The purpose of this study is to attempt to partially solve the 2-sigma problem by emphasizing the feedback-corrective procedure of mastery learning. This variable was studied in combination with enhanced student's prerequisite learning as well as with using organizers. The dependent variables are learning outcome : cognitive achievement. The subject for this study was the arithmetic in the first semester of 5th grade. The sample for this study consisted of 235 fifth grade students. These students were randomly assigned to one of five arithmetic classes, each of which made use of different instructional condition. These classes were as follows : Class 1, or Group C. Used conventional methods of instruction. This class was the control group. Class 2, or Group C+E. Used conventional instruction and enhancement of students' prerequiste learning prior to the instruction. Class 3, or Group C+O. Used conventional instruction and a set of organizers as a supplement to the textbook. Class 4, or Group ML. Used the feedback -corrective procedures of mastery learning. Class 5, or Group ML+E+O. Used the feedback-corrective procedures of mastery learning, enhancement of students' prerequisite learning, and the same organizers as in group C+O. The major findings of this study were as follows : First, the use of organizers was found to have a positive effect on student summative achievement. The mean achievement level for students in group C+O on the final summative test was significantly higher than the mean level of achievement of the students in the control group, Group C, at the 0.05 level of significance. In addition, the effect size of the use of organizers many times in the process of learning was larger than the effect size of the use advance organizer alone at the beginning of the learning task. Second, the size of effect of the combined use of mastery learning, organizers, enhancement of prerequisite learning on achievement(1.68) was much larger than that of other settings(ML, 1.13; C+O, .73; C+E, .52) This combined effect was larger than that of Avalos's study, which has been found to be about .85(Avalos, 1986), and also larger than the average effect of mastery learning on students' achievement in mathmatics, which has been found to be .89(Guskcy & Gates,l985) This result suggests the possibility of additive effects of the three strategies on students' achievement when these strategies are used together. An important consequence of these findings in relation to the 2- sigma problem is that organizers, enhancement of prerequisite learnings, and mastery learning added much to each other to improve students' achievement, when these strategies are used together Although 1.68 is a very large effect size, the combined use of organizers, enhancement of prerequisite learning, and mastery learning seemed to provide a solution to the 2-sigma problem. Third, the strongest relationship between measures of prior achievement and subsequent achievement was found in the conventional group, where the multiple correlation reached .92. The weakest relationship, on the other hand, was found under the ML + E +0 settings of this study, where the multiple correlation was .58. In the intermediate settings, the multiple con elation was .65 for the mastery learning groups, 42 for the organizer groups, and .77 for the enhanced prerequisite learning settings. These results show that the improved learning conditions provided to the students(groups ML+E+O, C+O, ML, C+E) reduce the predictive value of students' prior achievement on achievement over a series of learning task as well as on the summative achievement.