Sampling behaviour in estimating predictive validity in the context of selection and latent variable modelling: A Monte Carlo study

A small Monte Carlo study was conducted to determine whether MAXCOV analysis, a taxometric method for testing between discrete (“taxonic”) and continuous models of latent variables, is robust when indicators of the latent variable are skewed. Analysis of constructed data sets containing three levels of skew indicated that the MAXCOV procedure is unlikely to yield spurious findings of taxonicity even when skewness is considerable. However, care must be taken to distinguish low base-rate taxonic variables from skewed nontaxonic variables.

Download Full-text

Robustness of Taxometric Analysis with Skewed Indicators: I. a Monte Carlo Study of the Mambac Procedure

Psychological Reports ◽

10.2466/pr0.1996.79.1.243 ◽

1996 ◽

Vol 79 (1) ◽

pp. 243-248 ◽

Cited By ~ 24

Author(s):

Charles Cleland ◽

Nick Haslam

Keyword(s):

Monte Carlo ◽

Latent Variables ◽

Latent Variable ◽

Monte Carlo Study ◽

Data Sets ◽

Taxometric Analysis ◽

Continuous Models ◽

The Family ◽

Taxometric Method

A small Monte Carlo study was conducted to determine whether Meehl and Yonce's (1994) MAMBAC procedure—a taxometric method for testing between discrete and continuous models of latent variables—is robust when the latent variable and its manifest indicators are skewed Analysis of constructed data sets containing three levels of skew indicated that the MAMBAC procedure is highly unlikely to yield spurious findings of discreteness (“taxonicity”) even when skewness is considerable. MAMBAC appears to be a robust and promising addition to the family of taxometric procedures.

Download Full-text

Alternative Estimation Methods for Conjoint Analysis: A Monté Carlo Study

Journal of Marketing Research ◽

10.1177/002224378101800112 ◽

1981 ◽

Vol 18 (1) ◽

pp. 101-106 ◽

Cited By ~ 7

Author(s):

Dick R. Wittink ◽

Philippe Cattin

Keyword(s):

Monte Carlo ◽

Conjoint Analysis ◽

Predictive Validity ◽

Simulation Study ◽

Rank Order ◽

Monte Carlo Study ◽

Estimation Methods ◽

Preference Rank ◽

Estimation Procedures ◽

Simulation Results

Conjoint analysis has been applied in a large number of commercial projects as well as in many noncommercial studies. Often MONANOVA, a nonmetric technique, is applied to a preference rank order obtained for a set of hypothetical objects. The authors report simulation results obtained for four alternative estimation procedures, ANOVA, LINMAP, LOGIT, and MONANOVA. The results suggest, within the limitations of the simulation study, that ANOVA may be the preferred procedure for compensatory models, whereas LINMAP is most likely to provide the best predictive validity for models with a dominant attribute.

Download Full-text

An integral equation and Monte Carlo study of homo- and hetero nuclear square-well diatomic fluids

Molecular Physics ◽

10.1080/00268970050199770 ◽

2000 ◽

Vol 98 (24) ◽

pp. 2033-2043

Author(s):

Vít Jirásek, Stanislav Labík, Anatol Ma

Keyword(s):

Integral Equation ◽

Monte Carlo ◽

Monte Carlo Study ◽

Square Well

Download Full-text

A grand canonical ensemble Monte Carlo study of confined planar and homeotropically anchored Gay—Berne films

Molecular Physics ◽

10.1080/00268979809482254 ◽

1998 ◽

Vol 93 (4) ◽

pp. 681-692

Author(s):

THOMAS GRUHN ◽

MARTIN SCHOEN

Keyword(s):

Monte Carlo ◽

Canonical Ensemble ◽

Monte Carlo Study ◽

Grand Canonical Ensemble ◽

Ensemble Monte Carlo ◽

Grand Canonical

Download Full-text

Analyzing Observed Composite Differences Across Groups

Methodology ◽

10.1027/1614-2241/a000049 ◽

2013 ◽

Vol 9 (1) ◽

pp. 1-12 ◽

Cited By ~ 101

Author(s):

Holger Steinmetz

Keyword(s):

Monte Carlo ◽

Structural Equation Modeling ◽

Measurement Invariance ◽

Structural Equation ◽

Monte Carlo Study ◽

Equation Modeling ◽

Factor Loadings ◽

Latent Mean Differences ◽

Partial Invariance ◽

Mean Differences

Although the use of structural equation modeling has increased during the last decades, the typical procedure to investigate mean differences across groups is still to create an observed composite score from several indicators and to compare the composite’s mean across the groups. Whereas the structural equation modeling literature has emphasized that a comparison of latent means presupposes equal factor loadings and indicator intercepts for most of the indicators (i.e., partial invariance), it is still unknown if partial invariance is sufficient when relying on observed composites. This Monte-Carlo study investigated whether one or two unequal factor loadings and indicator intercepts in a composite can lead to wrong conclusions regarding latent mean differences. Results show that unequal indicator intercepts substantially affect the composite mean difference and the probability of a significant composite difference. In contrast, unequal factor loadings demonstrate only small effects. It is concluded that analyses of composite differences are only warranted in conditions of full measurement invariance, and the author recommends the analyses of latent mean differences with structural equation modeling instead.

Download Full-text