1970 | The stacking effect and methods of correcting for it.

Occasionally it is practicable to use an ESP test in which all the Ss make their calls for the same set of targets instead of for individual sets. When a single set is used for the group, it is necessary to make a statistical correction to compensate for the possibility that the Ss may share the same calling bias and that this may coincide with (or deviate from) the target order and give spuriously high (or low) results (the “stacking effect”). 1 of the methods of correction was worked out by T. N. Greville. Other Es have allowed for the stacking effect by merely reducing their critical ratios by 10%, a simplification which lacks theoretical justification. 2 other methods evaluating these multiple-calling experiments are the majority-vote technique and the use of the “index of preference.”

(PsycINFO Database Record (c) 2017 APA, all rights reserved)