COMPARISON OF 2 INDEXES OF ANSWER COPYING AND DEVELOPMENT OF A SPLICED INDEX

Two indices of answer copying, one using only identical wrong response s and the other using both right and wrong, were compared using natura lly occurring test responses (no simulation of copying). Each index po inted to the existence of about the same number of examinee pairs who had copied. However index B implicated more higher scoring pairs than the g(2) index, which, in turn, implicated more lower scoring pairs. A n adaptation of g(2) using only wrong-answer correspondences was found to identify even more pairs of highly likely copiers than B among hig her scoring examinees. A spliced index, the regular g(2) for lower sco ring pairs and g(2) based on identical wrong responses for higher scor ing pairs, was found to identify substantially more pairs of likely co piers than any single index. The methodological approach adopted made it possible to evaluate the indices without knowing which pairs of exa minees actually copied.