This tutorial will help you compute and interpret RV coefficients used to test the association between two matrices, in Excel using XLSTAT.
Not sure this is the statistical test you are looking for? Check out this guide.
Included inXLSTAT-Base XLSTAT-Sensory XLSTAT-Marketing XLSTAT-Forecast XLSTAT-Biomed XLSTAT-Ecology XLSTAT-Psy XLSTAT-Quality XLSTAT-Premium
Dataset to run an RV coefficient test in XLSTAT
An Excel sheet containing both the data and the results for use in this tutorial can be downloaded by clicking here.
The data used in this example correspond to a study where 12 cranberry juices have been evaluated by two sensory panels. Each panel evaluated the product with their own lexicon. Panel FR used 6 attributes and panel CA used 7 attributes. Data were analyzed by mean of PCA.
Goal of this tutorial
The aim of this tutorial is to evaluate how similar the information gathered by the two PCAs are.
Setting up an RV coefficient test in XLSTAT
After opening XLSTAT, click on Test a hypothesis / Correlation/Association tests / RV coefficients.
In the General tab, select the first quantitative dataset (only the variables) in the Matrix A data field. Then, in the Matrix B data field, select the second dataset.
In the Options tab, select the User defined option for the Selected Variables and enter 1 in the from field and 5 in the to field.
Click the OK button and the Continue button in the following dialog box.
Interpreting the results of a RV coefficient test in XLSTAT
A table displays the RV coefficients for the series of dimensions chosen, i.e. the first variable, the two first variables, the three first variables, etc. Together with the RV coefficients XLSTAT displays the adjusted RV coefficients, the standardized RV coefficients, the mean and the variance of the RV coefficient distribution and the corresponding p-values.
The closer the RV coefficient to 1, the more similar the matrices. The lower the p-value, the more different from 0 the RV coefficient.
A bar chart allow to visualize the RV coefficients. The color code allows to quickly check if the RV coefficients are significantly different from 0 or not and at what significance level (0.001, 0.01 or 0.05).