates made by a programme such as BILOG-MG. It is shown that, with the exception of parameter c (pseudo-chance) which does not benefit from the inclusion of the response time, the estimates for the model are always higher for parameters a (discrimination) and b (difficulty) of the items, always giving rise to a higher correlation with the original data, in spite of the inefficiency of the computation algorithm employed. The results suggest that the use of response time supplies as much information or more than the use of very informative a priori distributions. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Quality & Quantity Springer Journals

Increase in Precision when Estimating Parameters in Computer Assisted Testing Using Response Time

Loading next page...
 
/lp/springer_journal/increase-in-precision-when-estimating-parameters-in-computer-assisted-CwhJZ4dHJD
Publisher
Kluwer Academic Publishers
Copyright
Copyright © 2005 by Springer
Subject
Social Sciences; Methodology of the Social Sciences; Social Sciences, general
ISSN
0033-5177
eISSN
1573-7845
D.O.I.
10.1007/s11135-004-0437-y
Publisher site
See Article on Publisher Site

Abstract

A model is given in this article that associates response time to each item with its relative difficulty for the person examined. The properties of the model are presented with respect to the quantity of information, while deducing the parameter estimators. Likewise, an empirical study to prove the feasibility and plausibility of the model is carried out, feasibility being understood as the existence of adequate estimators for the parameters of the model and plausibility the adjustment of the model proposed to the items in a spatial aptitude test used to demonstrate this (“Rotation of solid figures”). For this, the model is tested with simulated data, comparing the behaviour of the estimates carried out on the model proposed with the estimates made by a programme such as BILOG-MG. It is shown that, with the exception of parameter c (pseudo-chance) which does not benefit from the inclusion of the response time, the estimates for the model are always higher for parameters a (discrimination) and b (difficulty) of the items, always giving rise to a higher correlation with the original data, in spite of the inefficiency of the computation algorithm employed. The results suggest that the use of response time supplies as much information or more than the use of very informative a priori distributions.

Journal

Quality & QuantitySpringer Journals

Published: May 28, 2004

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off