This study was conducted to examine if the current identification process of reading difficulty is valid for 3rd and 4th graders and investigate their characteristics in reading comprehension using a teacher rating scale. A total of 54 students and their teachers were participated in this study. First, we examined whether students diagnosed with poor readers at the schools were consistent with the students at risk of reading disabilities based on the theory of the standardized reading comprehension test battery. For this, we compared the percentage of children diagnosed with reading disabilities by the offices of school district with the percentile score of the standardized reading achievement test. The results showed that the two groups did not exactly coincide with each other, but the students who were diagnosed with poor reading generally showed low scores in the areas of reading. Second, the students were divided into reading comprehension risk, average, upper class according to reading comprehension achievement, and then compared the difference in the reading comprehension characteristics, which are the reading comprehension sub-skills, by the teacher s rating. The Kruskal Wallis test showed that there was a significant difference between the groups in all the items of the five sub-domains of reading comprehension. As a result of comparing the average ranking of the groups, the reading risk group showed significantly lower ability than the average group or the higher group in all items. However, there was no significant difference between the average level group and the advanced level group in all the items except the three items, ie, ‘realistic comprehension’, ‘inferential comprehension’, ‘ability to summarize correctly according to structure of writing’.