English is a useful language, and millions endeavor to become skilled sufficiently to enable them to fulfill the educational prerequisites of their local governmental jurisdictions or participate in the local, regional, or world economy. Instruction is often based on the results of an assessment measure, such as the Common European Framework for Languages. Assessments are generally unidimensional, and their level designations, such as 1–5, and their acronyms, such as ESL, ELL, EAL, ESOL, and EFL, obscure the diversity within them. The problem of acronyms and level designations and their relationship to diversity will be explored. I will argue that the label ESL—or English-language learner (ELL) or whatever acronym is used—is problematic because it masks significant underlying differences that have serious consequences. Acronyms do not represent diversity because they are unidimensional measures of English that obscure inter- and intra-subgroup diversity, e.g., there were significant differences among and within Mandarin-, Cantonese-, Tagalog-, and Spanish-speaking sub-groups of Canadian immigrants. The importance of disaggregating categorical data to assess diversity will be described, including a discussion of related unpublished findings. Concluding remarks will outline instruction design employing diversity features that are informative.