本研究針對遙遠聯想測驗（Remote Associates Test, RAT）的缺點，提出一種新的測驗方式，稱為RAT-like評測工具，主要特色是將發散性思考融入RAT中。搭配Williams繪圖式創造性思考測驗，在對學童進行施測後，利用相關性分析和因子分析探討各發散性思考指標之間的內部關聯性，運用K-means群集分類法，研究不同群集在不同面向所呈現的差異與特性。研究表明，對於高創造力學童與低創造力學童，RAT-like測驗結果與Williams創造性思考測驗結果是一致的，但是對於那些中等表現學童，在獨創性方面的表現卻有分歧，此說明RAT-like測驗之內涵不同於一般Guilford-Style創造力測驗。此研究成果在探討創造性問題解決時，提供一種全新的途徑，可用於評量創造性思考能力。

This study presents information about a new creative thinking test based on a Remote Associates Test (RAT) approach. Up until now the lack of divergent-thinking components in the RAT was considered a major drawback. However these components are included in the RAT-like tests. In this study the RAT-like tests were compared to the Williams’ creative thinking test. Correlation and factor analyses were employed to study the relationships among the divergent thinking components of the two different tests. K-means clustering was also employed. The results suggest that the divergent thinking skills in the RAT-like test may be distinct from those in the Williams’ test and other Guilford style tests. The results on the RAT-like tests coincide with those on the Williams’ test for gifted and lower performing students but are different in terms of originality for students with intermediate levels of performance.