Search

Using Monte Carlo simulations, this research examined the performance of four missing data methods in SEM under different multivariate distributional conditions. The effects of four independent variables (sample size, missing proportion, distribution shape, and factor loading magnitude) were investigated on six outcome variables: convergence rate, parameter estimate bias, MSE of parameter estimates, standard error coverage, model rejection rate, and model goodness of fit--RMSEA. A three-factor CFA model was used. Findings indicated that FIML outperformed the other methods in MCAR, and MI should be used to increase the plausibility of MAR. SRPI was not comparable to the other three methods in either MCAR or MAR.

This study examined how well elementary students' performance on a set of commonly-used reading assessments conformed to a model of automatic print processing. The assessments included measures of word recognition-untimed, word recognition-timed, oral reading accuracy, oral reading rate, silent reading rate, and spelling. The proposed print-processing model, based on the work of Perfetti (1992) and Share (1995), held that contextual reading accuracy is directly related to automatized word knowledge which is directly related to reading rate. Structural equation modeling showed that the performance data had an acceptable fit to the proposed model and to a second, post hoc model in which automatic word recognition is directly related to contextual reading accuracy. However, additional regression analyses tended to favor the initial model. Because the six print-processing components in this study proved to be reliable and related, and because two of the components (word recognition-timed and spelling) are seldom used in conventional reading assessments, the results have important implications for practice.

Users assume statistical software packages produce accurate results. In this article, the authors systematically examined Statistical Package for the Social Sciences (SPSS) and Statistical Analysis System (SAS) for 3 analysis of variance (ANOVA) designs, mixed-effects ANOVA, fixed-effects analysis of covariance (ANCOVA), and nested ANOVA. For each model, the authors examined 3 different data sets. With the mixed-effects design, results were always correct for SPSS syntax and SAS syntax. For SPSS point-and-click, the F and p values for the random-effect were always incorrect as the wrong error term is used. With the ANCOVA design, results varied both by software package and by type of sums of squares. With the nested design, the p values for the F and multiple comparison procedure did not agree for the nonnested factor in SPSS point-and-click. Recommendations were made regarding which package to use for each design. (Contains 3 tables.)

The American Fear Survey Schedule for Children (FSSC-AM; J. J. Burnham, 1995, 2005) has been used to measure fears of children and adolescents. The FSSC-AM is based on the 2nd revision of a psychometrically sound and well-known fear scale (i.e., FSSC-II; E. Gullone & N. J. King, 1992). In this study, age and gender differences, fear intensity scores, and fear prevalence scores were analyzed across race/ethnicity (i.e., White, African American, and Hispanic populations). Multivariate analyses of variance yielded significant effects for race/ethnicity.

Complex survey designs often employ multistage cluster sampling designs and oversample particular units to ensure more accurate population parameter estimates. These issues must be accommodated in the analysis to ensure accurate parameter estimation. Incorporation of sample weights in some statistical procedures has been studied. However, research on the behavior of sample weights on estimates, standard errors, and fit measures in latent variable models is negligible, and studies examining methodology on latent variable modeling applications using extant data are rare. Using the Beginning Postsecondary Students Longitudinal Study 1990/92/94, the authors found, with mixed results, that a statistically significant difference exists in estimates and fit indices when weights and designs are applied versus when they are ignored.

This simulation study investigated the robustness of structural equation modeling to different degrees of nonnormality under 2 estimation methods, generalized least squares and maximum likelihood, and 4 sample sizes, 100, 250, 500, and 1,000. Each of the slight and severe nonnormality degrees was comprised of pure skewness, pure kurtosis, and both skewness and kurtosis. Bias and standard errors of parameter estimates were analyzed. In addition, an analysis of variance was conducted to investigate the effects of the 3 factors on several goodness-of-fit indexes. The study found that standard errors of parameter estimates were not significantly affected by estimation methods and nonnormality conditions. As expected, standard errors decreased at larger sample sizes. Parameter estimates were more sensitive to nonnormality than to sample size and estimation method. Chi-square was the least robust model fit index compared with Normed Fit Index, Nonnormed Fit Index, and Comparative Fit Index. Sample sizes of 100 or more are recommended for accurate parameter estimates.

A person-process-context ecological systems model provides an organizing framework for examining student and parental perceptions of school safety. Overall, adolescent age, SES, and neighborhood predicted school characteristics. School characteristics predicted parental involvement in schools, peers' educational expectations, exposure to delinquent peers, and so forth, which, in turn, each predicted perceptions of school safety. (Contains 45 references.) (Authors/PKP)

Finds that both low-ability and average-ability first-grade readers, and to a lesser extent high-ability readers, from different socioeconomic backgrounds, select books that are beyond students' reading level. Discusses influences of gender and genre on these selections. Discusses the relation of frustration-level texts to interest and motivation, as well as to supporting children's conceptual and genre knowledge growth. (SR)

In the research paper considered, Stahl and Miller compare the effectiveness of whole language/language experience programs versus basal reader approaches. The article is criticized for misrepresenting concepts related to emergent literacy and inadequately defining whole language, the lack of equivalence between the two nonbasal approaches, and problems with conclusions concerning decoding. (TJH)