This study empirically examined the effectiveness of using mathematical texts as a means of enhancing student literacy regarding negative numbers. A quasi-experimental two-group pretest-posttest design was adopted to recruit students from a public senior high school in New Taipei City. Specifically, 91 seventh-grade students from four classes were assigned to the experimental group, and 45 seventh-grade students from two other classes were assigned to the control group. Over a course of 4 weeks, the experimental group attended 5 math sessions per week, with 1 of the sessions dedicated to reading mathematical texts. By contrast, the control group continued with their routine learning activities using standard textbooks throughout the 4 weeks. This study developed the mathematical texts studied by the experimental group, the zhi-xíng-shì (knowledge-capacity-comprehension) evaluation criteria, and a pretest and posttest based on these criteria. Accordingly, the following two main findings were obtained: First, after learning with the mathematical texts, the experimental group demonstrated significant differences between their pretest and posttest scores in terms of negative number literacy. Specifically, their knowledge comprehension increased from 3.5% to 25.9%, and the error occurrence decreased from 11.8% to 2.4%. Second, after the pretest scores were controlled for, a multivariate analysis of covariance was conducted to assess the posttest scores in the knowledge, capacity, and comprehension dimensions. The results revealed that the experimental group's mean score for the comprehension dimension was significantly higher than that of the control group. However, their mean scores for the other two dimensions differed nonsignificantly from those of the control group. According to the experimental findings, suggestions are provided for the application of the developed mathematical texts in teaching.