Random Dot E Stereogram-Reply

Abstract

In Reply. —We quite agree with Dr Schecter's observation on the matter of statistically significant end points. The use of Pascal's triangle or the formula (p + q) n brought us to the same conclusion— perhaps five correct answers out of six is too charitable. 1 Selection of pass/fail criteria when trying to establish an efficient, effective screening procedure for use on large populations while simultaneously minimizing false-positives and false-negatives is a process of balancing. As stated in our article, a disproportionate number of false-positives or false-negatives is undesirable and often decreases the effectiveness of the procedure. 2 However, increasing the number of trials in a procedure increases the time required to carry out the screening process. If time constraints are present, time also limits the number of individuals it is possible to screen. Increasing from six to eight trials is a 33% increase in time, or a 33% decrease in