The purpose of this study was to examine the operational definitions applied to the actual diagnosis of special education subjects with learning disabilities in each provincial and regional special education support center. As of 2020, 32 special education support centers, which account for 16.2% of the total 197 special education support centers, participated in the study. Four types (1: LD request - LD selection, 2: request for other disability - LD selection, 3: LD request - rejection of special education subjects, 4: LD request - selection for other disabilities) were examined. First, type 1 was the most common (92 persons, 61.7%), followed by type 3 (32 persons, 21.5%), type 4 (14 persons, 9.4%), and type 2 (11 persons, 7.4%). Second, the difference in the ratio of the presence or absence of characteristics (difficulty) between the LD request-LD selection (Type 1) and Rejection (Type 3) groups was found to be a statistically significant difference in reading, writing, and mathematics academic domains. And in 6 domains, As a result of comparing the averages with the sum of having difficulties (1) as basic data, the group selected for type 4 other disabilities had the highest difficulty across all areas. At the time of rejection, intelligence and adaptive behavior tests were used a lot, and the trend of test results was different for each type, and the final selection as a subject for special education with a learning disability and the statement method for writing the reason for rejection were also without special criteria. Based on these major results, implications for clarification of decision-making criteria used in diagnosing eligibility for special education subjects with learning disabilities and consistency in application, and the need for follow-up studies were discussed.