The quality of education as reflected in academic achievementrncan not be divorced from the complex set of social andrneducational factors which interplay to affect learningrnprocesses and outcomes. The qu e stion "what makes a differencernin academic performance?" has been addressed by differentrnresearchers from both the developed and the developingrncountries. Within the developed countries much is known aboutrnthe determinant-achievement relationships. In the developingrncountries, however, the findings are inconclusive andrninconsistent.rnThe purpose of the study was to examine the extent to whichrnvariables used in the Educational Production Function and thernProcess-Product research approaches explain differences inrngrade five and six pupils ' English achievement. Specifically,rnthe study attempted to investigate the relative impact onrnpupils' achievement of seven home, six pupil, six schoolclassroomrnand eleven teacher related variables.rnThis task was accompl ished by regressing thirty independentrnvariables for each grade level. The dependent variable used asrna criterion measure was achievement test prepared by therninvestigator based on the objectives and contents determined byrnthe Ministry of Education for the respective grade levels. Thernthirty independent variables were selected on the basis ofrnrevi ew of literature. Information on the independent variablesrnwas secured using three basic tools of data collection:rnQuestionnnaire. school and classroom inventory. and classroomrnevaluation rating scale.rnThe r e lati onship b e tween the thirty ind ependent variables andrnpu p ils ' a c hi evement at each grade l evel was analysed by thernme thod of multipl e r egression. A variable ' s statistical impactrnon the d epend ent va riabl e was a c cept ed as significant at 0.05rna level or better.rnResults of the analysis revealed that variables describingrnhome. pupil. school-classroom and teacher attributes haverndifferential impacts on stud ents ' English achievement~ somernvar iables (such as class repeti tion, father ' s education andrnplanning skill) show ed statistically significant impacts whilernothers ( s u ch as pupl sex, cla ss - ize, and family-size) exertedrnnonsignificant influence. Based on these findings some basicrnco n c lu~i o ns ar e r each ed and r e lat ed r ecomm endations forwarded