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

Investigations of differential item ...

University of California, Berkeley.

Linked to FindBook

Google Book

Amazon

博客來

Investigations of differential item functioning: Comparisons among approaches, and extension to a *multidimensional context /

Record Type:	Electronic resources : Monograph/item
Title/Author:	Investigations of differential item functioning: Comparisons among approaches, and extension to a *multidimensional context // Insu Paek.
Author:	Paek, Insu,
Description:	1 electronic resource (201 pages)
Notes:	Source: Dissertations Abstracts International, Volume: 65-01, Section: A.
Contained By:	Dissertations Abstracts International65-01A.
Subject:	Educational evaluation. -
Online resource:	https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3082356
ISBN:	9780496302079

Investigations of differential item functioning: Comparisons among approaches, and extension to a *multidimensional context /
Paek, Insu,

Investigations of differential item functioning: Comparisons among approaches, and extension to a *multidimensional context /Insu Paek. - 1 electronic resource (201 pages)

Source: Dissertations Abstracts International, Volume: 65-01, Section: A.

Differential item functioning (DIF) has been one of the major topics in the measurement field for over the last thirty years. As a result, research on DIF-detecting methodology has proliferated. Built on the accumulated work in DIF, a unidimensional parametric item response theory (IRT) DIF method based on multidimensional random coefficient multinomial logit (MRCML) model (Adams, Wilson, & Wang, 1997) and Moore's (1996) work is suggested. It is capable of estimating a DIF index and its standard error directly from the data, which is a unique feature of this DIF modeling compared other proposed parametric IRT DIF modeling. Its performance is compared with the simultaneous bias test (SIBTEST) model (Shealy and Stout, 1993a) by simulation studies. SIBTEST is a recently developed DIF method that is a non-parametric IRT DIF model. Since there is no prior work on which type of approach for detection of DIF is better, this comparison study to reveal the strength and weakness of different methods in various situations would be of help to theory and practice of DIF detection methodology. The controlled variables that are given prime roles to simulate data for comparison are item response function (IRF) conditions and population distribution conditions. The foci of the comparisons are the behavior of DIF estimators and significance test results such as Type I error power. For the Type II error and power investigation, (asymptotic) z-test and likelihood ratio test are used in the suggested parametric method. Using the flexibility of the MRCML model, a multidimensional extension of the unidimensional parametric DIF modeling is attempted and its performance is investigated (in rather limited conditions compared to the unidimensional study) by a simulation study. And the importance of using a multidimensional DIF model in DIF investigation for a multidimensional test is shown by a simulation study as well.

English

ISBN: 9780496302079Subjects--Topical Terms:

526425
Educational evaluation.
Subjects--Index Terms:

Differential item functioning

Investigations of differential item functioning: Comparisons among approaches, and extension to a *multidimensional context /
LDR:03326nmm a22004093i 4500 001 2430929
005 20250605050543.5
006 m o d
007 cr|nu||||||||
008 261011s2002 miu||||||m |||||||eng d
020 $a 9780496302079
035 $a (MiAaPQD)AAI3082356
035 $a AAI3082356
040 $a MiAaPQD $b eng $c MiAaPQD $e rda
100 1 $a Paek, Insu, $e author. $3 3819496
245 1 0 $a Investigations of differential item functioning: Comparisons among approaches, and extension to a *multidimensional context / $c Insu Paek.
264 1 $a Ann Arbor : $b ProQuest Dissertations & Theses, $c 2002
300 $a 1 electronic resource (201 pages)
336 $a text $b txt $2 rdacontent
337 $a computer $b c $2 rdamedia
338 $a online resource $b cr $2 rdacarrier
500 $a Source: Dissertations Abstracts International, Volume: 65-01, Section: A.
500 $a Publisher info.: Dissertation/Thesis.
500 $a Advisors: Wilson, Mark.
502 $b Ph.D. $c University of California, Berkeley $d 2002.
520 $a Differential item functioning (DIF) has been one of the major topics in the measurement field for over the last thirty years. As a result, research on DIF-detecting methodology has proliferated. Built on the accumulated work in DIF, a unidimensional parametric item response theory (IRT) DIF method based on multidimensional random coefficient multinomial logit (MRCML) model (Adams, Wilson, & Wang, 1997) and Moore's (1996) work is suggested. It is capable of estimating a DIF index and its standard error directly from the data, which is a unique feature of this DIF modeling compared other proposed parametric IRT DIF modeling. Its performance is compared with the simultaneous bias test (SIBTEST) model (Shealy and Stout, 1993a) by simulation studies. SIBTEST is a recently developed DIF method that is a non-parametric IRT DIF model. Since there is no prior work on which type of approach for detection of DIF is better, this comparison study to reveal the strength and weakness of different methods in various situations would be of help to theory and practice of DIF detection methodology. The controlled variables that are given prime roles to simulate data for comparison are item response function (IRF) conditions and population distribution conditions. The foci of the comparisons are the behavior of DIF estimators and significance test results such as Type I error power. For the Type II error and power investigation, (asymptotic) z-test and likelihood ratio test are used in the suggested parametric method. Using the flexibility of the MRCML model, a multidimensional extension of the unidimensional parametric DIF modeling is attempted and its performance is investigated (in rather limited conditions compared to the unidimensional study) by a simulation study. And the importance of using a multidimensional DIF model in DIF investigation for a multidimensional test is shown by a simulation study as well.
546 $a English
590 $a School code: 0028
650 4 $a Educational evaluation. $3 526425
650 4 $a Educational tests & measurements. $3 3168483
653 $a Differential item functioning
653 $a Multidimensional
653 $a Type I error
690 $a 0288
690 $a 0443
710 2 $a University of California, Berkeley. $e degree granting institution. $3 3765948
720 1 $a Wilson, Mark $e degree supervisor.
773 0 $t Dissertations Abstracts International $g 65-01A.
790 $a 0028
791 $a Ph.D.
792 $a 2002
856 4 0 $u https://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3082356