Effects of blind guessing on the success of passing true-false and multiple-choice tests are investigated under a stochastic binomial model. Critical values of guessing are thresholds which signify when the effect of guessing is negligible. By checking a table of critical values assembled in this paper, one can make a decision with 95% confidence as to whether the correction for guessing is necessary for grading a true-false or multiple-choice test. This useful tabulation also represents an intermediate step toward development of a more comprehensive model for non-blind guessing in Bayesian statistics. (Contains 18 references and 2 tables.) (Author)