Dimension Reduction - Diagnostic - Quality extension

From Q
Jump to: navigation, search

Computes measures of the quality of a correspondence analysis of a tablecorrespondence analysis of a table.


The table below shows that the first 2 dimensions of a correspondence analysis represent most brands very well, with the exception of Diet Coke.

Technical details

The percentages across the top of the table show the proportion of variance explained by each dimension.

The row labels show the quality score for each row and column category. That is, the proportion of variance explained by the first two dimensions for each row and column category.

The cells of the table show breakdown of the quality score by dimension. The first two columns sum to the quality score.


includeWeb("QScript R Output Functions");

var is_displayr = (!!Q.isOnTheWeb && Q.isOnTheWeb());
if (!is_displayr)
    if (Q.fileFormatVersion() >= 17.13)
        alert("Please update Q to use this feature from the extension button, or run it from the menu via Create > Dimension Reduction > Diagnostic > Quality.");

function main() {

    // The following 2 variables contain information specific to this diagnostic.
    var required_class = "CorrespondenceAnalysis";
    var output_name_suffix = "quality";
    var item = checkSelectedItemClass(required_class);
    if (item == null)
        return false;
    var r_name = stringToRName(item.referenceName);

    // The following lines contain the R code to run
    var expression = "CAQuality(" + r_name + ")"

    return createROutput(item, expression, output_name_suffix);