Learning disability identification consistency: The impact of methodology and student evaluation data

Learning disability (LD) identification has long been controversial and has undergone substantive reform. This study examined the consistency of school psychologists' LD identification decisions across three identification methods and across student evaluation data conclusiveness levels. Data were collected from 376 practicing school psychologists from 22 states. Eighty-three percent (n = 313) of participants were female. Ninety-one percent (n = 342) of participants identified as Caucasian, 4% (n = 15) Latino, 1.3% (n = 5) African American, .8% (n = 3) Asian/Pacific Islander, .3% (n = 1) Native American/Alaskan Native, and 1.3% (n = 5) 2 or more races. Participants were randomly assigned to 1 of 9 conditions and used 1 type of identification method and examined 1 type of student evaluation data to determine if a student should be identified with LD. Results showed that overall identification consistency was somewhat low (73.7%, κ = .45) There were no differences in identification consistency across identification methods X²(2, N = 376) = 3.78, p = .151, but there were differences in identification consistency across conclusiveness levels of student evaluation data X²(2, N = 376) = 50.40, p = .0001. Implications for practice, training, and research are also discussed, including the need of school psychologists to consider psychometric issues in LD identification as well as the need to further research the impact of student data conclusiveness in LD identification.