The desire of every teacher and every educator is to improve academic performance in schools and colleges. Meaningful teaching and learning requires flexibility, creativity and responsibility in order to provide a conducive learning environment able to respond to the learner's individual needs. A host of research has been conducted to investigate various factors that influence academic performance. Among such factors is the preferred learning styles of learners. A quasiexperimental pre-test post-test non-equivalent control group design was conducted to investigate how three learning styles (visual, auditory and kinaesthetic) and three teachers' instructional strategies (guided-inquiry, cooperative learning and direct instruction) predict academic performance (achievement test scores and acquisition of science process skills) in Physics practical work while controlling for gender. A sample of five hundred and nineteen (519) form three physics students were selected through multistage sampling procedures (purposive sampling, proportionate stratified random sampling and simple random sampling). Physics Achievement Tests 1 and 2 (PAT1 & PAT2), Learning Style Questionnaire and Physics Process Skills Checklist (PPSC).Teachers' Instructional Guides on Guided-Inquiry (TIGITS), Cooperative (TIGCTS) and Direct Instructional Strategy (TIGDITS) were used to collect quantitative data.. Validity and reliability of the instruments were assessed. Data were analysed using multiple Linear Regression test. Results indicated that while Teachers' Instructional Strategies contributed significantly to the model (B= 3.266, p<.05), Students' learning Styles did not (B= .717, p=.412). These findings have implications to educators, teachers, learners, and ministry of education stakeholders. Recommendations are provided.