Retrospective miscue analysis: improving student reading strategy selection.
Throughout 2009 and continuing into 2010, my school participated in an action learning project offered by the University of Wollongong. The aim of this project was to help us add what we called a 'forensic level' to our understanding of the reading process. From the perspective of classroom teaching, this meant that only data meeting stringent, rigorous, scientific standards of accuracy and validity would be used to evaluate our students' reading ability. From this, we would design subsequent instructional procedures to support our students' needs. At a whole staff level we believed that if all teachers were collecting and using the same data, our analysis and subsequent interpretations would not only be easily transferable but would be consistent across teachers.
One form of data that met a forensic standard is a process called Miscue Analysis (Goodman, Watson & Burke, 2005) and this is the focus of this article.
Miscue Analysis is a procedure which involves collecting three pools of data which when taken together provide a detailed forensic picture of the strategies and cognitive understandings students employ to read (Goodman, Watson & Burke, 2005). One pool of data comprises a Burke Reading Interview (Goodman, Watson & Burke, 2005), which questions students about their perceptions of themselves as readers and the processes that effective readers use. Another pool of data consists of an audio recording of a student reading aloud, unaided, a carefully selected text that puts the student's reading under strain, thus forcing them to display the strategies at their disposal. Another pool of data is the unaided retelling of the text, supported by teacher questioning to elicit additional information about the child's understanding of the text. The second and third pools of data, (the recorded reading and the unaided retelling) are then analysed and evaluated using the modified Miscue Framework developed by Goodman, Burke and Watson (2005). Ultimately the Burke interview, student reading and retelling combine to give a detailed forensic overview of each student's knowledge about and use of reading strategies.
Getting started: Preliminaries
In order to become more familiar with the miscue analysis procedures, it was suggested by the project leaders that I teach a small group of my Stage Three students how to complete a miscue analysis. This decision grew from my involvement in the action learning project and its influence on the way I taught reading in the classroom. My students had developed their own definition of reading and identified strategies they believed were essential in order to read for meaning.
This led the students to categorise two sets of strategies: those that assisted on a whole text level and those that were appropriate for use when meaning broke down at the word level. I hoped that while my students learned how to complete a miscue analysis, their knowledge of reading strategies and their conscious selection of appropriate strategies would improve.
I had a composite class of twenty-three Year 6 students and nine Year 5 students. I decided to work with only the Year 5 students for a number of reasons. The Years 5 students were an easily identifiable group with a wide range of reading abilities. The project was run in Term 4, making its success with Year 6 students, who were preparing to move onto high school and had a waning enthusiasm for primary school, somewhat doubtful. The Year 5 students would also be present the following year, allowing any skills they developed to be built on and possibly utilised as a part of our peer support and mentoring programs. Ultimately one of the Year 5 students did not participate due to absence from school.
The eight remaining Year 5 students were paired up for the project.
I started by outlining the purpose of the task and familiarising students with the Garageband program, which would be used to record readings. All students also completed a Burke interview.
The next major task involved teaching my students how to do a recorded reading and retelling of their partner using a text we were using in class, Mercury and the Woodman (Aesop, 2006).
Following this, I provided scaffolding for my students as they completed a miscue analysis of their partner's reading. During this process we only used the most common miscue notations:
* Skipped word
* Successful correction
* Unsuccessful correction
Students continued to raise any concerns and seek clarification on the use of notations throughout this process and additional notations were explained and demonstrated as the need arose.
Then I introduced my students to these original Miscue Framework questions:
* Is the sentence syntactically acceptable in the reader's dialect and within the context of the entire text?
* Is the sentence semantically acceptable in the readers' dialect and within the context of the entire text?
* Does the sentence, as finally produced by the reader, change the meaning of the entire text?
* How much does the miscue look like the text word?
* How much does the miscue sound like the expected response?
At this point, in response to student feedback I modified these questions so be more 'student-friendly':
* Does the sentence make sense grammatically? Y/N
* Does the sentence make sense based on what has already been read? Y/N
* Does it change the meaning of the story? Y/N
In order to evaluate each retelling I developed a retelling guide for Mercury and the Woodman, assigning values to each character, plot and theme within the story. Students used these guides to evaluate each retelling and assign a percentage mark.
After the notations, questions and an evaluation of each retelling were completed. Students shared their reflections on what each reader was doing and why.
Discussion revolved around what strategies were used, questioning the reader about what they were thinking at the time and evaluating whether the strategy selected by the reader was the most effective at the time. Students made quite sophisticated responses during these times. They were able to identify patterns in their own use of strategies and to recommend additional strategies to one another for given situations. After each session, students were encouraged to complete written reflection about what they observed, learned and to ask questions they may have.
From this first miscue analysis it was determined that students needed explicit instruction on how to retell. Following modeled examples, students practiced this skill by completing verbal and written retellings of their own novels. Individual chapters of these novels were also used to complete an additional miscue analysis. By this time, most of the students' discussion and evaluation of their own reading strategies occurred naturally, while listening to their own reading while completing the miscue notations.
The final phase of our project was to have each pair complete a miscue analysis of a Kindergarten student's reading. Four Kindergarten students were selected by the Kindergarten teacher and matched to each Year 5 pair. Appropriate texts were also selected by the Kindergarten teacher and were at each child's instructional level. The Year 5 students prepared for the interview by reading their Kindergarten student's text and completing their own retelling. They then completed the miscue analysis notations and questions. Each retelling was only evaluated informally, values were not assigned to each component and a retelling score was not achieved.
Each pair was then encouraged to discuss and identify the strengths and weaknesses of each reader and to suggest some strategies they may benefit from using more frequently. These observations were shared with the Kindergarten teacher who found the Year 5 students' recommendations strongly reflected her own evaluations of the Kindergarten readers.
My observations of the Year 5 students have led me to believe that the process of talking about reading using the miscue framework is a vital component of the guided reading strategy. As these students became aware of what readers do to make meaning, they naturally became critical of their own use of strategies and more selective in what they chose to do. Students also developed an awareness of the appropriate meta-language needed to comment on their own reading processes. The poorer readers in my groups also gained confidence as they realised that competent readers did not faultlessly reproduce every word on the page. They were more willing to predict and substitute unknown words, confident that this was a strategy that even good readers used.
This project was severely limited by the data collection process I used. All recorded student readings, retelling and notated texts have been kept, however the most valuable data were obtained from the informal discussion that occurred as students listened to each reading for the first time. Ideally these conversations should have been recorded. Compared to these informal discussions, the students' written reflections were very poor. Students seemed conscious of wanting to give the 'correct' answer rather than freely reflecting on what they thought. This inhibition did not seem to be present during the general conversations about each reading.
Where to now?
Based on my experience with this project in 2009, I have designed the guided reading sessions of my current composite Stage 1 class to improve on this model. A Burke interview with each student has revealed that the majority of students in the class recognise 'sounding it out' as their primary reading strategy. However, a miscue analysis of each student's reading has revealed that this is not necessarily the case. Many students are using additional strategies, however they are not aware of doing so and do not value these strategies as assisting them in making meaning. A small number do have a limited number of strategies, most often trying to decode individual words. Based on these observations and in light of the work completed with my Year 5 students last year, I felt that Retrospective Miscue Analysis would be an effective strategy to assist these students in developing a wider repertoire of strategies while improving their awareness of what effective readers do.
Currently each small group guided reading session with my Stage 1 students involves analysing a recorded reading. These guided reading sessions were preceded by whole class lessons to introduce the concept of a miscue and frequent recordings of my own reading to demonstrate that all readers regardless of ability make miscues. Students were then grouped not only based on their reading level but also their use of strategies as demonstrated in an individual miscue analysis.
Each session then runs as follows. While all students read the selected text, one student reads aloud and is recorded using an iPod. The iPod is then connected to a dock and replayed to all students. Students listen to the recording and follow along with their own text. As a miscue is noted, students pause the iPod and discuss what has occurred. They are also able to question the reader in order to gather additional information on what they were thinking or trying to do at the time. At present I lead this discussion, with an aim to reducing my input as students become familiar with the process. This entire conversation, including the iPod reading is recorded using Garageband. As a result all student observations and comments are preserved, as is the text they are commenting on, putting everything into context. My ultimate goal is to have my Stage 1 students aware of the whole range of strategies available to them, to understand that all readers utilise these same strategies and to evaluate when these strategies are most effective for them.
My hope is that by the end of this year, students will not only have demonstrated improvement in their reading through the effective use of a wide range of strategies but that a Burke Interview will reveal they have a greater awareness of the strategies used by a 'good' reader.
Goodman, Y., Watson, D. and Burke, C., (2005) Reading Miscue Inventory: From Evaluation to Instruction, 2nd Edition. Richard C. Owen Publishers, Inc., New York.
Aesop. (2006) Mercury and the Woodman, The Aesop for Children. Available URL: http://www.gutenberg.org/files/19994/19994h/19994-h.htm#Page_72
Melissa Dean is currently a Stage 1 teacher at Jamberoo Public School on the South Coast of NSW. Prior to this she taught in the Campbelltown area.