In education, examining students' learning in detail, determining their strengths and weaknesses and giving effective feedback have gained importance over time. The aim of this study is to determine the distribution of students' answers to the reading comprehension achievement test items which were written at different cognitive levels and to investigate the affective variables that are effective in classifying students based on their incorrect, blank, and unrelated answers identified via rubric. For this purpose, a reading comprehension achievement test, a student information form, the perceived academic self-efficacy scale and the learned helplessness tendency scale were used to collect data. The student information form included perseverance, achievement motivation, exposure to bullying and test anxiety subscales. A rubric was used to determine the students' response categories. According to the findings of the study, the rate of blank and incorrect answers increases as the cognitive level of the items become more complex. While the most correct response rates are decreasing, partially-correct answers are increasing relatively. While students' learned helplessness tendencies were effective in classifying their blank and unrelated answers at the most basic reading comprehension level, as the cognitive process became more complex, the affective characteristics classifying the student responses increased in number. It was concluded that these variables are important in improving the students' answers and in leading them to the partially correct and the most correct answer. It can be suggested to create trainings and classroom environments that will equip and improve students' features about these variables.