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

Latent variable modeling of faking: ...

Burns, Gary N.

Linked to FindBook

Google Book

Amazon

博客來

Latent variable modeling of faking: Techniques and illustration.

Record Type:	Language materials, printed : Monograph/item
Title/Author:	Latent variable modeling of faking: Techniques and illustration./
Author:	Burns, Gary N.
Description:	96 p.
Notes:	Adviser: Neil Christiansen.
Contained By:	Dissertation Abstracts International67-11B.
Subject:	Psychology, Industrial. -
Online resource:	http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3242143
ISBN:	9780542980527

Latent variable modeling of faking: Techniques and illustration.
Burns, Gary N.

Latent variable modeling of faking: Techniques and illustration. - 96 p.

Adviser: Neil Christiansen.

Thesis (Ph.D.)--Central Michigan University, 2006.

Applicant faking continues to be a concern in applied personality testing. A number of different indices of faking for observed data but not for latent variables have been identified in prior research. Utilizing two existing data sets, the current study sought to provide a didactic example and illustration of how to model faking with latent variables in both between-subject and within-subject experimental designs. In both designs the effects of faking on the latent factor mean and covariance structures were examined. The between-subject design focused specifically on the use of methods factors to capture the variance associated with faking. The within-subject design examined different methods of partialing faking from true score variance and examined faking in the framework of latent growth curve modeling.

ISBN: 9780542980527Subjects--Topical Terms:

520063
Psychology, Industrial.

Latent variable modeling of faking: Techniques and illustration.
LDR:03329nam 2200325 a 45 001 970813
005 20110921
008 110921s2006 eng d
020 $a 9780542980527
035 $a (UMI)AAI3242143
035 $a AAI3242143
040 $a UMI $c UMI
100 1 $a Burns, Gary N. $3 1294854
245 1 0 $a Latent variable modeling of faking: Techniques and illustration.
300 $a 96 p.
500 $a Adviser: Neil Christiansen.
500 $a Source: Dissertation Abstracts International, Volume: 67-11, Section: B, page: 6761.
502 $a Thesis (Ph.D.)--Central Michigan University, 2006.
520 $a Applicant faking continues to be a concern in applied personality testing. A number of different indices of faking for observed data but not for latent variables have been identified in prior research. Utilizing two existing data sets, the current study sought to provide a didactic example and illustration of how to model faking with latent variables in both between-subject and within-subject experimental designs. In both designs the effects of faking on the latent factor mean and covariance structures were examined. The between-subject design focused specifically on the use of methods factors to capture the variance associated with faking. The within-subject design examined different methods of partialing faking from true score variance and examined faking in the framework of latent growth curve modeling.
520 $a Overall, results indicated that faking significantly changed the latent mean structure and the pattern of covariance amongst latent factors. Examination of method factors in the between-subjects design indicated that this may be a suitable approach to combat the attenuation of discriminant validity caused by faking, reducing the amount of shared variance amongst distorted traits by as much as 21.8 or 26.5%. The results also indicated that between-subject designs may not be the most appropriate method to utilize to examine the effects of faking on factor loadings, that within-subject designs may offer a more promising solution.
520 $a Examination of different methods of partitioning faking in the within-subject design highlighted faking as a multidimensional construct. Results indicated that multiple faking factors were needed to most accurately capture the variance associated with faking but that these faking factors were highly correlated. Comparison of these faking factors with traditional indicators of faking indicated that they were similar. More specifically, these faking factors were negatively related to with-person correlations of honest and faked responses and positively correlated with various change scores.
520 $a Finally, the current study illustrated an approach to using latent growth curve modeling to examine the effects of faking. Results indicated that this method paralleled results seen with observed variables with fewer of the associated drawbacks in examining change. These results also highlighted the multidimensional nature of faking.
590 $a School code: 6006.
650 4 $a Psychology, Industrial. $3 520063
650 4 $a Psychology, Personality. $3 1017585
650 4 $a Psychology, Psychometrics. $3 1017742
690 $a 0624
690 $a 0625
690 $a 0632
710 2 0 $a Central Michigan University. $3 1019276
773 0 $t Dissertation Abstracts International $g 67-11B.
790 $a 6006
790 1 0 $a Christiansen, Neil, $e advisor
791 $a Ph.D.
792 $a 2006
856 4 0 $u http://pqdd.sinica.edu.tw/twdaoapp/servlet/advanced?query=3242143