Publication Date


Advisor(s) - Committee Chair

Doris Redfield, Carl Martray, Daniel Roenker

Degree Program

Department of Psychology

Degree Type

Master of Arts


This study was designed to develop an expressive language scoring system so that the creative quality of a person's written language could be quantified and measured. Sixty fifth -grade students, 31 males and 29 females, were given the opening line, i.e., story-starter, to a story and were asked to finish the stories. Each student received two story-starters concerning the same object ("content")-- either box, string or money--but were different in "context" (usual or unusual setting). Thus, a total of 120 stories were written, with 40 stories being from each content group. These stories were then rated according to their level of creativity by ten teachers having fourth, fifth or sixth grade teaching experience. These teachers were unfamiliar with the students who wrote the stories. The teachers were divided into two groups of five, and each group rated 60 stories on a seven point scale. The stories were divided equally so that each group rated one story from each child and so that the content and context of the stories were balanced. After the teachers rated the stories, they were asked to list the criteria they used for rating the stories. In addition to the teacher ratings of creativity, each student who wrote a pair of stories was rated on general creativity by his/her classroom teacher. The teachers' lists of criteria, along with past research by Guilford (1968) and Torrance (1974), helped determine what to include in the present experimenter-developed scoring system. There were seven sub-factors used, viz., ideational fluency, associational fluency, elaboration, relevant flexibility, irrelevant flexibility, originality, and organization. The subscale scores were combined to yield a total score for each story. The relationship among the total scores, sub-factor scores, teacher ratings of story creativity and classroom teacher ratings of student creativity was analyzed using regression analyses. Results showed that the total score was the best predictor for teacher ratings. The correlation between total score and teacher ratings of story creativity was .67. This suggests that the experimenter-developed scoring system validly reflects teacher ratings of story creativity. The correlation between the scoring system and classroom teacher ratings of student creativity as .36. The low correlation between the scoring system and classroom teacher ratings could be due to biases from knowing the students and other factors which could have influenced their judgments. Another possibility is the indication that teachers are, in general, poor judges of creativity.


Child Psychology | Psychology | School Psychology | Social and Behavioral Sciences