MULTISIB: A procedure to investigate DIF when a test is intentionally two-dimensional

William Stout, Hsin Hung Li, Ratna Nandakumar, Daniel Bolt

Research output: Contribution to journalArticle

15 Citations (Scopus)

Abstract

MULTISIB is proposed as a statistical test for assessing differential item functioning (DIF) of intentionally two-dimensional test data, such as a mathematics test designed to measure algebra and geometry. MULTISIB is based on the multidimensional model of DIF as presented in Shealy & Stout (1993), and is a direct extension of SIBTEST, its unidimensional counterpart. For an intentionally two-dimensional test, DIF is appropriately modeled to result from secondary dimensional influence from other than the two intended dimensions. Simulation studies were used to assess the performance of MULTISIB to detect DIF in intentionally two-dimensional tests. These results indicate that MULTISIB exhibited reasonably good adherence to the nominal level of significance and good power. Moreover, for each DIF model the average amount of DIF estimated over the 100 simulations of the model by MULTISIB was close to the true value, confirming its relative lack of statistical estimation bias in assessing true DIF. In addition, the simulation studies supported the importance of using the regression correction to adjust the scores on the studied item due to impact and the importance of matching examinees on two subtest scores instead of the total test score.

Original languageEnglish
Pages (from-to)195-213
Number of pages19
JournalApplied Psychological Measurement
Volume21
Issue number3
DOIs
Publication statusPublished - 1997 Sep

Fingerprint

Mathematics
simulation
mathematics
statistical test
regression
lack
trend
performance
Values

All Science Journal Classification (ASJC) codes

  • Social Sciences (miscellaneous)
  • Psychology (miscellaneous)

Cite this

Stout, William ; Li, Hsin Hung ; Nandakumar, Ratna ; Bolt, Daniel. / MULTISIB : A procedure to investigate DIF when a test is intentionally two-dimensional. In: Applied Psychological Measurement. 1997 ; Vol. 21, No. 3. pp. 195-213.
@article{badba7fbc9e343ffbe1ee92af272c6fd,
title = "MULTISIB: A procedure to investigate DIF when a test is intentionally two-dimensional",
abstract = "MULTISIB is proposed as a statistical test for assessing differential item functioning (DIF) of intentionally two-dimensional test data, such as a mathematics test designed to measure algebra and geometry. MULTISIB is based on the multidimensional model of DIF as presented in Shealy & Stout (1993), and is a direct extension of SIBTEST, its unidimensional counterpart. For an intentionally two-dimensional test, DIF is appropriately modeled to result from secondary dimensional influence from other than the two intended dimensions. Simulation studies were used to assess the performance of MULTISIB to detect DIF in intentionally two-dimensional tests. These results indicate that MULTISIB exhibited reasonably good adherence to the nominal level of significance and good power. Moreover, for each DIF model the average amount of DIF estimated over the 100 simulations of the model by MULTISIB was close to the true value, confirming its relative lack of statistical estimation bias in assessing true DIF. In addition, the simulation studies supported the importance of using the regression correction to adjust the scores on the studied item due to impact and the importance of matching examinees on two subtest scores instead of the total test score.",
author = "William Stout and Li, {Hsin Hung} and Ratna Nandakumar and Daniel Bolt",
year = "1997",
month = "9",
doi = "10.1177/01466216970213001",
language = "English",
volume = "21",
pages = "195--213",
journal = "Applied Psychological Measurement",
issn = "0146-6216",
publisher = "SAGE Publications Inc.",
number = "3",

}

MULTISIB : A procedure to investigate DIF when a test is intentionally two-dimensional. / Stout, William; Li, Hsin Hung; Nandakumar, Ratna; Bolt, Daniel.

In: Applied Psychological Measurement, Vol. 21, No. 3, 09.1997, p. 195-213.

Research output: Contribution to journalArticle

TY - JOUR

T1 - MULTISIB

T2 - A procedure to investigate DIF when a test is intentionally two-dimensional

AU - Stout, William

AU - Li, Hsin Hung

AU - Nandakumar, Ratna

AU - Bolt, Daniel

PY - 1997/9

Y1 - 1997/9

N2 - MULTISIB is proposed as a statistical test for assessing differential item functioning (DIF) of intentionally two-dimensional test data, such as a mathematics test designed to measure algebra and geometry. MULTISIB is based on the multidimensional model of DIF as presented in Shealy & Stout (1993), and is a direct extension of SIBTEST, its unidimensional counterpart. For an intentionally two-dimensional test, DIF is appropriately modeled to result from secondary dimensional influence from other than the two intended dimensions. Simulation studies were used to assess the performance of MULTISIB to detect DIF in intentionally two-dimensional tests. These results indicate that MULTISIB exhibited reasonably good adherence to the nominal level of significance and good power. Moreover, for each DIF model the average amount of DIF estimated over the 100 simulations of the model by MULTISIB was close to the true value, confirming its relative lack of statistical estimation bias in assessing true DIF. In addition, the simulation studies supported the importance of using the regression correction to adjust the scores on the studied item due to impact and the importance of matching examinees on two subtest scores instead of the total test score.

AB - MULTISIB is proposed as a statistical test for assessing differential item functioning (DIF) of intentionally two-dimensional test data, such as a mathematics test designed to measure algebra and geometry. MULTISIB is based on the multidimensional model of DIF as presented in Shealy & Stout (1993), and is a direct extension of SIBTEST, its unidimensional counterpart. For an intentionally two-dimensional test, DIF is appropriately modeled to result from secondary dimensional influence from other than the two intended dimensions. Simulation studies were used to assess the performance of MULTISIB to detect DIF in intentionally two-dimensional tests. These results indicate that MULTISIB exhibited reasonably good adherence to the nominal level of significance and good power. Moreover, for each DIF model the average amount of DIF estimated over the 100 simulations of the model by MULTISIB was close to the true value, confirming its relative lack of statistical estimation bias in assessing true DIF. In addition, the simulation studies supported the importance of using the regression correction to adjust the scores on the studied item due to impact and the importance of matching examinees on two subtest scores instead of the total test score.

UR - http://www.scopus.com/inward/record.url?scp=0031519834&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0031519834&partnerID=8YFLogxK

U2 - 10.1177/01466216970213001

DO - 10.1177/01466216970213001

M3 - Article

AN - SCOPUS:0031519834

VL - 21

SP - 195

EP - 213

JO - Applied Psychological Measurement

JF - Applied Psychological Measurement

SN - 0146-6216

IS - 3

ER -