An assessment of the possible effect of mathematics symbol instruction on achievement in word problems was the main thrust of the study. Two null hypotheses were tested in the study. The non-equivalent group pre-testpost-test quasi-experimentalresearch design was adopted. The multistage sampling technique was used to select a sample of 387 participants from 6,740 senior secondary class one students in the 81 government-owned coeducational secondary schools in Calabar education zone of Cross River State. The treatment class was exposed to mathematics symbol instructional strategy for solving word problems while the control class was exposed to regular word problem teaching, otherwise called the conventional method. "Students mathematics achievement test" was the instrument used for data collection after validation by experts. The reliability coefficient of 0.90 was determined using the Kuder-Richarson 20 formula. Primary data were obtained after informed consent from the participants. The null hypotheses were all tested at the 0.05 alpha level using the analysis of covariance. Findings revealed a significant effect of the mathematics symbol instructional strategy on students' word problem achievement in secondary schools. Male students perform significantly better in word problems than their female counterparts when taught using the mathematics symbol instruction. These findings discussed fundamental theoretical, practical, and research implications. It was recommended, among other things, that mathematics teachers concentrate on demystifying the abstraction in mathematics by deliberately adopting mathematics symbol instructional strategy in interpreting mathematical statements and sentences into their symbolic notations before engaging in their computation.