This study aims to develop a scientific creativity task which science-gifted elementary students can conduct on a field trip to a botanical garden, and to analyze the results from conducting the task. For this, 38 science-gifted fifth-graders from the Science-Gifted Education Center, located at the Office of Education, participated in a field trip to a botanical garden, as a part of their program. Prior to the program, researchers developed a scientific creativity task for outdoor education program, along with science education specialists and teachers. The tasks were to observe plants, and to create something new and useful, or, in other words, scientifically creative, based on the plants' characteristics. The students could submit at most three ideas. Also, they assessed their own ideas, and selected an idea that they thought was the most creative. The results were analyzed by using the scientific creativity formula. The main findings from this study are as follows. First, it was found that the scientific creativity formula had an upward bias in assessing originality. Second, the students tended to assess the usefulness of their own ideas more generously. Third, the correlation between self-assessment results and scores from the scientific creativity formula for originality was r=.43. Fourth, in formula-based assessments, the correlation between originality scores and usefulness scores was relatively high, at r=.56. Fifth, the correlation between a student's scientific creativity score and the number of his or her ideas was very low, at r=.23. Sixth, when the ideas chosen as the most creative by students were compared with the ideas that had the highest scores in formula-based assessments, it was shown that 8 out of 19 students (42.1%) did not choose the idea that appeared to be the most creative when graded by the formula. This study is concluded by discussing the lessons from the scientific creativity task analysis for primary science education and gifted education.