Kaiser-Meyer-Olkin measure of sampling adequacy (KMO)

  • tests whether the partial correlations among variables are small
  • used to compare the magnitures of the observed correlation coefficients to the magnitures of the partial correlation coefficents
  • large values suggest that a of the variables is a good idea
  • range from 0 to 1
  • 0 means the sum of partial correlations is large compared to the sum of correlations – factor analysis will not yield reliable factors
  • .5 to .7 mediocre
    .7 to .8 is good
    .8 to .9 is great
    .9+ superb
  • Hutcheson & Sofroniou, 1999
  • if you have low values, you may wish to collect more data or examine the variables you want to include.

Was this helpful?

0 / 0

Leave a Reply 0

Your email address will not be published. Required fields are marked *