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

The effects of construct shift on gr...

Martineau, Joseph Arnold.

Linked to FindBook

Google Book

Amazon

博客來

The effects of construct shift on growth and accountability models.

Record Type:	Electronic resources : Monograph/item
Title/Author:	The effects of construct shift on growth and accountability models./
Author:	Martineau, Joseph Arnold.
Description:	196 p.
Notes:	Source: Dissertation Abstracts International, Volume: 65-04, Section: A, page: 1335.
Contained By:	Dissertation Abstracts International65-04A.
Subject:	Education, Tests and Measurements. -
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3129517
ISBN:	0496767763

The effects of construct shift on growth and accountability models.
Martineau, Joseph Arnold.

The effects of construct shift on growth and accountability models. - 196 p.

Source: Dissertation Abstracts International, Volume: 65-04, Section: A, page: 1335.

Thesis (Ph.D.)--Michigan State University, 2004.

It is desirable for many measurement applications, particularly in education, to have developmental scales that span wide developmental levels for the purpose of modeling growth. Many test companies provide such scales and claim they are adequate for growth modeling. However, scales spanning wide grade ranges also span wide content ranges. This shift in the constructs being measured from grade to grade jeopardizes the validity of inferences and interpretations made from growth models based upon developmental scales. It is shown how observed growth trajectories, the magnitude of effects on growth, and the ability to detect statistically significant effects on growth are affected by construct shift. It is shown how construct shift may cause distorted results in educational experiments, quasi-experiments, correlational studies, and accountability analyses. These distortions may be understatement or overstatement of effects on initial status or growth, detection of an effect when there is none, failure to detect a true effect, attribution of effects on initial status to effects on growth, attribution of effects on growth to effects on initial status, identification of highly effective teachers as ineffective simply because their students achievement is outside the range of achievement measured well by a given grade-level test, and the attribution of prior teacher effects to current teachers. Therefore, theories and policies based upon studies of growth that use developmental scale scores are likely to be invalid because of distorted conclusions about the shape of growth trajectories, about the effects of interventions on growth, about the effects of environmental variables on growth, or about educator effectiveness in eliciting growth.

ISBN: 0496767763Subjects--Topical Terms:

1017589
Education, Tests and Measurements.

The effects of construct shift on growth and accountability models.
LDR:02641nmm 2200277 4500 001 1846446
005 20051103093510.5
008 130614s2004 eng d
020 $a 0496767763
035 $a (UnM)AAI3129517
035 $a AAI3129517
040 $a UnM $c UnM
100 1 $a Martineau, Joseph Arnold. $3 1934559
245 1 4 $a The effects of construct shift on growth and accountability models.
300 $a 196 p.
500 $a Source: Dissertation Abstracts International, Volume: 65-04, Section: A, page: 1335.
500 $a Adviser: Mark D. Reckase.
502 $a Thesis (Ph.D.)--Michigan State University, 2004.
520 $a It is desirable for many measurement applications, particularly in education, to have developmental scales that span wide developmental levels for the purpose of modeling growth. Many test companies provide such scales and claim they are adequate for growth modeling. However, scales spanning wide grade ranges also span wide content ranges. This shift in the constructs being measured from grade to grade jeopardizes the validity of inferences and interpretations made from growth models based upon developmental scales. It is shown how observed growth trajectories, the magnitude of effects on growth, and the ability to detect statistically significant effects on growth are affected by construct shift. It is shown how construct shift may cause distorted results in educational experiments, quasi-experiments, correlational studies, and accountability analyses. These distortions may be understatement or overstatement of effects on initial status or growth, detection of an effect when there is none, failure to detect a true effect, attribution of effects on initial status to effects on growth, attribution of effects on growth to effects on initial status, identification of highly effective teachers as ineffective simply because their students achievement is outside the range of achievement measured well by a given grade-level test, and the attribution of prior teacher effects to current teachers. Therefore, theories and policies based upon studies of growth that use developmental scale scores are likely to be invalid because of distorted conclusions about the shape of growth trajectories, about the effects of interventions on growth, about the effects of environmental variables on growth, or about educator effectiveness in eliciting growth.
590 $a School code: 0128.
650 4 $a Education, Tests and Measurements. $3 1017589
650 4 $a Psychology, Psychometrics. $3 1017742
690 $a 0288
690 $a 0632
710 2 0 $a Michigan State University. $3 676168
773 0 $t Dissertation Abstracts International $g 65-04A.
790 1 0 $a Reckase, Mark D., $e advisor
790 $a 0128
791 $a Ph.D.
792 $a 2004
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3129517