It adopted quasi- experimental approach. A multistage sampling technique was used to draw a sample of200 basic six (6) pupils from four schools. The instrument used for the data collection was Basic Science achievement Test (BSAT). It was developed by the researchers. Trial test of the instrument was done using Kuder Richardson 21. The reliability coefficient is 0.80. The result of the study shows that the bilingual mode of instruction had significant effect on Basic Science F (1.191) = 57.10, p< 0.05 Partial Eta Square = .230. Therefore the effect size (23%) is moderate. The bilingual mode of instruction also has significant main effect on gender in achievement of pupils in Basic Science F (1191)=9.76, p= .002 partial eta squared = .049 . However the effect size (4.9%) is low. Therefore, it was recommended that policy makers should include the use of bilingual mode of instruction in policy formulation for basic schools.
