![]() PCA.A standalone program that calculates the reliability of difference scores, given the reliability of two tests and the correlation between those two tests.Ī standalone program that calculates the Body Mass Index for adults given height and weight.Ī standalone program that computes the confidence intervals of obtained eigenvalues as per Larsen and Warne (2010).Ī standalone program for evidence-based practice that computes posterior probability when given prior probability and likelihood ratio.Ī standalone Macintosh program to compute the statistical significance of chi-square difference tests as typically used in comparisons of nested SEM models in EQS, LISREL, and Mplus SEM software as per Bryant and Satorra (2012).Ī standalone program to compute omega and omega hierarchical, measures of measurment precision or reliability, using standardized loadings from a confirmatory factor analysis. See Stat class for the general statistical methods and functions, RankAnalysis for matrix rank analysis and Cronbach for Cronbach's alpha analysis. PCA is a subclass of the superclass Scores. Not all of the methods descibed in this page form part of a conventional PCA but as the latter methods dominate the class title PCA has been retained for convenience and compatibility with earlier versions. See Brian Ripley's Principal Component Analysis and Factor Analysis for an explanation of the differences. PCA is closely related to Factor Analysis but they are not identical. The denominator of the varaiances, covariances, standard deviations, skewnesses and kurtoses may be set as n or as n−1. Statistics notes: The terms mean, standard deviation, variance and covariance used in this page and in the method names refer to the sample means, sample standard deviations, sample variances and sample covariances. the method standardizedAlpha() may also be spelt standardisedAlpha() and, similarly, Curtosis and Curtoses may replace Kurtosis and Kurtoses respectively. as ' standardised' or ' Standardised', e.g. Spelling note: All methods in this class containing ' standardized' or ' Standardized' in their name may be spelt with an s replacing the z, i.e. and whether the responding agency is human or inanimate. In this documentation responses or measurements will be referred to as responses or scores, questions or environmental changes as items and the responding agencies as persons irrespective of whether the data is scientific, sociological, psychometric, educational etc. ![]() 'No response' handling options: deletion, replacement (mean of the item, mean of the persons responses, total mean, zero, user supplied value) are available. The data may be entered from a text file or as a 2D array (String, double, float, int, char, boolean or Matrix) ![]() Extraction criteria based on eigenvalues greater than unity, greater than a Monte Carlo eigenvalue percentile or greater than the Monte Carlo eigenvalue means are available. A parallel analysis, using Monte Carlo simulations, is performed. Options are available for an analysis using either the covariance or the correlation martix. This class contains the methods necessary for a basic Principal Component Analysis with a varimax rotation. Michael Thomas Flanagan's Java Scientific Library Michael Thomas Flanagan's Java Scientific Library: Principal Component Analysis ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |