The purpose of this research is to find the most reasonable originality grading system by analyzing
various originality scoring mehtods of Wallach & Kogan Creativity Test. Hence, this study evaluated
reliability and validity of the 4 most commonly used originality scoring methods. Discriminant analysis
was operated to find out which of the 4 originality scoring methods predicts better the creative and
non-creative groups of children. The subject of this study consists of 4, 5, and 6 grade elementary school
students in Incheon. The summary of the result is as follows. First, method 3 (1 point to responses less
than 5%, and one additional point to those less than 1%) showed the most reliability. Second, the
correlations between other creativity tests and the 4 different originality scores suggest that three following
originality scoring methods show correlation on a similar level: method 1 (less than 5% rated responses
were considered to be creative and were given one point), method 2 (responses rated less than 5% were
given a weighting on a scale of 1~3 as follows: 2.5~5% - 1 point, 1~2.5% - 2 points, 0~1% - 3
points.) and method 3 (1 point to responses less than 5%, and one additional point to those less than
1%). However, method 4 (dividing up the originality score on method 1 with fluency score) appeared to
be inappropriate to be used for originality standard. Third, on discriminant analysis that speculates
creativity group (TTCT highest 20% and lowest 20%), in order of method 2, 1, 3, 4 predicted the
creativity group, and creativity group prediction rate adopting method 1 and 2 were over 70%, which
was relatively high. Based on these results, this research aims to reveal the significance and the current
limits of various originality scoring methods.