Methods: The study is descriptive and methodological type with aim of  psychometric Early Learning Observation and Rating Scale. The first population of study was all children of 4 to 6 years in kindergartens and preschools in Bojnourd, 400 children's were selected through cluster random sampling method. The second population of the research included the first grade students of Bojnord city. Using purposefull sampling, 20 first-grade students referred to learning disabilities centers and 20 first grade students without learning problems were selected. It was used tools of the Early Learning Observation and Rating Scale, Denver Developmental Screening Test-II, and the Raven intelligence test to collect data. It was evaluated convergent and divergent validity, construct validity (exploratory factor analysis), diagnostic validity and internal reliability of the scale.

Results:The Cronbach's alpha reliability of the Early Learning Observation and Rating subscales was in the range of 0.86 to 0.92. Exploratory factor analysis confirmed the existence of seven factors in Early learning observation and rating scale. Early Learning Observation and Rating Scale were correlated positively with the Denver test(r=0.70) and negatively with Raven's IQ test(r=-0.30) (P<0.01) indicating a convergent validity and divergent validity confirmation. Also, the first-grade students with specific learning problems had significantly high scores in the subscales of Early Learning Observation and Rating Scale (P<0.0001) indicating a diagnostic validity.