An empirical test of the modified C index and SII, O*NET, and DHOC occupational code classifications

The present study investigated new approaches for assessing Holland's congruence hypothesis by (a) developing and applying four sets of decision rules for assigning Holland codes of varying lengths for purposes of computing Eggerth and Andrew's modified C index; (b) testing the modified C index computed using these four approaches against Brown and Gore's C index; and (c) comparing the validity of SII, O*NET, and DHOC taxonomies for assigning Holland codes to work environments. Results obtained from an archived sample of 239 employed young adults suggested that the modified C index performed as well as, but not better than, the standard C index. The SII was associated with larger mean congruence scores and stronger congruence-satisfaction relations than was the O*NET, which in turn was associated with larger mean congruence scores and stronger congruence-satisfaction relations than was the DHOC. Implications for research and practice related to Holland-type congruence are discussed.