6533b863fe1ef96bd12c78ba

RESEARCH PRODUCT

The Scree Test and the Number of Factors: a Dynamic Graphics Approach

Rubén Daniel LedesmaGuillermo MacbethPedro M. Valero-mora

subject

Linguistics and LanguagePsychometricsMachine learningcomputer.software_genreLanguage and LinguisticsCIENCIAS SOCIALESSCREE TESTData visualizationStatisticsComputer GraphicsHumansScreeDATA VISUALIZATIONGraphicsGeneral PsychologyPrincipal Component Analysisbusiness.industryPsicologíaExploratory factor analysisVisual inspectionRange (mathematics)FACTOR ANALYSISData Interpretation StatisticalPrincipal component analysisData analysisArtificial intelligenceFactor Analysis StatisticalPsychologybusinesscomputer

description

Exploratory Factor Analysis and Principal Component Analysis are two data analysis methods that are commonly used in psychological research. When applying these techniques, it is important to determine how many factors to retain. This decision is sometimes based on a visual inspection of the Scree plot. However, the Scree plot may at times be ambiguous and open to interpretation. This paper aims to explore a number of graphical and computational improvements to the Scree plot in order to make it more valid and informative. These enhancements are based on dynamic and interactive data visualization tools, and range from adding Parallel Analysis results to "linking" the Scree plot with other graphics, such as factor-loadings plots. To illustrate our proposed improvements, we introduce and describe an example based on real data on which a principal component analysis is appropriate. We hope to provide better graphical tools to help researchers determine the number of factors to retain. Fil: Ledesma, Ruben Daniel. Universidad Nacional de Mar del Plata; Argentina. Consejo Nacional de Investigaciones Científicas y Técnicas; Argentina Fil: Valero Mora, Pedro. Universidad de Valencia; España Fil: Macbeth, Guillermo Eduardo. Universidad Nacional de Entre Ríos; Argentina. Consejo Nacional de Investigaciones Científicas y Técnicas; Argentina

https://doi.org/10.1017/sjp.2015.13