Understanding and evaluating English Learners’ oral reading with miscue analysis

Keh, M.L. (2017). Understanding and evaluating English Learners’ oral reading with miscue analysis. Journal of Adolescent & Adult Literacy, 60(6), 643-653.


Miscue analysis procedures can help teachers better understand the reading processes of English Learners, and this article provides helpful specifics and examples. The article caught my eye because I have often used miscue analysis procedures in my own practice when I needed in-depth, qualitative data on students’ reading that the usual assessment tools didn’t provide. I’ve also appreciated how miscue analysis focused on the strengths of readers, helping me see what those readers brought to the reading process that I could build on. Miscue analysis has often helped me understand and appreciate the thought processes of readers who puzzled me, and it also can help students better understand themselves as readers.

Yeh’s work with English Learners demonstrated how all of the benefits I’ve seen from miscue analysis can help us better understand the reading of a group that we may be puzzled about: English Learners. Gaining understanding is important, because reading assessments that only show differences between what English Learners read and what is in the text may underestimate those readers’ strengths and abilities. Miscue analysis can show us what those differences are, but also can show us when those differences matter when it comes to comprehension, and when they don’t.

Yeh worked with eight high school students who were English Learners. These students came from Bulgaria, China, Mongolia, and Japan, and they represented a range of English proficiency. We are not given information on socioeconomic status, and the students participating were volunteers, and there may be factors that affected the results, so they must be interpreted with caution. Even so, it’s clear that not only did Yeh gain an increased awareness of the students’ reading processes, but (perhaps more importantly) these eight students understood their own reading processes better.

Retrospective Miscue Analysis procedures were used with the students. In this version of miscue analysis, miscues are discussed with the reader after a text has been read orally and coded using the standard miscue analysis procedures. That is, individual miscues have been coded for Graphic Similarity (how much the miscue looks like the printed text), and each sentence (with miscues) has been coded for Syntactic Acceptability (Does it sound like language?), Semantic Acceptability (Does it make sense?), and Meaning Change.

Here, examples of the conversations with students about their miscues are provided, as well as a list of sample questions (Table 1 on page 645) used as prompts for those conversations. These examples clearly show the thinking, and the strengths, of these readers. Yeh discovered five themes in the data. By far the most commonly occurring theme was that students would miscue on words that they had limited, or partial, vocabulary knowledge about. However, students who might not know how to pronounce a word might still have some understanding of the word’s meaning. Assessments that only look at how words are pronounced during oral reading would not capture that understanding.

A second theme involved a pattern where differences between the phonology of English and the phonology of the English Learners’ languages accounted for deviations from the printed text (for example, reading what sounds like “smelled” for the printed English word “smiled”), but may or may not have affected meaning construction. Certain sounds in English may be difficult for speakers of other languages to perceive and/or produce. The procedures used in Retrospective Miscue Analysis helped Yeh to assess when a miscue was a problem and when it reflected a language difference but not a meaning change.

A third theme that surfaced was when these English Learners omitted “grammatical morphemes” (p. 649) from words as they read orally. For example, a plural suffix like –s or a tense suffix like –ed might be omitted. Typical quantitative assessments would count these omissions as equally problematic with other miscues, yet they may not always indicate a problem with comprehending the text. Rather, they show where the reader is in developing English proficiency. Grammar does need to be addressed as its own issue, but it does not necessarily mean a reader is not comprehending a text.

A fourth theme related to attention; some English Learners reported fatigue while trying to read the English texts. It is understandable that increased mental effort required by reading aloud in a new language might sometimes produce miscues. A fifth theme showed how past instruction might produce miscues. For example, one student often inserted “the” unnecessarily; in the conversation after reading she volunteered that her English teachers in China constantly reminded them to put “the” in sentences.

I was struck by how much I could learn about the languages of English Learners through miscue analysis, and I plan to investigate further. I was also struck by how dangerous it is to rely only on quantitative reading assessments, particularly oral reading assessments. Just counting deviations from the printed text, and failing to look at the nature of those deviations, and whether or not they impact meaning construction, can lead to a “deficit” mindset that devalues readers and the strengths they bring to the reading process. Miscue analysis can be a valuable tool to help us find and utilize all learners’ strengths.