The Application of the Canonical Correlation Analysis to the Psychological Data
Journal Title: International Journal of Innovation in Science and Mathematics - Year 2018, Vol 6, Issue 5
Abstract
Many applied behavioral researchers are not aware that there is a general linear model (GLM) that governs most classical univariate such as analysis of variance (ANOVA), regression and multivariate statistical methods. Accordingly, many persons view these statistical methods as separate entities rather than conceptualizing their distinct similarities within the GLM. Canonical correlation analysis (CCA) studies associations between two sets of random variables. CCA represents the highest level of the general linear model (GLM) and can be rather easily conceptualized as a method closely linked with the more widely understood Pearson correlation coefficients r. An understanding of CCA can lead to a more global appreciation of other univariate and multivariate methods in the GLM. Statistics anxiety notwithstanding, the GLM provides a framework for understanding all classical analyses in terms of the simple Pearson r correlation coefficient. The GLM can also be conceptualized as a hierarchal family, with CCA serving as the parent analysis. Contrary to the compartmentalized understanding of statistical methods held by many researchers, CCA subsumes both univariate and multivariate methods as special cases. The researchers demonstrate CCA with basic language, using technical terminology only when necessary for understanding and use of the method. The researchers use psychological data: personality test and intelligence quotient (I.Q.) results, using MANOVA procedure in Statistical Packages for Social Sciences (SPSS) version 14.0. The purpose of this article is to reduce potential statistical barriers and open doors to CCA for applied behavioral scientists and personality researchers.
Authors and Affiliations
Aslayn H. Datu-Dacula, et al.
Keywords
Related Articles
- EP ID EP507942
- DOI -
- Views 75
- Downloads 0
Matched Pairs Models
Matched pairs data can always arise from measuring a response at two variables on different level and it presented in square contingence table way, where the variables have the same category levels....
The Current Situation and Enlightenment of Korean Middle School Mathematics Education Under the New Curriculum Standard
During the time from China first proposed mathematics reform in 1958 to initially establishing a mature and completing mathematics curriculum system in 2001, we have experienced a lot of difficulties and setbacks in the...
Roman Fibonomial Numbers
In the beginning of 20thcentury, the generalization of binomial coefficient had been proposed by substituting arbitrary sequence in place of natural numbers. Using this, the Fibonomial coefficien...
Multi-hop HF Radio Propagation Model Based on Multi-Paths Attenuation
In this paper, we mainly solve the multi hop frequency radio propagation problem for radio propagation on the sea surface and establish model to solve. In high frequency radio communication, two paths are mainly consider...
Oscillating Pressure Gradient-driven Unidirectional Viscous Flow in a 2D Channel
A unidirectional viscous flow in a two-dimensional channel is driven by an oscillating pressure gradient. The channel walls are at y = 0, 2h and the pressure gradient is –rho with respect to x = G*cos(sigma*t). We determ...