Academic Exchange Quarterly    Summer   2005    ISSN 1096-1453    Volume 9, Issue 2

To cite, use print source rather than  this on-line version which  may not  reflect print copy format requirements or   text lay-out and pagination.



Reflecting on Miscues in Content Area Readings


Michele Ebersole, University of Hawai`i at Hilo


Ebersole, Ph.D., is an Assistant Professor in the Education Department



This article describes a case study of a struggling middle school reader.  Different content area texts (science, math, literature) were selected and three oral reading miscue analysis cycles were conducted.  After each miscues analysis the reader reflected upon select miscues using Retrospective Miscue Analysis.  Miscue readings indicated some difference between the readings and RMA discussions showed increased knowledge about the reading process.   


“I don’t see myself as a good reader ‘cause I make a lot of mistakes.  When we read orally I’m shame and I read fast.”  - Kyle


Struggling middle school readers like Kyle (pseudonym) often view themselves as poor readers based on the number of “mistakes” made during reading.  Successful reading is perceived as reading words accurately.  Students like Kyle are embarrassed by their inability to read accurately and fluently, thus creating feelings of anxiety about their reading.  By the time they are in middle school these readers are overwhelmed by the reading demands required across content areas.


Kyle’s words hit deep as I had observed many students in my own teaching experience [a]“shame”[ed] to read because of their perception of themselves as readers.  As a result of my initial observations, I wanted to explore how struggling readers might learn to revalue their knowledge about reading and come to understand more about their successes.  I became interested in Retrospective Miscue Analysis (RMA) as a process which gives readers an opportunity to see their own miscues and build upon their strengths. 


RMA has been used with primary aged children (Martens, 1998) and intermediate, middle school, high school, and adult readers in helping them revalue the reading process and understand it as a process of constructing meaning (Y. Goodman, & Marek, 1996).  It has also been used to conduct reflective conversations with youth in juvenile correction centers (Moore & Aspengren, 2001).  In this inquiry I wanted to see what kind of miscues a struggling middle school reader makes when reading different types of texts encountered in school and how knowing about the quality of his miscues might influence his confidence and attitude toward reading.  The questions I explored were:  How do text features affect the patterns of miscues?  How do insights about these patterns increase a student’s understanding of the reading process?  Could RMA help a reader revalue his own reading process and perception of himself as a reader?  I hoped insights about readers struggling to process and make sense of content area texts would provide teachers with information to support content area literacy instruction.


In this article, I give a brief background on RMA and Kyle.  Then I explain how I used RMA with Kyle.  This case study looks at kinds of miscues (observed responses that differ from the expected response) Kyle made when reading different types of text and identifies his insights when he reflected upon the different miscues made during the readings.


Miscue Analysis and Retrospective Miscue Analysis

Miscue Analysis is a tool that provides teacher/researchers with increased understanding about the reading process (Y. Goodman, Watson, & Burke, 1987).  It is used to aid researchers in developing deeper insights into the reading process and it can help teachers analyze the oral reading of individual students.  This tool allows the researcher to qualitatively analyze why miscues are made and allows the researcher to interpret and understand what miscues reveal about the reading process. 


RMA is a follow up miscue procedure which provides the reader with information about the reading process and asks the reader to reflect upon select miscues made during the reading (Y. Goodman & Marek, 1996).  Developed by Chris Worsnop in the 1970s, RMA utilizes metacognition as it incorporates the ability to reflect upon one’s reading process. 


Earlier studies of content area miscues across science, social studies, math, and basal reader texts found no significant difference among miscues made across different texts, indicating the reading process is consistent across content areas (Kolczynski, 1978).  In another miscue study (K. Goodman & Y. Goodman, 1978) readers used culturally relevant materials with second language Samoans and Hawaiian pidgin (HCE) speakers in Hawai‘i.  The findings in this study revealed complex results depending on the nature of the text and grade level of the students in the study.  The researchers found that “the degree of relevance of the literature to the reader aids considerable in its predictability.  However, relevance to readers is a complex set of relationships” (Y. Goodman, 1982, p. 302).   Another study found that dialect speakers’ miscues do not influence ability to comprehend text (Eberwein, 1982).


Meeting Kyle

Driving up to this rural middle school of 600 students, I found myself wondering about our initial encounter:  What kind of reader is he?  How will he respond to me as a researcher and to RMA as a process?  I knew the school was designated Title I and labeled in corrective action which meant the school had not shown adequate yearly progress in standardized and state tests over the past four years.  Kyle’s teacher, Mrs. Smith (pseudonym,) was highly respected for her practice as a reading teacher.  At the time of the study she was working on a reading endorsement and was interested in miscue analysis as a classroom assessment tool. 


Mrs. Smith described Kyle as a struggling reader who was failing most of his classes.  She described Kyle as “charming,” yet he had a volatile temper and would sometimes explode in class.  He was easily distracted by others and did not turn in his work.  When Kyle learned there was a possibility he might fail his sixth grade year, he seemed to give up on school.  By the time I started working with him, in the fourth quarter, attendance had started to become a problem.  Mrs. Smith said she wanted me to work with him because, since he lacked confidence, he might need the one-to-one work.  She seemed to see me as her last effort to help this child. 


At our first meeting Kyle immediately started a conversation by asking me questions about who I was and what I wanted to do with him.  Within a few minutes of our conversation we established that we had mutual acquaintances and he seemed very comfortable talking about his experiences.  I asked Kyle questions using a modified Burke Reading Interview (Y. Goodman & Marek, 1996).  The interview revealed that Kyle viewed reading as “pronouncing the words correctly” and that he saw the road to improvement as “not making errors.”  He shared that he felt “pretty good” about reading in math and science because “I hardly make mistakes and the teacher usually reads to us.” 


Text Selection and Procedure

For the miscue analysis, textbooks Kyle used in social studies, science, math and reading were gathered.  The purpose for collecting current textbooks was to represent the range of books read by a typical middle school reader.   Due to time constraints only math, science and reading were used.  500 – 1000 word passages were selected and the readings were unfamiliar and unpracticed.  Each selection represented a complete short story or a complete lesson.  The Flesh-Kincaid Grade Level for the three texts were as follows:  Science 5.2, Math 4.5, and Literature 3.3. These books were used for average sixth graders at the school with exception of the literature text which was used by his reading teacher with struggling readers like Kyle.   


Miscue Analysis and RMA Procedures

Three miscue cycles were conducted, each followed by a RMA session within two weeks after the miscue reading.  The science text was read first, then literature, and lastly math.  A post interview followed these three miscue/RMA cycles.  The whole study took place between April and June. 


For each miscue cycle, Kyle read the entire selection from the original source without any assistance.  While the he read, every miscue was marked directly on a prepared typescript.  The miscue coding system was used to identify substitutions, omissions, insertions, repetitions, and corrections.  Each miscue was coded using the miscue analysis Procedure I coding form (Y. Goodman, Watson, & Burke, 1987).  Following each reading, Kyle retold what he recalled about the selection to assess reading comprehension.  Each reading was approximately 15 minutes in length and was audio-taped.


Each of Kyle’s miscues were qualitatively analyzed for syntactic, semantic, and graphophenemic acceptability by the researcher.  Select miscues were identified and discussed with the Kyle in a follow up RMA interview.  Questions guiding the RMA sessions were: Does the miscue make sense?  Should the miscue be corrected?  Why did you (the reader) miscue? Did the miscue affect understanding?  Kyle’s responses from the RMA interviews were qualitatively analyzed.  Patterns were identified based on

responses from the RMA sessions.


Reading and Reflecting on a Science Textbook

The science reading was on the topic of acceleration.  About one-half of each page included written text and the other half of each page was filled with photographs.  The miscue analysis of Kyle’s science textbook revealed he was making semantically acceptable miscues with no loss 81% of the time which was a strength in his reading.  92% of his miscues had a high degree or some degree of graphic similarity to the text.  For this RMA session, I selected miscues that focused on semantic and syntactic acceptability.  These miscues showed no meaning change and were not corrected.  We also talked about miscues which were self-corrected because the meaning was lost or changed. The original reading was played for Kyle, he marked his miscues and we discussed the high quality miscues he made.  In the example that follows, Kyle inserted the after until, then proceeded to repeat the line (italicized text), this time self-correcting his miscue.  This regression could indicate that he was reflecting on which strategy should be used to retain meaning.  Due to loss of meaning or meaning change, he went back to read it without his miscue.


twists and turns, going faster and faster

(self-correct)  the

until           you splash down into the pool at the


The following transcript comes from the RMA tape discussion. 


M:        You see what you are doing? You are doing something good readers do, you are correcting most of the miscues that don’t make sense.  But the miscues that still make sense, you still read on. Did you know that about yourself?

K:        Uh uh

M:        How does that make you feel about yourself?

K:        Good.  Because when I make mistakes I go back, if I know when I make mistakes and stuff.


Kyle learned he was making miscues and was going back to self-correct them if they did not make sense.  Through RMA he developed an awareness of his strengths as a reader and to appreciate the strategies he uses.     


Reading and Reflecting on a Math Textbook

The math text was three pages long and the content focus was coordinate graphs.  In addition to the written text, the reading included a picture of a student labeling his arm span and height, a chart of height and arm span measurements, and a coordinate graph.  Approximately one to two-thirds of each page was filled with text. 


Kyle’s miscue analysis revealed that 90% of his miscues were semantically acceptable and 60% of his miscues had syntactic acceptability.  85% of his miscues had graphic similarity. 


One particular pattern that emerged was overcorrection of miscues.  For the math reading, he exhibited overcorrection behavior in 30% of his miscues.  This was a high percentage in comparison to the science reading with 6% and the literature reading where he did not exhibit that behavior at all.  The following excerpt from his miscue analysis shows an example of this type of miscue. 

                        the (self-corrected)

Each point on this graph indicates both the height and the arm span for one student.


This kind of miscue indicates semantic and syntactic acceptability with little or no change in meaning and high graphic similarity, yet the miscue is self-corrected or overcorrected.  When we discussed this pattern in the RMA session, Kyle realized his miscue still made sense but went back to self-correct.  Due to the higher percentage of overcorrection with the math text and his labored oral reading of the text, it could indicate that he was paying too close attention to the graphic information in the text.  This may be an indication of his ability to correct miscues that are semantically and syntactically acceptable however, it could also suggest that he is somewhat inefficient at processing this text.  Sometimes, as possibly in this case, the effort to correct acceptable miscues outweighs the benefits of reading for meaning.  Kyle claimed that he focused on the text so closely and overcorrected because math is something he needed to learn and read carefully.


Reading and Reflecting on Literature

One short story, The Obake, was selected from the literature reading, Talkies: Ghost Stories of Hawaii (Roy, 1998). The speaking parts are written in “pidgin” or Hawai`i Creole English (HCE) which was very familiar to Kyle and is spoken by the locals in Hawai`i. Obake is a Japanese word and in Hawai’i is commonly used to refer to ghosts.  This reading was 14 pages long and included one picture.  It is written as a play therefore most of the text was written as dialogue in HCE.  This particular text can be defined as culturally relevant literature or “local literature” (Lum, 1986) which includes literature by and for people of Hawai‘i.  It reflects a commitment to a shared sense of place, shared history in the islands, and use of HCE.


This unique reading presented different types of miscues from the other readings.  There were more miscues that were graphically and phonetically different from the text, something Kyle had not done in math or science.  21% of the semantic miscues had meaning loss as opposed to 1% meaning loss in science and 10% in math. 17% of his miscues indicated syntactic weakness yet his retelling was stronger than the other readings.  This could indicate that he used his overall prior knowledge of the context to support his understanding of the text.   The focus for this RMA session was to show Kyle how he used his predicting strategies to help him make meaning of the reading.  He read this description in the text, substituted the word knew for saw, repeated the beginning of the sentence (italicized text) and corrected his miscue. 


Driving to my cousin’s plantation home, the dirt looked blood red with the rich green

                                                                                              (self-correct)    knew

trees and hibiscus everywhere.  The contrast was over-whelming.  I never saw grass so




The following discussion about this miscue shows that Kyle is using his knowledge of his spoken language or “pidgin” to predict what he reads.  He expresses that his substitution knew for saw would have been acceptable in the English language if the word “the” was part of the sentence.  Kyle also realizes that if he read, “I never knew grass was so green” it would have an acceptable syntactic structure in “pidgin.”  


K:        I said knew instead of saw.

M:        So you said the word knew.  Okay does it make sense if you would have said, I never knew?

K:        I never knew grass so green.   NO!  If I would have said, I never knew THE grass was so green it would make sense.

M:        Think about this.  Usually when you talk in pidgin you say “I never knew the . . .” You could have been predicting the next word was going to be knew because when you talk you usually say, “I never knew or I never know . . .”  But you realized that I never knew grass so green wasn’t right so you went back and you corrected it.

K:        Plus if they were speaking pidgin, it supposed to be I never knew grass was so green.  

M:        Right.

K:        That’s three predicting in a row!

M:        So what does that show about you?

K:        That I am a predictor!  That I think of things before it happens.


Through our discussion Kyle discovered he used his knowledge of “pidgin” to predict and he self-corrected when it did not make sense.


Comparing Content Area Readings

Kyle had 1.9 miscues per hundred words (MPHW) in the math reading, 2.9 MPHW in the literature reading, and 3.1 MPHW in the science reading.  He overcorrected 30% of his miscues in math whereas in the literature reading did not overcorrect at all and in the science reading he overcorrected only 6% of the time.  He had the least MPHW in the math and the highest percentage of overcorrection.  This could indicate Kyle was attending very closely to the print.  Although Kyle made more MPHW in the literature reading, he did not self-correct many of his miscues.  It seemed Kyle used his predicting strategies and background knowledge to support his understanding of text.  Upon sharing some of this information with Kyle in the RMA session he shared the following reflections:


M:        How come in the math reading, you hardly made any miscues?

K:        ‘Cause this [math] is like directions and that is like a story.  This [math] you can’t understand it because the paragraphs are more long too and that story it ain’t something that I have to learn but this [math] is something that I am going to have to learn about.  I don’t have to worry about if I made mistakes with the story. 


Kyle explains why the math text is difficult; it is something he is required to learn as opposed to something he wants to learn.  This seems to indicate choice and ownership is a critical aspect for him.  


As a result of his RMA session, it appears that even though he used syntactic, semantic and graphophonic cues to gain meaning Kyle used some text features differently.  With the science text, he focused on vocabulary and did not read headings or focus on text which was not in the main body.  He had some understanding of text as evidenced in his retelling.  The literature reading had a higher number of MPHW than math, high quality miscues, and the story was summarized more completely than the major concepts in the science or math text.  The math reading had the least MPHW, overcorrection of miscues, and Kyle missed the major point about relationship between two sets of data.


In meeting with Kyle’s teacher during the summer, she was very interested in looking back at what was discovered through the miscue work with him.  She was interested in looking at the various miscues he made using the different content area texts.  As a reading and literature teacher, much of her focus is really on her content area but she could see how differently he read his science and math texts and wanted to think of a way she could help other content area teachers see why Kyle struggled with his content area textbooks.  We also noticed how the instruction of the teacher influenced Kyle’s reading.  Kyle read all the headings in math and later shared he did that because the teacher made them do it.  During our discussion, we talked about different implications for content area reading instruction such as the value of using RMA to help students understand their miscues, using miscue analysis as a tool to help content area teachers understand where students struggle with their required readings and how content area teachers can demonstrate different reading strategies through “Think alouds.”


Conclusions and Recommendations

Classroom teachers play an important role in determining the curriculum which is taught and how the instruction is delivered.  Due to the nature of the study, the major implications for this particular study are written for the classroom teacher as related to curriculum and instruction.


Understand How Readers Use Text Features

The miscue insights Kyle had through RMA sessions can provide additional information about instructional implications for content area reading.  This information is useful in showing content area teachers the features of texts readers attend.  Content area textbooks have different text structures and styles which should be explicitly taught to improve comprehension of course readings (Fisher & Frey, 2004).  Through “Think alouds” teachers can demonstrate strategies and options for readers such as how to use text features, read to chunk information, and to pause for reflection and processing information.


Create Opportunities to Use Culturally Relevant Literature

Through the RMA sessions Kyle realized he had increased background knowledge about the literature and was better able to use his prediction strategies.  K. Goodman (1996) writes, “When we construct meaning from reading, we must draw on what we know, what we believe and what we value . . . The more we know about what we’re reading, the easier it will be to read” (p. 106).  Through reading the “local literature” Kyle may have experiences he could relate to and so connect with experiences and language he lived.


Create Opportunities for RMA Using Different Content Area Readings

By discussing different miscues, Kyle was able to understand some of his miscues and see some of the strategies he used while reading different content area texts.  The RMA sessions provided a space for Kyle to talk about his reading and his use of reading strategies.  Kenneth Goodman (1996) states,


Students must be helped to revalue themselves as learners.  They must revalue the process of reading as the construction of meaning in response to print.  The must come to appreciate their own strengths, to recognize the productive strategies they already can use, and to build positively on those (p. 17). 


Although this was one reader in isolation, it may be useful for teachers to consider both individual and group settings where students can come to reflect upon and revalue the strategies they are using when reading different content area texts. 


Final Words

Kyle provided valuable insights for both his teacher and for me.  He helped us see how one reader utilized different text features to support his understanding of the different content area readings.  As part of this inquiry I wondered about the impact Kyle’s reflections had upon his understanding of himself as a reader and his attitude toward reading.  In his final interview Kyle shared:  


M:        How do you feel about yourself as a reader now?

K:        Surprised.  Happy.  Like I know stuff that I didn’t know before about reading.  Like I knew that sometimes that miscues are not bad.  I learned that I am I predictor.  Before I didn’t care before ‘cause I hardly used to read before.

M:        Do you have any different attitudes than you had at the beginning?

K:        Yeah, I like to

M:        How do you feel about your ability to continue your reading?

K:        Strong, strongly.  I think I am going to continue reading and improving in my reading.


RMA gave Kyle the opportunity to reflect upon his miscues and helped him come to learn about the reading strategies he used and the new insights about the reading process.  The RMA sessions on “local literature” gave Kyle the opportunity to reflect upon his knowledge of language and gave him confidence that he was using effective strategies.  Through the RMA process, Kyle was able to step out of his reading struggles and revalue his understandings about reading.


Limitations and Further Research

This research provided me with a number of challenges.  Some concerns are my own cultural beliefs and biases, my personal interest in language and the imposition of this study upon Kyle.  I acknowledge this is one study working with one subject over a short period of time and I encourage further study of RMA using a variety of texts.  I have considered several different options for related areas of study such as using RMA within a classroom setting with students or engaging a group of teachers in the RMA process using different content area textbooks to see the different kinds of miscues readers make.


Teachers of reading are constantly searching for ways to help struggling readers like Kyle.  In the middle school, such as Kyle’s where holistic teaching is not valued, content area teachers focus on coverage of content and sometimes do not see the importance of teaching students strategies to deal with content area texts.  RMA was a powerful process for a middle school reader like Kyle.  I believe as teachers of reading we need to keep searching for ways to help students come to revalue their strengths and deepen their understandings about the reading process.  




Eberwein, L. (1982).  Do dialect speakers’ miscues influence comprehension?  Reading World, 21, 255 – 263.


Fisher, D. & Frey, N. (2004).  Improving adolescent literacy.  Columbus, OH:  Pearson. 


Goodman, K. (1996).  On Reading. Portsmouth:  Heinemann. 


Goodman, K. S. & Goodman, Y. M. (1978). Reading of American children whose language is a stable rural dialect of English or a language other than English (Project NIE-C-00-3-0087). U.S. Department of Health, Education, and Welfare.  


Goodman, Y. M. (1982). Retellings of literature and the comprehension process. Theory into Practice, 21(4), 301-307.


Goodman, Y. & Marek, A. (1996).  Retrospective miscues analysis:  Revaluing readers and reading.  Katonah, NY:  Richard C. Owen.


Goodman, Y., Watson, D., & Burke, C. (1987).  Reading miscue inventory:  Alternative procedures. Katonah, NY:  Richard C. Owen.


Kolcynski, R. (1978).  A comparative analysis of miscues on content area reading.  Muncie, IN. (ERIC Document Reproduction Service No. ED 214107).


Lum, D. (1986). Local literature and lunch. In E. Chock & D. Lum (Eds.). The best of bamboo ridge. Honolulu: Bamboo Ridge Press.


Martens, P. (1998).  Using retrospective miscue analysis to inquire:  Learning from Michael.  The Reading Teacher, 52, 176 – 180.


Moore, R. & Aspegren, C. (2001).  Reflective conversations between two learners:  Retrospective Miscue Analysis. Journal of Adolescent and Adult Literacy, 44, 492-503. 


Roy, R. (1998).  Talkies:  Ghost Stories of Hawaii.  Honolulu: Island Heritage.