Program FACTOR at 10 : origins, development and future directions
01 May 2017-Psicothema (Colegio Oficial de Psicólogos del Principado de Asturias)-Vol. 29, Iss: 2, pp 236-240
TL;DR: A conceptual view of the origins, development and future directions of FACTOR, a popular free program for fitting the factor analysis (FA) model, which has attained a degree of technical development comparable to commercial software, and offers options not available elsewhere.
Abstract: espanolAntecedentes: se pretende dar una vision conceptual del origen, desarrollos y futuras lineas de investigacion de FACTOR, un popular programa no comercial de analisis factorial (AF). Metodo: el estudio se organiza en tres partes. En la primera se discute FACTOR en su etapa inicial (2006-2012) como un programa AF tradicional con opciones novedosas. En la segunda se discute la etapa actual (2013-2016) en la que FACTOR se presenta ya como un programa general enmarcado tanto en los modelos de ecuaciones estructurales como en la teoria de respuesta a los items. En la tercera parte, finalmente se discute la esperada evolucion futura del programa. Resultados: en la actualidad FACTOR ha alcanzado un grado de desarrollo tecnico comparable al software comercial, ofreciendo opciones no disponibles en otros programas. Discusion: se discuten algunas limitaciones, asi como varios puntos que requieren cambios o mejoras. Se discute tambien el funcionamiento del programa dentro de la comunidad de usuarios. EnglishBackground: We aim to provide a conceptual view of the origins, development and future directions of FACTOR, a popular free program for fitting the factor analysis (FA) model. Method: The study is organized into three parts. In the first part we discuss FACTOR in its initial period (2006-2012) as a traditional FA program with many new and cutting-edge features. The second part discusses the present period (2013-2016) in which FACTOR has developed into a comprehensive program embedded in the framework of structural equation modelling and item response theory. The third part discusses expected future developments. Results: at present FACTOR has attained a degree of technical development comparable to commercial software, and offers options not available elsewhere. Discussion: We discuss several shortcomings as well as points that require changes or improvements. We also discuss the functioning of FACTOR within its community of users.
Citations
More filters
TL;DR: The Italian version of the FCV-19S is valid and reliable in assessing fear of COVID-19 among the general Italian population, and its unidimensional properties are confirmed.
Abstract: The advent of COVID-19 worldwide has led to consequences for people’s health, both physical and psychological, such as fear and anxiety. This is the case in Italy, one of the countries most affected by the pandemic. Given the heightened fear concerning COVID-19 in Italy., the present study analyzed the psychometric properties of the Italian version of the Fear of COVID-19 Scale (FCV-19S). The sample comprised 250 Italian participants who were administered Italian versions of the FCV-19S, the Hospital Anxiety and Depression Scale (HADS), and the Severity Measure for Specific Phobia–Adult (SMSP-A). Several psychometric tests were performed to investigate the validity and reliability of the test including confirmatory factor analysis. Analysis of the data showed satisfactory psychometric characteristics and confirmed the scale’s unidimensional properties. The seven FCV-19S items had acceptable correlations with the test total (from .443 to .784). Furthermore, the loadings on the factor were significant and strong (from .684 to .897). The internal consistency was very good (α = .871). Construct validity for the FCV-19S was supported by significant and positive correlations with the HADS (r=.649) and SMSP-A (r=.703). The Italian version of the Fear of COVID-19 Scale is valid and reliable in assessing fear of COVID-19 among the general Italian population.
314 citations
Cites methods from "Program FACTOR at 10 : origins, dev..."
...…RMSEA 0.06, and SRMR 0.08 (i.e., Browne and Cudeck 1993) The analysis was carried out using the following statistical packages: FACTOR v. 10.10.01 (Ferrando and Lorenzo-Seva 2017), SPSS Statistics v.25 (IBM Corporation 2017), and “R” software (R Core Team 2014) with the lavaan package (Yves…...
[...]
TL;DR: In this paper, a conceptual and practical guide for estimating internal consistency reliability of measures obtained as item sum or mean is presented as a byproduct of the measurement model underlying the item responses, including descriptive data analysis, test of relevant measurement models, and computation of internal consistency coefficient and its confidence interval.
Abstract: Based on recent psychometric developments, this paper presents a conceptual and practical guide for estimating internal consistency reliability of measures obtained as item sum or mean. The internal consistency reliability coefficient is presented as a by-product of the measurement model underlying the item responses. A three-step procedure is proposed for its estimation, including descriptive data analysis, test of relevant measurement models, and computation of internal consistency coefficient and its confidence interval. Provided formulas include: (a) Cronbach’s alpha and omega coefficients for unidimensional measures with quantitative item response scales, (b) coefficients ordinal omega, ordinal alpha and nonlinear reliability for unidimensional measures with dichotomic and ordinal items, (c) coefficients omega and omega hierarchical for essentially unidimensional scales presenting method effects. The procedure is generalized to weighted sum measures, multidimensional scales, complex designs with multilevel and/or missing data and to scale development. Four illustrative numerical examples are fully explained and the data and the R syntax are provided.
297 citations
TL;DR: This article proposes a comprehensive approach for assessing the quality and appropriateness of exploratory factor analysis solutions intended for item calibration and individual scoring by assessing three groups of properties: strength and replicability, determinacy and accuracy of the individual score estimates, and closeness to unidimensionality in the case of multidimensional solutions.
Abstract: This article proposes a comprehensive approach for assessing the quality and appropriateness of exploratory factor analysis solutions intended for item calibration and individual scoring. Three groups of properties are assessed: (a) strength and replicability of the factorial solution, (b) determinacy and accuracy of the individual score estimates, and (c) closeness to unidimensionality in the case of multidimensional solutions. Within each group, indices are considered for two types of factor-analytic models: the linear model for continuous responses and the categorical-variable-methodology model that treats the item scores as ordered-categorical. All the indices proposed have been implemented in a noncommercial and widely known program for exploratory factor analysis. The usefulness of the proposal is illustrated with a real data example in the personality domain.
209 citations
Cites background or methods from "Program FACTOR at 10 : origins, dev..."
...For this reason, we decided to implement a unified treatment in FACTOR in which bootstrap-based confidence intervals are available for all the indices proposed here....
[...]
...However, the procedures proposed here can also be used with more restricted approaches based on Procrustres transformations against fully specified or semispecified targets, which are also available in FACTOR (e.g., Ferrando & Lorenzo-Seva, 2013)....
[...]
...All the indices proposed in this article have been implemented in version 10.5 of the program FACTOR (Ferrando & Lorenzo-Seva, 2017), a well-known, free exploratory factor analysis program that can be downloaded at http://psico.fcep.urv.cat/utilitats/ factor/....
[...]
...For decades, the dominant view regarding item FA has been that confirmatory FA is the way to go, while EFA is at best a rough precursor that can be useful only in the preliminary stages of the analysis (see, e.g., Ferrando & Lorenzo-Seva, 2017)....
[...]
...In both cases a two-factor solution was fitted by using robust unweighted least squares estimation as implemented in FACTOR....
[...]
TL;DR: The results of this study present a trade‐off between a potential acquiescence bias when items are positively worded and a potential different understanding when combining regular and reversed items in the same test.
Abstract: This
research was funded by the Spanish Association of Methodology
of Behavioral Sciences and Health (AEMCCO), member of the
European Association of Methodology (EAM), and by the FPI
programme from the Ministry of Economy and Competitiveness
of the Government of Spain (PSI2014-56114-P, BES2012-053488,
and PSI2017-85724-P).
189 citations
Additional excerpts
...The data were analyzed with SPSS 20 (IBM, 2011), FACTOR 9.2 (Lorenzo-Seva & Ferrando, 2013; Ferrando & Lorenzo-Seva, 2017), TAP 12 (Brooks & Johanson, 2003), MPLUS 7.3 (Muthén & Muthén, 2012), FlexMIRT 2 (Cai, 2013) and ResidPlots-2 (Liang, Han, & Hambleton, 2009)....
[...]
University of La Rioja1, University of Valencia2, University College London3, Chinese Academy of Sciences4, University of Hawaii at Manoa5, University of British Columbia6, University of Wollongong7, University of Otago8, University of North Carolina at Greensboro9, Autonomous University of Barcelona10, Louisiana State University11, University of Pennsylvania12, Lenox Hill Hospital13, Georgia State University14, Ohio University15, University of Oviedo16, University of La Laguna17, University of Crete18, Southern Methodist University19, University of Bergen20, Macquarie University21, University of Western Australia22, Anglia Ruskin University23, University of Vienna24
TL;DR: The validity and utility of the SPQ-B, a brief and easy tool for assessing self-reported schizotypal traits, in cross-national research are supported, and theoretical and clinical implications for diagnostic systems, psychosis models, and cross- national mental health strategies are derived.
56 citations
References
More filters
TL;DR: Principles for reporting analyses using structural equation modeling are reviewed, and it is recommended that every report give a detailed justification of the model used, along with plausible alternatives and an account of identifiability.
Abstract: Principles for reporting analyses using structural equation modeling are reviewed, with the goal of supplying readers with complete and accurate information. It is recommended that every report give a detailed justification of the model used, along with plausible alternatives and an account of identifiability. Nonnormality and missing data problems should also be addressed. A complete set of parameters and their standard errors is desirable, and it will often be convenient to supply the correlation matrix and discrepancies, as well as goodness-of-fit indices, so that readers can exercise independent critical judgment. A survey of fairly representative studies compares recent practice with the principles of reporting recommended here. Structural equation modeling (SEM), also known as path analysis with latent variables, is now a regularly used method for representing dependency (arguably “causal”) relations in multivariate data in the behavioral and social sciences. Following the seminal
3,834 citations
TL;DR: Bifactor latent structures were introduced over 70 years ago, but only recently has bifactor modeling been rediscovered as an effective approach to modeling construct-relevant multidimensionality in a set of ordered categorical item responses.
Abstract: Bifactor latent structures were introduced over 70 years ago, but only recently has bifactor modeling been rediscovered as an effective approach to modeling construct-relevant multidimensionality in a set of ordered categorical item responses. I begin by describing the Schmid-Leiman bifactor procedure (Schmid & Leiman, 1957), and highlight its relations with correlated-factors and second-order exploratory factor models. After describing limitations of the Schmid-Leiman, two newer methods of exploratory bifactor modeling are considered, namely, analytic bifactor (Jennrich & Bentler, 2011) and target bifactor rotations (Reise, Moore, & Maydeu-Olivares, 2011). In section two, I discuss limited and full-information estimation approaches to confirmatory bifactor models that have emerged from the item response theory and factor analysis traditions, respectively. Comparison of the confirmatory bifactor model to alternative nested confirmatory models and establishing parameter invariance for the general factor also are discussed. In the final section, important applications of bifactor models are reviewed. These applications demonstrate that bifactor modeling potentially provides a solid foundation for conceptualizing psychological constructs, constructing measures, and evaluating a measure's psychometric properties. However, some applications of the bifactor model may be limited due to its restrictive assumptions.
1,508 citations
TL;DR: ESEM, an overarching integration of the best aspects of CFA/SEM and traditional EFA, provides confirmatory tests of a priori factor structures, relations between latent factors and multigroup/multioccasion tests of full (mean structure) measurement invariance.
Abstract: Exploratory factor analysis (EFA) and confirmatory factor analysis (CFA), path analysis, and structural equation modeling (SEM) have long histories in clinical research. Although CFA has largely superseded EFA, CFAs of multidimensional constructs typically fail to meet standards of good measurement: goodness of fit, measurement invariance, lack of differential item functioning, and well-differentiated factors in support of discriminant validity. Part of the problem is undue reliance on overly restrictive CFAs in which each item loads on only one factor. Exploratory SEM (ESEM), an overarching integration of the best aspects of CFA/SEM and traditional EFA, provides confirmatory tests of a priori factor structures, relations between latent factors and multigroup/multioccasion tests of full (mean structure) measurement invariance. It incorporates all combinations of CFA factors, ESEM factors, covariates, grouping/multiple-indicator multiple-cause (MIMIC) variables, latent growth, and complex structures that typically have required CFA/SEM. ESEM has broad applicability to clinical studies that are not appropriately addressed either by traditional EFA or CFA/SEM.
1,052 citations
TL;DR: In this paper, the authors considered the most appropriate parallel analysis procedure to assess the number of common factors underlying ordered polytomously scored variables, and proposed minimum rank factor analysis (MRFA) as an extraction method, rather than the currently applied principal component analysis (PCA) and principal axes factoring.
Abstract: Parallel analysis (PA) is an often-recommended approach for assessment of the dimensionality of a variable set. PA is known in different variants, which may yield different dimensionality indications. In this article, the authors considered the most appropriate PA procedure to assess the number of common factors underlying ordered polytomously scored variables. They proposed minimum rank factor analysis (MRFA) as an extraction method, rather than the currently applied principal component analysis (PCA) and principal axes factoring. A simulation study, based on data with major and minor factors, showed that all procedures consistently point at the number of major common factors. A polychoric-based PA slightly outperformed a Pearson-based PA, but convergence problems may hamper its empirical application. In empirical practice, PA-MRFA with a 95% threshold based on polychoric correlations or, in case of nonconvergence, Pearson correlations with mean thresholds appear to be a good choice for identification of the number of common factors. PA-MRFA is a common-factor-based method and performed best in the simulation experiment. PA based on PCA with a 95% threshold is second best, as this method showed good performances in the empirically relevant conditions of the simulation experiment.
971 citations
TL;DR: The role of latent variables in multiple regression, probit and logistic regression, factor analysis, latent curve models, item response theory, latent class analysis, and structural equation models is reviewed.
Abstract: ▪ Abstract The paper discusses the use of latent variables in psychology and social science research. Local independence, expected value true scores, and nondeterministic functions of observed variables are three types of definitions for latent variables. These definitions are reviewed and an alternative “sample realizations” definition is presented. Another section briefly describes identification, latent variable indeterminancy, and other properties common to models with latent variables. The paper then reviews the role of latent variables in multiple regression, probit and logistic regression, factor analysis, latent curve models, item response theory, latent class analysis, and structural equation models. Though these application areas are diverse, the paper highlights the similarities as well as the differences in the manner in which the latent variables are defined and used. It concludes with an evaluation of the different definitions of latent variables and their properties.
883 citations