東華大學圖書館 |

Language: English

Help

回圖書館首頁

手機版館藏查詢

Back

Switch To: Labeled | MARC Mode | ISBD

Statistical tests in the DFIT framew...

University of Missouri - Columbia.

Linked to FindBook

Google Book

Amazon

博客來

Statistical tests in the DFIT framework: A Monte Carlo evaluation of conventional methods and a bootstrap alternative.

Record Type:	Language materials, printed : Monograph/item
Title/Author:	Statistical tests in the DFIT framework: A Monte Carlo evaluation of conventional methods and a bootstrap alternative./
Author:	Petroski, Gregory F.
Description:	111 p.
Notes:	Source: Dissertation Abstracts International, Volume: 66-09, Section: A, page: 3212.
Contained By:	Dissertation Abstracts International66-09A.
Subject:	Education, Educational Psychology. -
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3189946
ISBN:	9780542337215

Statistical tests in the DFIT framework: A Monte Carlo evaluation of conventional methods and a bootstrap alternative.
Petroski, Gregory F.

Statistical tests in the DFIT framework: A Monte Carlo evaluation of conventional methods and a bootstrap alternative. - 111 p.

Source: Dissertation Abstracts International, Volume: 66-09, Section: A, page: 3212.

Thesis (Ph.D.)--University of Missouri - Columbia, 2005.

Item and test bias are forms of test invalidity where unfair advantage is given to some groups of examinees. The judgment of bias requires statistical evidence of differential functioning. Differential item and test functioning (DIF and DTF) occur when individuals with the same ability but from different subgroups of the examinee population have differing probabilities of success on a test item or test. Raju, van der Linden, and Fleer (1992) proposed methods for detecting differential functioning of items and tests (DFIT) in the context of parametric item response theory. The DFIT framework includes tests of significance for DIF and DTF. The purpose of this research was to use simulation methods to study Type I error rates and power characteristics of the conventional DFIT tests for DIF and DTF and several bootstrap alternatives.

ISBN: 9780542337215Subjects--Topical Terms:

1017560
Education, Educational Psychology.

Statistical tests in the DFIT framework: A Monte Carlo evaluation of conventional methods and a bootstrap alternative.
LDR:02112nam 2200289 a 45 001 969851
005 20110920
008 110921s2005 eng d
020 $a 9780542337215
035 $a (UnM)AAI3189946
035 $a AAI3189946
040 $a UnM $c UnM
100 1 $a Petroski, Gregory F. $3 1293907
245 1 0 $a Statistical tests in the DFIT framework: A Monte Carlo evaluation of conventional methods and a bootstrap alternative.
300 $a 111 p.
500 $a Source: Dissertation Abstracts International, Volume: 66-09, Section: A, page: 3212.
500 $a Supervisor: Steven J. Osterlind.
502 $a Thesis (Ph.D.)--University of Missouri - Columbia, 2005.
520 $a Item and test bias are forms of test invalidity where unfair advantage is given to some groups of examinees. The judgment of bias requires statistical evidence of differential functioning. Differential item and test functioning (DIF and DTF) occur when individuals with the same ability but from different subgroups of the examinee population have differing probabilities of success on a test item or test. Raju, van der Linden, and Fleer (1992) proposed methods for detecting differential functioning of items and tests (DFIT) in the context of parametric item response theory. The DFIT framework includes tests of significance for DIF and DTF. The purpose of this research was to use simulation methods to study Type I error rates and power characteristics of the conventional DFIT tests for DIF and DTF and several bootstrap alternatives.
520 $a Simulation results indicate that the conventional tests have extremely high Type I error rates and their use can no longer be recommended. Bootstrap versions of the tests for DIF and DTF exhibited good control of Type I error and reasonable power when DIF items are excluded from the parameter equating step.
590 $a School code: 0133.
650 4 $a Education, Educational Psychology. $3 1017560
650 4 $a Psychology, Psychometrics. $3 1017742
690 $a 0525
690 $a 0632
710 2 0 $a University of Missouri - Columbia. $3 1017522
773 0 $t Dissertation Abstracts International $g 66-09A.
790 $a 0133
790 1 0 $a Osterlind, Steven J., $e advisor
791 $a Ph.D.
792 $a 2005
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3189946