Elementary education is critical as it lays the foundation for learning, critical thinking, social and emotional development, individual growth, and self-confidence. Hence, studying the elementary students' educational progress is immensely important. This manuscript aims to investigate the factors that impact elementary students' academic performance, predict their academic performance, and use the above factors to identify the student's appropriate skilled level group based on their academic performance. In this study, the Dynamic Indicators of Basic Early Literacy Skills (DIBELS) score, a measurement to assess students' early literacy skills for K-6 graders, is used to quantify students' academic performance. A sample of 185 first and second graders and their features, including student's BMI (body mass index), amount of time the student spends watching television, student's gender, number of hours that a student sleeps each night, amount of time that the student spends reading books, student's race, and amount of time that the student spends on physical activity are used. Based on the regression analysis, second graders' academic performances are significantly impacted by their BMI values ([beta] = - 4.002, p < 0.05) and the amount of time students spend reading books ([beta] = 29.14, p < 0.05). The first graders' academic performances are significantly impacted by the amount of sleeping time ([beta] = 41.89, p < 0.01) and their gender ([beta] = - 37.129, p < 0.05). Furthermore, the experimental findings indicate that machine-learning techniques accurately predict the students' appropriate academic group. In the naive Byers classifier, students in the lowest academic performance group can be identified successfully with a sensitivity of 92%, and the students in the highest academic group can be identified with a specificity value of 100%.