Overall performance in written English in Kenya Certificate of Secondary Education (KCSE) examination has been fluctuating below 50% between 1998 to 2012. Empirical evidence shows that mastery learning (ML) pedagogy is more effective than traditional pedagogy (TP) in instruction. Thus the effect of ML pedagogy on achievement in writing in English, in Nyandarua County was examined. This was achieved by comparing the effects of ML pedagogy on achievement to those of TP. The target population was all the Form One students in the location. Form ones were chosen to exercise control on the TP of their secondary schools to which they were not habituated, being new in the secondary level of education. The target population was a total of 5,779 students consisting of 2,702 girls and 3,077 boys. The stratified sampling technique was used, to ensure that all subgroups of the population were represented. In this way, a sample of 989 subjects was drawn, including 430 girls and 559 boys. This study used quasi – experimental research employing Solomon Three – Group Design. A Composition Achievement Test (COMPAT) and a ML pedagogy module for teachers were used for data collection. These instruments were developed by the researcher. Pilot testing with respect to the COMPAT was done for reliability of the instrument. Pilot testing yielded .9 Cronbach`s alpha with respect to reliability. Content and construct validity were ensured through systematic construction of the research instruments. Validity of the instruments was further established by five Egerton University test experts. Data was analysed using the Statistical Package for Social Sciences (SPSS) version 17.0 for Windows. One way ANOVA, Scheffe post hoc test and the t-test were the inferential statistics that were used. All levels of significance were fixed at = .05 level. The bases on which comparison was made included undifferentiated wholesale grouping. The null hypothesis was not accepted. The students taught English composition through ML pedagogy had a COMPAT achievement that was statistically significantly higher than that of the students taught through TP throughout the investigation. This study has findings that indicate possible benefits for secondary school teachers of English and their students, the research community, curriculum development institutes and university departments of English language education. The main recommendation put forth was that ML pedagogy should be used in the teaching of English language composition writing in Kenya