Massive open online courses (MOOCs) are online courses for multiple learners with different backgrounds, including English-as-a-second-language (ESL) learners. In a MOOC, course concepts are important for diverse learners to grasp what they can learn in the course and its prerequisite knowledge. Previous studies have explored methods to automatically extract concepts from course videos or identify prerequisite concepts in a course. However, as a concept typically consists of several words, it could be difficult for ESL learners to understand what a concept means if they do not know the words in the concept. For example, for "geospatial data," many of them may need an additional explanation of what "geospatial" means in addition to the explanation of the concept. This paper extensively analyzes the readability of MOOC concepts using a openly-available manually-annotated MOOC-concept dataset on computer science and a vocabulary test result dataset of ESL learners with different English skills. We found that the percentage of concepts for which an ESL learner is likely to know all the words is only 25.8\%, implying that ESL learners usually require additional vocabulary explanation to understand MOOC concepts. We also show qualitative analyses and that almost half of the concepts are unreadable to ESL learners.