2001
DOI: 10.1348/096317901167415
|View full text |Cite
|
Sign up to set email alerts
|

The construct validity of a Belgian assessment centre: A comparison of different models

Abstract: This study investigates the construct validity of a Belgian assessment centre through a comprehensive set of con rmatory factor analysis models. In particular, the general con rmatory factor analysis approach as well as the correlated uniqueness approach are tted to the same data. Results replicate Sagie and Magnezy's (1997) nding that the correlated uniqueness model is appropriate to represent exercise eVects in assessment centres, as a good t and no estimation problems are obtained. In this model the dimensi… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
5
0
1

Year Published

2002
2002
2009
2009

Publication Types

Select...
6
1

Relationship

2
5

Authors

Journals

citations
Cited by 8 publications
(6 citation statements)
references
References 12 publications
0
5
0
1
Order By: Relevance
“…The appropriateness of the multiple group structural equation modeling analysis was determined by goodness-of-fit indices. The model fit was assessed using the Tucker-Lewis Index, Comparative Fit Index, and root-mean-square error of approximation (Kenny & McCoach, 2003;Lievens & Van Keer, 2001). Coefficients of paths between constructs were examined to see whether path coefficients were invariant between the full-and part-time employees.…”
Section: Discussionmentioning
confidence: 99%
“…The appropriateness of the multiple group structural equation modeling analysis was determined by goodness-of-fit indices. The model fit was assessed using the Tucker-Lewis Index, Comparative Fit Index, and root-mean-square error of approximation (Kenny & McCoach, 2003;Lievens & Van Keer, 2001). Coefficients of paths between constructs were examined to see whether path coefficients were invariant between the full-and part-time employees.…”
Section: Discussionmentioning
confidence: 99%
“…In addition, they have been found in ACs conducted all over the world. Apart from the United States (Bycio, Alvares, & Hahn, 1987;Harris, Becker, & Smith, 1993;Kudisch, Ladd, & Dobbins, 1997;Reilly, Henry, & Smither, 1990;Sackett & Dreher, 1982;Schneider & Schmitt, 1992;Silverman, Dalessio, Woods, & Johnson, 1986), these results have been established in the United Kingdom (Anderson, Lievens, van Dam, & Born, 2006;Crawley, Pinder, & Herriot, 1990;Robertson, Gratton, & Sharpley, 1987), Germany (Kleinmann & Ko¨ller, 1997;Kleinmann, Kuptsch, & Ko¨ller, 1996), Belgium (Lievens & Van Keer, 2001), France (Borteyrou, 2005;Rolland, 1999), Australia (Atkins & Wood, 2002), New Zealand (Jackson, Stillman, & Atkins, 2005), China (Wu & Zhang, 2001), and Singapore (Chan, 1996).…”
Section: Exercises As Key Components In Assessment Centersmentioning
confidence: 95%
“…The RMSEA is particularly well suited to determining the number of latent subconstructs in a construct because it decreases when an additional sub-construct reduces the F 0 (minimal population discrepancy function) substantially, but increases, showing poorer model fit, if including the additional subconstruct only reduces F 0 slightly (Steiger and Lind 1980). Thus, it is used to provide guidance on the number of optimal subconstructs to use (Lievens and Van Keer 2001), which was central to our objective of confirming the constituent dimensions of the high-level trust constructs. Convergent validity was assessed using three criteria: individual item lambda coefficients greater than 0.70, a significant (0.05 level) t statistic for each path (Gefen et al 2000), and each path loading greater than twice its standard error (Anderson and Gerbing 1988).…”
Section: Phase 2: Confirmatory Factor Analysis (Cfa)mentioning
confidence: 99%