Clark and Mirels reported a “correction procedure” for fluency in scoring figural divergent thinking measures, which resulted in a reduction from .40 to .20 in the average inter‐correlation of scores excluding fluency. Since their correction involved shortening the test, at least part of the reduction of the inter‐correlations can be explained by the reduced reliability of the test. Estimates of the magnitude of this effect were provided. It was concluded that the “correction” reported is at best a partial correction, and that the potential merit of other procedures should be considered.
|Original language||English (US)|
|Number of pages||2|
|Journal||Journal of Educational Measurement|
|State||Published - Jun 1971|