Published: 12-14-2015 06:34 am
Updated: 12-14-2015 06:35 am
"lpsative measures are multiple measures, where the data are collected, or are modified, in such a way that all subject totals across the measures are equal. Much has been written about factor analysis with such data, however, no clear consensus has been reached regarding the suitability of ipsative measures for factor analysis. The purpose of the present article is to show analytically the fundamental problems that ipsative measures impose for factor analysis. The expected value of the correlation between ipsative measures is shown to equal - 1/ ( k - I), where k is the number of measures. The rank of the resulting correlation matrix is reduced by one to k - 1, and ipsativity alone produces k - 1 artifactual bipolar factors, which will obscure any actual interrelations among the measures. If the data are known to be ipsative or if the tell-tale signs of ipsativity are seen, factor analysis should not be done."
Despite that it's unrecommended to use multivariate analysis on ipsative data, this is what is done in MBTI.
Published: 03-04-2016 05:12 am
Updated: 03-04-2016 05:12 am
"We are left with recommending against the use of principal component, principal factor or maximum likelihood factor analysis with ipsative measures. The separation of artifactual bipolar factors induced by ipsativity from any true underlying relationship will be difficult at best, and not worth the danger of a largely incorrect interpretation. We also realize that in some instances measures may actually be ipsative, or be nearly ipsative, without the researcher's awareness. Ifipsativity is suspected, the researcher may perform principal components analysis followed by the rotation of k - 1 factors to look for the telltale sign of artifactual ipsative factors like those in Table 1. If the data are ipsative, factor analysis will not be interpretable."