Psychometric properties of a silent word reading test (LEO-1-min)

Abstract

Introduction. Teachers and researchers often need to evaluate word decoding skill in group-wise and in a short time. The LEO-1-min test is created to measure word reading through a lexical decision procedure where the examinee identifies pseudowords in a list of frequent words. Objetive. To examine the reliability and validity of LEO-1-min, a silent word reading test, suitable for quick assess of reading abilities in a wide age range of students. Method. Participants were 284 children from 1st to 6th grade of a subsidized Primary School. We created four alternate forms of the LEO-1-min, each with 180 stimuli (132 words and 48 pseudowords). Results. The results show an adequate parallel forms reliability of the scores (range rs = from .57 to .81). High correlations were found between the scores on the LEO-1-min and the scores on a standardized reading aloud test. The discriminant analysis of the scores on the LEO-1-min shows a high level of success in predicting the oral word decoding performance. Discussion and Conclusion. LEO-1-min reliability is acceptable to good. Lexical decision in LEO-1-min and oral reading are highly correlated, which support using lexical decision as a groupwise test to screen for poor word readers. Form A of the test and provisional scales are presented for each primary grade.