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

A comparison of tests for adverse im...

Meyer, Lia Loveland.

Linked to FindBook

Google Book

Amazon

博客來

A comparison of tests for adverse impact on a cognitive ability test.

Record Type:	Language materials, printed : Monograph/item
Title/Author:	A comparison of tests for adverse impact on a cognitive ability test./
Author:	Meyer, Lia Loveland.
Description:	120 p.
Notes:	Adviser: Jane Halpert.
Contained By:	Dissertation Abstracts International66-12B.
Subject:	Psychology, Cognitive. -
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3200266
ISBN:	9780542444821

A comparison of tests for adverse impact on a cognitive ability test.
Meyer, Lia Loveland.

A comparison of tests for adverse impact on a cognitive ability test. - 120 p.

Adviser: Jane Halpert.

Thesis (Ph.D.)--DePaul University, 2005.

Alternative tests of adverse impact are available, but their accuracy in estimating adverse impact in the population varies according to the characteristics of the sample being evaluated, and investigations into when they are more or less accurate have not evaluated the phenomena using real-world data.

ISBN: 9780542444821Subjects--Topical Terms:

1017810
Psychology, Cognitive.

A comparison of tests for adverse impact on a cognitive ability test.
LDR:03337nam 2200349 a 45 001 969879
005 20110920
008 110921s2005 eng d
020 $a 9780542444821
035 $a (UnM)AAI3200266
035 $a AAI3200266
040 $a UnM $c UnM
100 1 $a Meyer, Lia Loveland. $3 1293935
245 1 2 $a A comparison of tests for adverse impact on a cognitive ability test.
300 $a 120 p.
500 $a Adviser: Jane Halpert.
500 $a Source: Dissertation Abstracts International, Volume: 66-12, Section: B, page: 6960.
502 $a Thesis (Ph.D.)--DePaul University, 2005.
520 $a Alternative tests of adverse impact are available, but their accuracy in estimating adverse impact in the population varies according to the characteristics of the sample being evaluated, and investigations into when they are more or less accurate have not evaluated the phenomena using real-world data.
520 $a In response to criticisms of the 4/5 rule (endorsed by the EEOC) for detecting the presence of adverse impact, alternative tests of adverse impact have been proposed. Some of these alternative tests include the Z D test, the ZIR test, the chi-square test, Upton's corrected chi-square test, Fisher's exact test, and confidence intervals. These tests do take into account sample characteristics (one criticism of the 4/5 rule), however unfortunately these same sample characteristics also appear to reduce the power of these tests, thus leading to errors.
520 $a Investigations of these alternative tests of adverse impact have been conducted using analytical solutions and Monte Carlo studies. The current study sought to confirm these estimates and simulations using real-world data, which may or may not perform as predicted.
520 $a Consistent with past research, analyses confirmed that the test adversely impacted African American and Hispanic participants. Contrary to past research, the test also adversely impacted Asian participants. Consistent with past research, all participants with a college degree or more scored higher than participants with less than a college degree. Similarly, most participants applying for higher complexity jobs scored higher than participants applying for lower complexity jobs. However, adverse impact was more severe in high complexity jobs than in medium and low complexity jobs.
520 $a The pattern of adverse impact detected by the ZD, ZIR, chi-square, Upton's corrected chi-square, and Fisher's exact tests were significantly influenced by sample size. The pattern of adverse impact detected by the ZIR test and the 4/5 rule were significantly influenced by selection rate.
520 $a Hypotheses related to specific relationships among the different tests of adverse impact were not supported. However, the 4/5 rule and confidence intervals were consistently more likely to detect adverse impact than the other five tests of adverse impact, across all conditions. Implications of these findings are discussed and potential areas of future research suggested.
590 $a School code: 0937.
650 4 $a Psychology, Cognitive. $3 1017810
650 4 $a Psychology, Industrial. $3 520063
650 4 $a Psychology, Psychometrics. $3 1017742
690 $a 0624
690 $a 0632
690 $a 0633
710 2 0 $a DePaul University. $3 1032307
773 0 $t Dissertation Abstracts International $g 66-12B.
790 $a 0937
790 1 0 $a Halpert, Jane, $e advisor
791 $a Ph.D.
792 $a 2005
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3200266