1994
DOI: 10.1037/h0088287
|View full text |Cite
|
Sign up to set email alerts
|

Structural analysis of correlated factors: Lessons from the verbal-performance dichotomy of the Wechsler scales.

Abstract: Although the constructs of interest in psychological assessment and applied research often are correlated with other variables, a fundamental question has not been adequately resolved: "Just how high can the correlations among supposedly discrete constructs or factors be?" Logically, based on the application of convergent and discriminant validity criteria, the factors should have higher correlations with variables that they are intended to measure than with variables that they are not intended to measure. The… Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1

Citation Types

1
28
0

Year Published

1995
1995
2016
2016

Publication Types

Select...
5
2
1

Relationship

0
8

Authors

Journals

citations
Cited by 51 publications
(29 citation statements)
references
References 125 publications
(230 reference statements)
1
28
0
Order By: Relevance
“…This procedure presents advantages in terms of the statistical processing of relatively small samples (200 to 500 subjects) and consequently, the fitness index seems to work better with ML than with other statistical estimation procedures (Hoyle, 1998). Regarding our choice of fit indices, our decision is essentially based on the chisquare analysis, GFI (Goodness-of-Fit Index), RMR (Root Mean Residual), RMSEA (Root Mean Squared Error of Approximation) and AIC (Akaike Information Criterion), taking the indices suggested in the literature (Akaike, 1987;Hancock & Freeman, 2001;Macmann & Barnett, 1994). Note: M = mean; SD = standard deviation.…”
Section: Resultsmentioning
confidence: 99%
“…This procedure presents advantages in terms of the statistical processing of relatively small samples (200 to 500 subjects) and consequently, the fitness index seems to work better with ML than with other statistical estimation procedures (Hoyle, 1998). Regarding our choice of fit indices, our decision is essentially based on the chisquare analysis, GFI (Goodness-of-Fit Index), RMR (Root Mean Residual), RMSEA (Root Mean Squared Error of Approximation) and AIC (Akaike Information Criterion), taking the indices suggested in the literature (Akaike, 1987;Hancock & Freeman, 2001;Macmann & Barnett, 1994). Note: M = mean; SD = standard deviation.…”
Section: Resultsmentioning
confidence: 99%
“…The following guidelines could be used concerning model fit criteria and acceptable fit interpretation [5,16,20,35]: the Goodness-of-Fit Index (GFI) and the Adjusted Goodness-of-Fit Index (AGFI), that is, GFI adjusted for the number of variables and the degrees of freedom, where values greater than 0.90 and close to 0.95 reflect a good fit. The Comparative Fit Index (CFI) with values closer to 1.0 implying good model fit and also the advantage of reflecting the degree of fit relatively well at all sample sizes.…”
Section: Discussionmentioning
confidence: 99%
“…Accordingly, we tried to confirm the model obtained with estimation procedures of maximum likelihood, by using the following fit indices: chi-square, comparative fit index (CFI), standardised root mean residual (SRMR), and root mean square error of approximation (RMSEA). According to Macmann and Barnett (1994), the chi-square allows fit testing between the model and the observed covariance matrix, in the sense that the smaller the value, the better the fit. In this case, a greater proximity to the degree of freedom (df) is required and therefore, the null hypothesis may be accepted (p !.05).…”
Section: Construct Validitymentioning
confidence: 99%