Teacher based is the usual instructional method used by most teachers in high school. Traditionally, teachers direct the learning and students work individually and assume a receptive role in their education. Student based learning approach is an instructional use of small groups of students working together to accomplish shared goals to increase everyone's and their own achievement. With the advert of progressive education in the 19th century and the influence of psychologists, educators have largely replaced teacher based curriculum approaches with "hands-on" activities and group work where the child determines on his own what he/she wants to do in class. The current study investigated the effect of "teacher" and "student based" instructions on probability achievement outcomes and attitudes of secondary school students in Bungoma North district, Kenya. It tries to explore the possibilities of exposing students to pedagogical practice which will actively engage them into the learning process and improve their performance in probability. The study will also inspire mathematics teachers to use appropriate teaching approaches that incorporates constructivist teaching characteristics that can enhance active participation of students and at the same time offer insight into patterns of student thought (understanding of mathematical concepts and processes) as well as their perception of, and attitudes towards solving and mathematics in general. Seymour Paper's theory of constructionist provides the frame work for this study. The study utilized the matching only pretest-posttest control group design. The experimental group (N = 100) used student based learning approach, while the control group (N = 100) used teacher based approach. The study targeted 660 Form three students. Attitude and probability scores were collected from (N = 200) students in five secondary schools. It was analyzed by descriptive and inferential statistics with the use of SPSS (Statistical Package for Social Sciences) guide and hypothesis tested by one- way analysis of variance (ANOVA). The analysis revealed that "student based" method of instruction enhanced students' attitudes toward probability and probability achievement.