This article develops an analytical framework for analysing college (tertiary) statistics textbooks in terms of text accessibility by integrating the text, the reader, and the content into the framework. Five accessibility attributes of science texts were adapted to conceptualize the accessibility of statistics texts. For each accessibility attribute, two components were proposed by referring to the literature on the readability of mathematics texts as well as the characteristics of statistics. The feasibility of the framework is demonstrated by analysing sample statistics texts. The contributions and potential of the framework are discussed.