Correctly prepared semantic feature analysis charts can be used to concretize abstract mathematical concepts. This study, it is aimed to develop a valid and reliable rubric that can be used to evaluate the accuracy of the semantic feature analysis charts created by mathematics teachers. The data of the research consists of 36 semantic feature analysis charts created by mathematics teachers who have been trained in preparing a semantic feature analysis chart. During the validity study, the opinions of the experts about the rubric prepared were examined individually in the terms of content, structure and criteria. The percentages of concordance of expert opinions on sub-dimensions vary between 83% and 91%. The expert opinion concordance rate for the graded scoring key as a whole is found to be 87%. These calculated concordance percentages show that the rubric developed is valid. During the reliability study, the consistency between the scores evaluated for the semantic feature analysis charts by two mathematicians was calculated. The consistency of the raters in the sub-criteria was calculated with "Cohens's Kappa". A perfect fit in the sub-dimensions of instruction writing and text writing, and an acceptable fit in the sub-dimension of table design were observed. The concordance between the total scores of the two scorers was calculated with the "Spearman Correlation Coefficient" and a positive, significant and high level of concordance was found. These results show that the rubric is a valid and reliable tool in evaluating the semantic feature analysis charts used in maths lessons. The developed rubric can be used in other fields of education by making necessary changes.

Mathematics education, Semantic Feature Analysis, Performance evaluation, Rubric