NotesFAQContact Us
Search Tips
Showing all 12 results Save | Export
Peer reviewed Peer reviewed
Direct linkDirect link
Hardin, Andrew M.; Chang, Jerry Cha-Jan; Fuller, Mark A.; Torkzadeh, Gholamreza – Educational and Psychological Measurement, 2011
The use of causal indicators to formatively measure latent constructs appears to be on the rise, despite what appears to be a troubling lack of consistency in their application. Scholars in any discipline are responsible not only for advancing theoretical knowledge in their domain of study but also for addressing methodological issues that…
Descriptors: Structural Equation Models, Measurement, Statistical Data, Meta Analysis
Peer reviewed Peer reviewed
Direct linkDirect link
Ding, Cody S.; Davison, Mark L. – Educational and Psychological Measurement, 2010
Akaike's information criterion is suggested as a tool for evaluating fit and dimensionality in metric multidimensional scaling that uses least squares methods of estimation. This criterion combines the least squares loss function with the number of estimated parameters. Numerical examples are presented. The results from analyses of both simulation…
Descriptors: Multidimensional Scaling, Least Squares Statistics, Criteria, Computation
Peer reviewed Peer reviewed
Direct linkDirect link
Weitzman, R. A. – Educational and Psychological Measurement, 2009
Building on the Kelley and Gulliksen versions of classical test theory, this article shows that a logistic model having only a single item parameter can account for varying item discrimination, as well as difficulty, by using item-test correlations to adjust incorrect-correct (0-1) item responses prior to an initial model fit. The fit occurs…
Descriptors: Item Response Theory, Test Items, Difficulty Level, Test Bias
Peer reviewed Peer reviewed
Lix, Lisa M.; Keselman, H. J. – Educational and Psychological Measurement, 1998
Comparison of six procedures to test for location equality among two or more groups when population variances are heterogeneous suggests that, when the variance homogeneity and normality assumptions are not satisfied, and the design is unbalanced, the use of any of these test statistics with the usual least squares estimators is not recommended.…
Descriptors: Comparative Analysis, Estimation (Mathematics), Least Squares Statistics, Research Design
Peer reviewed Peer reviewed
Kromrey, Jeffrey D.; Foster-Johnson, Lynn – Educational and Psychological Measurement, 1998
Provides a comparison of centered and raw-score analyses in least squares regression. The two methods are demonstrated with constructed data in a Monte Carlo study to be equivalent, yielding identical hypothesis tests associated with the moderation effect and regression equations that are functionally equivalent. (SLD)
Descriptors: Hypothesis Testing, Least Squares Statistics, Monte Carlo Methods, Raw Scores
Peer reviewed Peer reviewed
Bentler, Peter M.; Yuan, Ke-Hai – Educational and Psychological Measurement, 2000
Proposes generalized least squares methods in structural equation models to estimate potential mean structure parameters and to evaluate whether the given model can be augmented successfully with a mean structure. A simulation shows that a method that takes variability due to the estimate of covariance structure parameters into account performs…
Descriptors: Evaluation Methods, Least Squares Statistics, Simulation
Peer reviewed Peer reviewed
Sturman, Michael C. – Educational and Psychological Measurement, 1999
Compares eight models for analyzing count data through simulation in the context of prediction of absenteeism to indicate the extent to which each model produces false positives. Results suggest that ordinary least-squares regression does not produce more false positives than expected by chance. The Tobit and Poisson models do yield too many false…
Descriptors: Attendance, Individual Differences, Least Squares Statistics, Models
Peer reviewed Peer reviewed
Brown, R. L. – Educational and Psychological Measurement, 1992
A Monte Carlo study explores the robustness assumption in structural equation modeling of using a full information normal theory generalized least-squares estimation procedure on Type I censored data. The efficacy of the following proposed alternate estimation procedures is assessed: asymptotically distribution free estimator and a latent…
Descriptors: Computer Simulation, Equations (Mathematics), Estimation (Mathematics), Least Squares Statistics
Peer reviewed Peer reviewed
Wolins, Leroy – Educational and Psychological Measurement, 1995
From 105 samples of 300 observations each and 87 samples with 3,000 observations each, constrained factor analyses of 96 normally distributed variables in a three-stage hierarchical structure were computed by maximum likelihood and unweighted least squares (ULS). ULS took less time and computer resources and led to better estimates. (SLD)
Descriptors: Estimation (Mathematics), Factor Analysis, Least Squares Statistics, Maximum Likelihood Statistics
Peer reviewed Peer reviewed
Levin, Joseph – Educational and Psychological Measurement, 1991
A reanalysis of the intercorrelation matrix from a principal components analysis of the Life Styles Inventory was conducted using a Canadian sample. Using nonmetric multidimensional scaling, analyses show an almost perfect circumplex pattern. Results illustrate the inadequacy of factor analytic procedures for the analysis and representation of a…
Descriptors: Attitude Measures, Correlation, Factor Analysis, Foreign Countries
Peer reviewed Peer reviewed
Elliott, Steven D. – Educational and Psychological Measurement, 1989
The method of unweighted means (MUM) was studied to determine Type I error rates associated with various degrees of imbalance among the cell frequencies. Conditions under which the MUM is a reasonable alternative to the least squares method are discussed, and the issue of power is considered. (SLD)
Descriptors: Analysis of Variance, Least Squares Statistics, Multivariate Analysis, Statistical Analysis
Peer reviewed Peer reviewed
Wasik, John L. – Educational and Psychological Measurement, 1981
The use of segmented polynomial models is explained. Examples of design matrices of dummy variables are given for the least squares analyses of time series and discontinuity quasi-experimental research designs. Linear combinations of dummy variable vectors appear to provide tests of effects in the two quasi-experimental designs. (Author/BW)
Descriptors: Least Squares Statistics, Mathematical Models, Multiple Regression Analysis, Quasiexperimental Design