Biases and Standard Errors of Standardized Regression Coefficients

The paper obtains consistent standard errors (SE) and biases of order O(1/n) for the sample standardized regression coefficients with both random and given predictors. Analytical results indicate that the formulas for SEs given in popular text books are consistent only when the population value of the regression coefficient is zero. The sample standardized regression coefficients are also biased in general, although it should not be a concern in practice when the sample size is not too small. Monte Carlo results imply that, for both standardized and unstandardized sample regression coefficients, SE estimates based on asymptotics tend to under-predict the empirical ones at smaller sample sizes.     

An Investigation of the Accuracy of Standardized Path Coefficients

The present study investigated potential sex differences in preferences for spicy, hot or unusual foods and in food aversions or neophobia. This study also reexamined the issue of a gender difference in preference for sweet foods. Questionnaires concerning past and current food use and preferences as well as food and condiment use in one actual meal were completed by 148 people between 17 and 32 years of age. Their responses clearly support the prediction that men tend to have a stronger preference than women for spicy, hot foods. The results also support the prediction that men are more likely than women to seek unusual and new foods. Both sexes showed the same degree of preference for condiment use.     

Standardized Regression Coefficients and Newly Proposed Estimators for $${R}^{{2}}$$R2 in Multiply Imputed Data

Psychometrika - Whenever statistical analyses are applied to multiply imputed datasets, specific formulas are needed to combine the results into one overall analysis, also called combination rules....     

Determination of the Regression Coefficients and Their Associated Standard Errors in Hierarchical Regression Analysis

Hierarchical regression analysis is potentially a very useful statistical technique for establishing the significance of sets of predictor variables. However, when a hierarchical analysis which is based on theory is performed, some estimation procedures for the regression coefficients and their associated standard errors are potentially inappropriate. Specifically, the hierarchical regression equations, the incremental or hierarchical tests, and the parameter estimation of this procedure may not correspond. This problem is investigated by the development of four approaches (simultaneous, stagewise, orthogonal, and hierarchical) of estimation to the analysis. For each method, regression, coefficient estimators and their standard errors are determined. By comparison of these approaches, the use of orthogonalized sets of predictor variates or a modification to a series of simultaneous analyses are recommended as the most sensible technique for a theory driven hierarchical analysis.     

In Spite of Indeterminacy Many Common Factor Score Estimates Yield an Identical Reproduced Covariance Matrix

It was investigated whether commonly used factor score estimates lead to the same reproduced covariance matrix of observed variables. This was achieved by means of Schönemann and Steiger’s (1976) regression component analysis, since it is possible to compute the reproduced covariance matrices of the regression components corresponding to different factor score estimates. It was shown that Thurstone’s, Ledermann’s, Bartlett’s, Anderson-Rubin’s, McDonald’s, Krijnen, Wansbeek, and Ten Berge’s, as well as Takeuchi, Yanai, and Mukherjee’s score estimates reproduce the same covariance matrix. In contrast, Harman’s ideal variables score estimates lead to a different reproduced covariance matrix.     

A Simple Simulation Technique for Nonnormal Data with Prespecified Skewness,Kurtosis, and Covariance Matrix

We present and investigate a simple way to generate nonnormal data using linear combinations of independent generator (IG) variables. The simulated data have prespecified univariate skewness and kurtosis and a given covariance matrix. In contrast to the widely used Vale-Maurelli (VM) transform, the obtained data are shown to have a non-Gaussian copula. We analytically obtain asymptotic robustness conditions for the IG distribution. We show empirically that popular test statistics in covariance analysis tend to reject true models more often under the IG transform than under the VM transform. This implies that overly optimistic evaluations of estimators and fit statistics in covariance structure analysis may be tempered by including the IG transform for nonnormal data generation. We provide an implementation of the IG transform in the R environment.     

The Systematic Distortion Hypothesis: A Confirmatory Test of the Implicit Covariance and General Impression Models

Supervisory ratings of subordinate performance were used to compare two variations of the systematic distortion hypothesis: an implicit covariance model and a general impression (GI) model. Our purpose was to assess whether a model specifying a set of correlated performance dimensions or a model specifying a single, GI factor provided the best representation of actual supervisory performance ratings (N = 491). Confirmatory factor analysis was used to test a first-order correlated factor (i.e., implicit covariance) model against a second-order hierarchical (i.e., GI) model. The results generally supported the latter perspective whereby a GI is thought to serve as the basis for subsequent dimensional judgments, which in turn serve as the basis for individual item ratings. The implications of these results with respect to rater training and rating accuracy are discussed.     

Skepticism revisited: Chalmers on The Matrix and brains-in-vats

Thought experiments involving The Matrix, brains-in-vats, or Cartesian demons have traditionally thought to describe skeptical possibilities. Chalmers has denied this, claiming that the simulations involved are real enough to at least sometimes defeat the skeptic. Through an examination of the meaning of kind terms in natural language I argue that, though the Chalmers view may be otherwise attractive, it is not an antidote to skepticism.     

The Acceptability and Representativeness of Standardized Parent-Child Interaction Tasks

Analogue behavioral observation of structured parent-child interactions has often been used to obtain a standardized, unbiased measure of child noncompliance and parenting behavior. However, for assessment information to be clinically relevant, it is essential that the behavior observed be similar to that which the child normally experiences and elicits. Furthermore, assessment procedures should be socially appropriate and acceptable to participants in order to facilitate investment in the assessment, and potentially treatment, process. Although analogue tasks have been shown to have reasonable psychometric properties, the social validity of these methods has not been evaluated. This study examined the acceptability and representativeness of 4 widely utilized, laboratory-based parent-child interaction tasks in a sample of 43 nonreferred, 3- to 6-year old children and their mothers, who were of European-American or African-American ethnicity. Mothers rated the acceptability of each task and the degree to which it was representative of “typical” interactions occurring with the child. Repeated-measures analyses of variance revealed high levels of acceptability and representativeness for all tasks, but consistent differences across tasks. These ratings were predominantly stable over a 2-week period. Additionally, social validity ratings were significantly associated with parenting self-esteem, maternal ethnicity, child gender, and child age, but not child behavior problems. Limitations and implications of these findings for the assessment of parent-child interaction are discussed.