The Effects of Choice Weights and Item Weights on the Reliability and Predictive Validity of Aptitude-Type Tests. Final Report [electronic resource] / Robert J. Bayuk.

An investigation was conducted to determine the effects of response-category weighting and item weighting on reliability and predictive validity. Response-category weighting refers to scoring in which, for each category (including omit and "not read"), a weight is assigned that is proporti...

Full description

Saved in:
Bibliographic Details
Online Access: Full Text (via ERIC)
Main Author: Bayuk, Robert J.
Corporate Author: University of Pensylvania
Format: Electronic eBook
Language:English
Published: [S.l.] : Distributed by ERIC Clearinghouse, 1973.
Subjects:

MARC

LEADER 00000cam a22000002u 4500
001 b6505635
003 CoU
005 20081118081318.7
006 m d f
007 cr un
008 730301s1973 xx |||| o ||| | eng d
035 |a (ERIC)ed078061 
040 |a ericd  |c ericd  |d MvI 
099 |f ERIC DOC #  |a ED078061 
099 |f ERIC DOC #  |a ED078061 
100 1 |a Bayuk, Robert J. 
245 1 4 |a The Effects of Choice Weights and Item Weights on the Reliability and Predictive Validity of Aptitude-Type Tests. Final Report  |h [electronic resource] /  |c Robert J. Bayuk. 
260 |a [S.l.] :  |b Distributed by ERIC Clearinghouse,  |c 1973. 
300 |a 77 p. 
500 |a ERIC Document Number: ED078061. 
500 |a Sponsoring Agency: National Center for Educational Research and Development (DHEW/OE), Washington, DC. Regional Research Program.  |5 ericd. 
500 |a Contract Number: OEG-3-71-0108.  |5 ericd. 
520 |a An investigation was conducted to determine the effects of response-category weighting and item weighting on reliability and predictive validity. Response-category weighting refers to scoring in which, for each category (including omit and "not read"), a weight is assigned that is proportional to the mean criterion score of examinees selecting that category. Item weighting refers to the application of multiple regression techniques to maximize the relationship between a composite of item scores and a criterion. The study of the effects of weighting on reliability indicated that scores resulting from response-category weighting were significantly more reliable than scores corrected for chance success. Response-category weighting in concert with item weighting resulted in scores significantly less reliable than scores corrected for chance success. The study of the effects of the weighting on predictive validity indicated that no gain in predictive validity accrued through the use of response-category weighting as opposed to scores corrected for chance success. Response-category weighting with item weighting resulted in scores significantly more reliable than scores corrected for chance success. (Author/CK) 
650 1 7 |a Aptitude Tests.  |2 ericd. 
650 0 7 |a Correlation.  |2 ericd. 
650 1 7 |a Predictive Validity.  |2 ericd. 
650 0 7 |a Research Reports.  |2 ericd. 
650 0 7 |a Response Style (Tests)  |2 ericd. 
650 0 7 |a Scoring Formulas.  |2 ericd. 
650 0 7 |a Statistical Analysis.  |2 ericd. 
650 0 7 |a Test Interpretation.  |2 ericd. 
650 1 7 |a Test Reliability.  |2 ericd. 
650 0 7 |a Test Results.  |2 ericd. 
650 1 7 |a Weighted Scores.  |2 ericd. 
710 2 |a University of Pensylvania. 
856 4 0 |u http://files.eric.ed.gov/fulltext/ED078061.pdf  |z Full Text (via ERIC) 
907 |a .b65056358  |b 07-06-22  |c 10-19-10 
998 |a web  |b 10-19-12  |c f  |d m   |e -  |f eng  |g xx   |h 4  |i 1 
956 |a ERIC 
999 f f |i d09ffce9-9516-5874-b7f6-fed23f0b2d42  |s 45a91601-befa-57cb-83cb-02ff449284aa 
952 f f |p Can circulate  |a University of Colorado Boulder  |b Online  |c Online  |d Online  |e ED078061  |h Other scheme  |i web  |n 1