Samkvæmni í mati á Myndflötum og Völundarhúsum í WPPSI-RIs

Abstract

Four experienced psychologists (three males and one female) scored the same 60 forms on two performance subtests, Geometric Design and Mazes, of the Icelandic standardization version of WPPSI-R. Forms were randomly sampled from the standardization sample. Scoring was based on scoring rules given in the Icelandic handbook. Objectivity of the scoring rules was tested by evaluating interscorer reliability. To assess interscorer reliability a type of Intraclass correlation coefficient was used that takes account of differences in scorer leniency as well as random error. Interscorer reliability coefficients were 0,82 on Geometric Design and 0,87 on Mazes. These results indicate that the scoring rules for these two performance subtests are objective enough for different scorers to reach similar results when scoring the same responses.