The revised search strategies did not have an overall effect on the number of relative documents retrieved. That is, for the whole experiment 24 relevant documents were retrieved by the revised search strategies and 24 relevant documents were retrieved by the original search stragegies. However, the revised search strategies had an effect of reducing the number of nonrelevant documents retrieved from 139 to 60 or a reduction of 56.8. Another way of analyzing the effect of the search strategy revisions would be to state that under the revised search strategies 24 of 84 documents or 27.6 of the documents retrieved were relevant while under the original search strategies 24 of 163 documents or 17.3 of the documents retrieved were relevant. The revision of search strategies did have a satistically significant effect of increasing the percentage of relevant documents retrieved. In other words, the method of revising search strategies can be used to advantage in reducing the noise. This is particularly true if the system user is willing to miss one or two relevant documents to gain the advantage of not retrieving as many non-relevant documents. The interaction or dependency of the search strategy revision on the questions did not have a statistically significant effect on the percentage of relevant documents retrieved.