Skip to main content
eScholarship
Open Access Publications from the University of California

Scholarly Works (50 results)

  • Article
  • Peer Reviewed
UC Riverside Previously Published Works (2018)
Common factor analysis (FA) and principal component analysis (PCA) are commonly used to obtain lower-dimensional representations of matrices of correlations among manifest variables. Whereas some experts argue that differences in results from use of FA and PCA are small and relatively unimportant in empirical studies, the fundamental rationales for the two methods are very different. Here, FA and PCA are contrasted on four key issues: the range of possible dimensional loadings, the range of potential correlations among dimensions, the structure of residual covariances and correlations, and the relation between population parameters and the correlational structures with which they are associated. For decades, experts have emphasized indeterminacies of the FA model, particularly indeterminacy of common factor scores. Determinate in most respects, a heretofore unacknowledged, pernicious indeterminacy of PCA is demonstrated: the indeterminacy between PCA structural representations and the correlational structures from which they are derived. Researchers are often advised to use either FA or PCA in exploratory rounds of data analysis to understand and refine the dimensional structure of a domain before moving to Structural Equation Modeling in later theory-testing, confirmatory, replication studies. Results from the current study suggest that PCA is an unreliable method to use for such purposes and may lead to serious misrepresentation of the structure of a domain. Hence, PCA should never be used if the goal is to understand and represent the latent structure of a domain; only FA techniques should be used for this purpose, as only FA provides reliable structural representations as the basis for confirmatory tests in future studies.
    Cover page: On Common Factor and Principal Component Representations of Data: Implications for Theory and for Confirmatory Replications
    • Article
    • Peer Reviewed
    UC Riverside Previously Published Works (2023)
    Indices of cumulative risk (CR) have long been used in developmental research to encode the number of risk factors a child or adolescent experiences that may impede optimal developmental outcomes. Initial contributions concentrated on indices of cumulative environmental risk; more recently, indices of cumulative genetic risk have been employed. In this article, regression analytic methods are proposed for interrogating strongly the validity of risk indices by testing optimality of compositing weights, enabling more informative modeling of effects of CR indices. Reanalyses of data from two studies are reported. One study involved 10 environmental risk factors predicting Verbal IQ in 215 four-year-old children. The second study included an index of genetic CR in a G×E interaction investigation of 281 target participants assessed at age 15 years and then again at age 31 years for observed hostility during videotaped interactions with close family relations. Principles to guide evaluation of results of statistical modeling are presented, and implications of results for research and theory are discussed. The ultimate goals of this paper are to develop stronger tests of conjectures involving CR indices and to promote methods for improving replicability of results across studies.
      Cover page: Interrogating the validity of cumulative indices of environmental and genetic risk for negative developmental outcomes
      • Thesis
      • Peer Reviewed
      UC Riverside Electronic Theses and Dissertations (2016)

      Students with low reading achievement experience difficulty accessing content area instruction. English learners from Hispanic homes traditional lag behind their English speaking counterparts. Previous research has identified various teaching methods that support reading achievement in English learners with particular emphasis on early grades. The current study adds to the existing literature by examining the effects of a small-group intervention cycle on middle school aged students. A small-group intervention was developed with components identified in the research literature on effective teaching methods for English learners. This cycle employed a gradual release of responsibility for instruction from the expert teacher to the students. Areas of reading acquisition receiving particular emphasis included phonemic awareness, phonics, vocabulary, and comprehension. Students were assigned to treatment groups randomly from a stratified sample of English language proficiency and Lexile reading achievement scores. Analyses of variance and covariance were employed to evaluate the differences in reading outcomes for the two treatment groups. The results suggest that research-based small-group intervention can have positive effects on low performing middle school students and would be a useful strategy for intervention.

        Cover page: Effects of a Research-Based Intervention on Reading Achievement of Middle School English Learners
        • Article
        • Peer Reviewed
        UC Riverside Previously Published Works (2023)
        Measurement is fundamental to all research in psychology and should be accorded greater scrutiny than typically occurs. Among other claims, McNeish and Wolf (Thinking twice about sum scores. Behavior Research Methods, 52, 2287-2305) argued that use of sum scores (a) implies that a highly constrained latent variable model underlies items comprising a scale, and (b) may misrepresent or bias relations with other criteria. The central claim by McNeish and Wolf that use of sum scores requires the assumption that a parallel test model underlies item responses is incorrect and without psychometric merit. Instead, if a set of items is unidimensional, estimators of reliability are available even if the factor model underlying the set of items does not have a highly constrained form. Thus, dimensionality of a set of items is the key issue, and whether strict constraints on parameter estimates do or do not hold dictate the appropriate way to estimate reliability. McNeish and Wolf also claimed that more precise forms of scoring, such as estimating factor scores, would be preferable to sum scores. We provide analytic bases for reliability estimation and then provide several demonstrations of reliability estimation and the relative advantages of sum scores and factor scores. We contend that several claims by McNeish and Wolf are questionable and that, as a result, multiple recommendations they made and conclusions they drew are incorrect. The upshot is that, once the dimensional structure of a set of items is verified, sum scores often have a solid psychometric basis and therefore are frequently quite adequate for psychological research.
          Cover page: Thinking thrice about sum scores, and then some more about measurement and analysis
          • Thesis
          • Peer Reviewed
          UC Riverside Electronic Theses and Dissertations (2017)

          Since there is limited research on the applicability of the Theory of Planned Behavior (TPB) model in educational contexts, the current cross-national comparison study aimed to investigate how non-cognitive factors, specifically, the TPB model components, affect students’ math outcomes in USA and their peers in three other countries, Germany, Japan, and Korea. Using internationally representative samples from the 2012 Programme for International Student Assessment (PISA) data, various structural equation models (SEM) were employed to examine the TPB applicability in the academic domain across four national groups.

          One key strength of the current investigation was the use of two random samples from each country (the derivation and cross-validation samples). The results of the factor structural analyses from both the derivation (N = 10681) and cross-validation (N = 10682) samples indicated that the measurement of the TPB model components were invariant across four countries. From both 1st/2nd samples, two findings were consistent with “West vs. Asia” contrast: (a) the influence of Attitude, among the TPB model components, on Math Achievement was much stronger in Japan and Korea than USA and Germany; (b) the estimated effect size of Math Achievement was much greater in Japan and Korea than USA and Germany.

          However, most other findings were contrasts showing that one Western country or one Asian country differed from the other three countries, inconsistent with a generic “West vs. Asia” cultural comparison model: (a) the influence of Control on Math Intention in USA was distinctively greater than for the other three countries; (b) the direct effect of Control on Math Achievement was strong in USA, Germany, and Korea, but weak in Japan; (c) lastly, the influence of Attitude on Math Intention and Math Intention on Math Achievement in Korea were much stronger than in the other three countries. In addition, among three Subjective Norm factors (Friend, Parent, and Teacher), the effect of SN-Teacher on Math Intention was much stronger than the other two SN factors in all four countries. The present cross-national comparison study concluded with implications for math education and future research directions.

            Cover page: Examining the Effects of Non-Cognitive Factors on Mathematics Achievement Across National Groups: USA, Germany, Japan, and Korea
            • Article
            • Peer Reviewed
            UC Riverside Previously Published Works (1979)
            Argues that there are a number of potential problems with the methodological employed in the study of T. J. Donahue and J. W. Costar (see record 1978-24142-001) and that these issues were not appropriately addressed in their article in order to provide the reader with a means of evaluating the correctness of the measurement instrument, the hypotheses tested, and the data analysis procedures. The problems are discussed. (5 ref) (PsycINFO Database Record (c) 2006 APA, all rights reserved). © 1979 American Psychological Association.
              Cover page: Using nonorthogonal analysis of variance designs in psychological research: Comment on Donahue and Costar
              • Article
              • Peer Reviewed
              UC Davis Previously Published Works (2015)
              Manuck, Craig, Flory, Halder, and Ferrell (2011) reported that a theoretically anticipated effect of family rearing on girls' menarcheal age was genetically moderated by two single nucleotide polymorphisms (SNPs) of the estrogen receptor-α gene. We sought to replicate and extend these findings, studying 210 White females followed from birth. The replication was general because a different measure of the rearing environment was used in this inquiry (i.e., maternal sensitivity) than in the prior one (i.e., family cohesion). Extensions of the work included prospective rather than retrospective measurements of the rearing environment, reports of first menstruation within a year of its occurrence rather than decades later, accounting for some heritability of menarcheal age by controlling for maternal age of menarche, and using a new model-fitting approach to competitively compare diathesis-stress versus differential-susceptibility models of Gene × Environment interaction. The replication/extension effort proved successful in the case of both estrogen receptor-α SNPs, with the Gene × Environment interactions principally reflecting diathesis-stress: lower levels of maternal sensitivity predicted earlier age of menarche for girls homozygous for the minor alleles of either SNP but not for girls carrying other genotypes. Results are discussed in light of the new analytic methods adopted.
                Cover page: Genetic moderation of effects of maternal sensitivity on girl's age of menarche: Replication of the Manuck et al. study
                • Article
                • Peer Reviewed
                UC Davis Previously Published Works (2018)
                How Latino youth cope with stressors may have implications for their adjustment. We examined how a temperamental characteristic (effortful control) and a contextual factor (economic hardship) were associated with Latino youth's coping. Individual differences in effortful control, a core facet of self-regulation, may contribute to coping as effortful control is consistently linked to adaptive behaviors during adolescence. We examined relations of effortful control and economic hardship to active coping in a sample of Mexican-origin youth (N = 674) across three time points (fifth to ninth grades). Although economic hardship negatively predicted coping and effortful control, effortful control positively predicted coping (controlling for prior levels). Findings support a resilience perspective by suggesting that effortful control may contribute to coping and thus counteract the negative effects of economic hardship.
                  Cover page: Longitudinal Relations of Economic Hardship and Effortful Control to Active Coping in Latino Youth
                  • Article
                  • Peer Reviewed
                  UC Riverside Previously Published Works (2013)
                    Cover page: Abstract: Uncrossing the Correlated Trait-Correlated Method Model for Multitrait-Multimethod Data
                    • Article
                    • Peer Reviewed
                    Proceedings of the Annual Meeting of the Cognitive Science Society, Volume 7 (1985)
                    Simple and complex addition problems were presented for true/false verification to 22 subjects across two times of measurement to test the general model for simple and complex addition proposed by Widaman, Cormier, & Geary (1985). Models fit to average RT data revealed that subjects were processing complex problems columnwise, beginning with the units column. Column sums seemed to be obtained through an incrementing process, and subjects exited problems as soon as a colimn error was encountered. Group level models were the same across complex problem types and for both times of measurement. However, individual level analyses suggested that nearly half of the subjects used a different processing strategy to obtain column sums for the second time of measurement. Results support the multi-staged model proposed by Widaman et al. (1985), but individual level results suggest that information processing models developed from group data may not represent the processing strategies used by all subjects, or the same subjects at different times.
                      Cover page: Cognitive Processing Strategies for Complex Addition
                      Top