An IRT Scale Transformation Method for Parameters Calibrated from Multiple Samples of Subjects. ACT Research Report Series 96-2 [electronic resource] / Lingjia Zeng.

A problem frequently confronted in item response theory (IRT) applications is that the item parameters calibrated using more than two independent samples of subjects must be expressed on the same scale. The existing methods were developed for a pairwise transformation, that is, from one scale to the...

Full description

Saved in:
Bibliographic Details
Online Access: Full Text (via ERIC)
Main Author: Zeng, Lingjia
Corporate Author: American College Testing Program
Format: Electronic eBook
Language:English
Published: [S.l.] : Distributed by ERIC Clearinghouse, 1996.
Subjects:

MARC

LEADER 00000nam a22000002u 4500
001 b6404772
003 CoU
005 20080220152133.9
006 m d f
007 cr un
008 960501s1996 xx |||| ot ||| | eng d
035 |a (ERIC)ed410268 
040 |a ericd  |c ericd  |d MvI 
099 |f ERIC DOC #  |a ED410268 
099 |f ERIC DOC #  |a ED410268 
100 1 |a Zeng, Lingjia. 
245 1 3 |a An IRT Scale Transformation Method for Parameters Calibrated from Multiple Samples of Subjects. ACT Research Report Series 96-2  |h [electronic resource] /  |c Lingjia Zeng. 
260 |a [S.l.] :  |b Distributed by ERIC Clearinghouse,  |c 1996. 
300 |a 19 p. 
500 |a ERIC Document Number: ED410268. 
500 |a Availability: ACT Research Report Series, P.O. Box 168, Iowa City, IA 52243-0168.  |5 ericd. 
520 |a A problem frequently confronted in item response theory (IRT) applications is that the item parameters calibrated using more than two independent samples of subjects must be expressed on the same scale. The existing methods were developed for a pairwise transformation, that is, from one scale to the other. The purpose of this study is to introduce a common scale transformation method that can simultaneously find a vector of transformation functions for placing the parameter estimated from two or more item pools on a specified common scale. Two examples are presented to illustrate the usefulness of the method. The first is from 2 forms of a mathematics test administered to nonequivalent groups of examinees (1,637 and 1,636 examinees). The second used the item parameter estimates of the 11 common items for the first item pool of the first study. The proposed common scale transformation method provides an alternative approach for solving the problem of incompatible IRT-parameter estimates calibrated with two or more independent samples of subjects. (Contains 2 tables and 11 references.) (Author/SLD) 
650 1 7 |a Estimation (Mathematics)  |2 ericd. 
650 1 7 |a Item Response Theory.  |2 ericd. 
650 0 7 |a Mathematics Tests.  |2 ericd. 
650 1 7 |a Scaling.  |2 ericd. 
650 1 7 |a Test Items.  |2 ericd. 
710 2 |a American College Testing Program. 
856 4 0 |u http://files.eric.ed.gov/fulltext/ED410268.pdf  |z Full Text (via ERIC) 
907 |a .b64047726  |b 07-06-22  |c 10-16-10 
998 |a web  |b 10-26-12  |c f  |d m   |e -  |f eng  |g xx   |h 3  |i 1 
956 |a ERIC 
999 f f |i 3220bcb7-cd7e-5a65-afcd-f24be034d9ac  |s 90b2b51a-4851-5534-be35-fe035cb109ed 
952 f f |p Can circulate  |a University of Colorado Boulder  |b Online  |c Online  |d Online  |e ED410268  |h Other scheme  |i web  |n 1