Cloze experiments to change readers' ideas about reading

Struggling readers try to make sense of text by relying heavily on the graphic and phonic features of words. This means they often make poor predictions about what text means because they are busy substituting words that look or sound similar but don't mean the same. For example, substituting dimple or simple won't help a reader understand what an author means by pimple.

One way to move students past this habit would be to give them a cloze text to work with. In a cloze text the reader has a passage with blanks where words were removed. The reader's task is to insert words that make sense. This exercise puts readers in a position where they have to take risks to make meaning and they can't rely on graphophonic cues.

For example:

• Once upon a ________,
• The little pig built _____ house out of straw
• The wolf walked ________ the road

Apparently a traditional cloze text eliminates every seven words. Another approach would be to leave an introductory section intact and then eliminate words according to a pattern, such as every ten words.

In Debra Goodman's chapter, The Reading Detective Club, in Retrospective Miscue Analysis by Goodman and Marek (1996), she describes using cloze exercises she built from predictable stories or folk tales in order to reap the benefits of retrospective miscue analysis with a larger group of students. Goodman had students work the exercises in pairs and then share responses. She would put all responses on the board and ask:

• Do these responses make sense?
• Why did you come up with this response?

The second question opens the door to the reading process. Following questions might include:

• What did you see in the text that led you to this response? 
• What clues did you use to come up with this response?
• What were you thinking about when you came up with this response?

Take-aways and more teaching advice from Retrospective Miscue Analysis book

Here are additional take-aways and bits of teaching advice...

1. When students finish reading, always focus on what they understand before discussing any other aspect of reading. Follow this by asking what, if anything, gave students trouble. 
2. If a reader pauses at an unknown word, wait it out, and watch what they do.
3. If a reader asks for the meaning of a word, instead of providing the answer, ask them what they think should go in that place. If they make a reasonable guess, they can move on. If the guess is not reasonable or they won't make a guess, encourage them to keep reading. Watch what they do when they see the word again or when they make sense of the meaning.
4. Text selection is key.
5. Gaining control of the reading process may be a light switch that flips on, or a gradual zigzag.
6. "Matching" is a good way to talk about reproduction of the letters/sounds. Did what the reader say out loud match what was written?
7. Miscue analysis is a problem-solving, inquiry-oriented experience. It can't be turned into a formula and will lose power if procedures are followed slavishly.
8. After the first 200 words of a text, a reader starts to make more effective predictions.
9. Don't underestimate the ability of any reader to read a substantial amount of text (500 words is a minimum to gain understanding).
10. Students assume that they have said something incorrect if the teacher asks a follow-up question.

Getting readers ready for retrospective miscue analysis in small groups

Retrospective miscue analysis is often done in individual sessions with students, but in a college classroom where students attend for a maximum of 10 weeks (and some fewer than that) there isn't time to schedule the number of individual sessions needed.

The analysis can also be done within student groups during class. Sarah Costello's chapter titled A Teacher/Researcher Uses RMA in Retrospective Miscue Analysis by Goodman and Marek, has some useful ideas for a series of seven background lessons to prepare students to do the analysis in small groups. I summarized these below with notes about implementation, and with instructional objectives (derived from Costello's description). Note that Costello developed the series of seven after trying three lessons exploring:
1) what is a miscue (unexpected response, not a mistake, a window into reader thinking),
2) all readers miscue
3) reading process: what readers are doing to make meaning (sample/predict/confirm/deny)

Lesson 1: What is the process of learning?
Objective: Identify and become aware of the pre-requisites for learning. Costello used a group brainstorming activity. I have used think/writes to call up wisdom gained in past experiences (successes and failures), sometimes combined with a dating line where students share their wisdom with others in turn. Pre-requisites to look for:

• students must want to learn
• students must be in an environment where they can take risks and make mistakes
• students must have opportunities to practice

Lesson 2: What is "reading"?
Objective: Define reading and list examples of the variety of student reading experiences. Costello assigned students to "log all the reading they did in one day," categorizing it on a chart as informational (e.g. textbook, magazine, instructions, cookbook), recreational (e.g. a phone text, TV commercial) or environmental (e.g. a stop sign). The class discussed the data collected on the next day.

Lesson 3: What do we perceive?
Objective: Evaluate the quality and nature of individual perception. People can perceive the same thing differently and we often see what we think we see.  Costello showed the students optical illusions - images with more than one meaning. My colleague Phyllis Nissila demonstrated a similar lesson on perception for our faculty inquiry group using geometric optical illusions.

Costello also used riddles to discuss multiple interpretations "because the author expects the reader will think one way and then provides a twist for the purpose of surprise."  Examples from Costello:

1: Some months have thirty days; some have thirty-one. How many have twenty-eight? (All)
2: You are lost in the woods. You are freezing and you find a cabin. You have only one match. You enter the cabin and find a kerosene lamp, an oil heater and a wood-burning stove. Which would you light first? (The match) (Goodman & Marek, Retrospective Miscue Analysis p. 171).

Costello closed the lesson with a final exercise and discussion about "what effective and efficient readers do when reading." This is a dramatic example of a miscue using a text that purposefully leads the good reader to read what is meant instead of what is written.

Efficient readers will miscue when reading these two samples because they don't even see (or read) the repeat the or a. If someone were reading word for word they would see and read it, but it's not important when reading for meaning. Efficient readers aren't making a mistake, it's a miscue. The efficient reader samples only as much text as the reader needs to make meaning, predicting what will be there and confirming or denying what he/she predicts by asking if it makes sense, then correcting as needed.

Possible discussion supports after reading the sample out loud (to a partner?) and having the partner listen and then compare the text to the words read out loud.

• Did the reader read the text out loud exactly how it was written on the page?
• What was different about the reader's version?
• Why didn't the reader say all the words?
• How did the reader know that those words weren't needed?
• What is the reader doing?
• Does it matter if the reader's version is different from the author's version (what is read out loud is different from what is typed on the page)?
• When would that matter?

Terms to highlight for students (note that Goodman calls these the strategies but if using a more detailed list of strategies perhaps these would need another name):

• sample - the reader chooses what to read next to make sense of the text
• predict - the reader thinks about what the author will say next
• confirm/deny - the reader checks to see if the prediction is correct
• correct - the reader samples and repeats if the prediction is not correct

In my experience students will need several more experiences in order to re-examine the assumption that decoding words is the same as reading (or making meaning). 

Lesson 4: What do readers do in the search for meaning, if they aren't just decoding the words?
Objective: Identify the steps in the reading process (sample, predict, confirm/deny, correct) and how students already use those steps.

[According to Costello, in this lesson students discussed how "miscues are influenced by what readers expect to see, relating to the previous lesson on perception." How is this different from the previous lesson? Is it necessary or is this just relation back to the previous lesson to move ahead?]

Next, Costello and her students discussed the way everyone uses the steps/strategies. For students to reach the lesson objectives they would need to use the information presented to make decisions on their own - perhaps an assignment to find examples and bring them back to class, or brainstorm them with a partner, after hearing a few examples. Possible examples:

• looking at a book cover and predicting what the book will be about
• watching a sample of a movie (trailer) to predict what the movie will be about
• comparing the trailer to the movie (confirm/deny/correct)
• sizing up a teacher on the first day of class
• meeting a dog at the dog pound to consider for adoption

Finally, Costello surfaced student background knowledge of reading strategies used to find word meaning, including sounding out the word, skipping it, asking someone, using the dictionary and stopping reading. I wonder why surfacing word level strategies is meaningful at this early stage, when a central goal is to help students re-examine their assumption that finding word meaning is the key task in reading. Perhaps she surfaced this schema early on so her students could see how it related to the new information she presented about the reading process. This also feels like another lesson. 

[Would it be fruitful to compare the traditional word level strategies to the sample/predict/confirm etc. steps.? For example, a reader who skips a word leaves it out of the sample, but can return later with more information to predict/confirm word meaning. Or, a reader who sounds out the word is sampling every letter in an attempt to find meaning. Or, a reader who asks someone or looks in the dictionary samples the word, but skips prediction - looking immediately for the correct answer. Perhaps that's why this technique fails so many readers, because often the dictionary just provides a new set of variables requiring new predictions and confirmation...]

Lesson 5: What information do readers rely on to make predictions and confirm or deny?

Objective: Identify the four cueing systems that readers use to make meaning.

Costello defined the four cueing systems and gave students examples of how they use their intuition about language to make meaning by using these systems.

Terms to define - the reader figures out meaning by looking at cues that are...

1. Semantic  - based on meaning, context - what the word could mean here
2. Syntactic - based on language, grammar - what function the word has here
3. Graphophonic - based on letters, sound - what the word looks like or sounds like
4. Pragmatic - based on word usage - what people use words like this for

Examples of syntax that students would pick up - many more examples needed:

Readers know that these words can't fit: Jack ran up the have. Give me the put.

To grasp this objective students would need to do something with the information - perhaps analyzing miscues (after watching a demonstration and doing a reciprocal teaching exercise to practice the technique). Costello has this lesson followed by the practice - will it work to have these on separate days?

This feels like technical information presented out of order - at a time when students do not yet understand what to do with the information or why it's important. If the driving purpose is to focus students on whether a miscue does/doesn't affect meaning, could Lesson 6 come before Lesson 5, with explanation of the technical terms following the analysis of examples?

Lesson 6: What do readers use to make predictions and confirm or deny?
Objective: Analyze the quality of a variety of miscues, using the cueing systems.

Costello gave students miscues to analyze and discussed each one while students responded on a chart.

Here is a miscue example provided by Costello (with miscue in bold italics, and original text crossed out): "I'm really bored," said Judy Julie, "Why don't we go out and play?"

Questions for analysis/charting:

1. Did what the reader said mean the same thing as what was written in the text? [semantics]
2. Did what the reader said sound like language [syntax]
3. Did what the reader said look like or sound like what was written in the text? [graphophonics]
4. Did the reader self-correct?
5. Did the reader need to self-correct?
6. What is the reader's strength?

Lesson 7: How can we notice and analyze miscues in a small group?
Objective: Analyze miscues from a recorded oral reading of text with help from a group. Try out the routine for small groups.

Costello originally used this lesson to review the procedures for small group retrospective miscue analysis session and the expectations for group members. She would ask a student volunteer to read for recording while the whole class followed the reading with a typescript, marking miscues. She found this unworkable because the audio wasn't loud enough on replay and the reader became defensive about the miscues. She recommends having students work only in small groups.

Another way to accomplish this would be for the teacher to do the reading for the whole class demonstration, recording and playing back using the classroom projector and speakers. To work in small groups students need instruction cards with steps to follow and roles to carry out.

Costello's routines for retrospective miscue analysis in small groups:

Size: Costello settled on four students as the optimal group size. Groups of three might be better for the 10-week term so that each student had three opportunities to read if this was a weekly routine.

Routine: On the first day one student read a selected text. All students had a typescript copy of the reading and a chart to mark the miscues (see above). The group would listen to the tape, and have stop the tape as soon as someone heard a miscue. The students would listen to it again and work together to analyze and log the miscue on their charts. This process would start on the first day and continue on a second day. At the end of each cycle a student would volunteer to read at the next session. The students turned in their charts and noted the name of the next reader.

Roles: Each group had a chairperson who would set up the recording equipment, assemble the group, hand out the materials, and get the group started.

Groundrules/etiquette: Allow the reader to read without interruption. Allow the reader to self-correct - let them figure it out for themselves (like figuring out a riddle). Allow readers to volunteer. Giver readers the option to read to teacher solo outside of class instead if more comfortable.

Teacher preparation: Costello formed heterogenous groups. She prepared a basket for each group with copies of the text, charts, and recording equipment. A log identifying readers could be added.

To investigate further...

Materials to investigate and for further collection of teaching ideas:

1. Atwell, Nancie, In the Middle (1987, 217) - using riddles to look at reading from a different perspective - author misdirecting the reader?
2. New Zealand Ministry of Education (1985, 1990), New Zealand model of teaching reading and writing - experiences for students to explore the reading and writing process
3. "Materials for teachers to explore the reading and writing process" - for experiences for students in addition to cloze exercises
4. Goodman, Bird, Goodman, The Whole Language Catalog (1991) and Assessment Supplement (1992)
5. Davenport, Ruth, Miscues not Mistakes - description of reader with business degree miscueing as he reads his daughter Blueberries for Sal
6. Burke, Jim, Reading Reminders - more reader experiences to explore the process?
7. Reading Apprenticeship curriculum - experiences moving students from word to concept level, role of vocabulary discussion, use of student goals, assessment for student feedback
8. Dudley-Marling and Rhodes, Readers and Writers with a Difference: A Holistic Approach to Teaching Struggling Readers and Writers - role of vocabulary, experiences
9. dehaene, Stanislas, Reading in the Brain - text examples where decoding doesn't lead to meaning and vice versa

Evaluating possible inquiries for whole-class Reading/Writing units

 The holy grail of inquiries for whole-class Reading/Writing units:

• controversial - inquiry with no right/wrong answer that sparks emotional engagement
• relevant - requires a decision that is relevant to students' lives and social contexts
• limited - can be investigated to reach an evidence-based conclusion in 3 weeks
• builds GED content knowledge - especially life science or government

I'm working on two inquiry units for Spring term, timed to coincide with teachers in other classrooms, with the first one grounded in Social Studies and the second in Science.

Looking for relevant controversies related to Social Studies I've been searching Oregon ballot initiatives, since during Spring term there will be a May vote-by-mail election, including voting on ballot initiatives. Past Oregon Voter's pamphlets describing ballot measures with descriptions of the controversies surrounding the measures are available online, along with the election history. The pamphlets provide an authentic and meaningful reading challenge. (Unfortunately the current pamphlet won't be published until midway through the term, or in the final days of this unit.)

I'm leery of investigating this spring's hottest ballot controversies (legalization of marijuana and gay marriage). Students need practice investigating questions to arrive at conclusions based on textual evidence. I assume that this is more difficult when students start the investigation with a ready and deeply-held emotional answer, which makes the spring controversies a bad match for me as a beginning inquiry leader. 

Another one predicted to show up on this Spring's ballot: Should Oregon require food to carry labels if the food contains genetically modified organisms? This would involve study of both social studies and science, but I'm pondering whether students will find this relevant and engaging.

Some older controversies from election history looked appealing at first:

• Should Oregon repeal Measure 57 (providing longer sentences for certain crimes)?
• Should Oregon repeal Measure 11 (providing mandatory minimum sentences for certain crimes)?
• Should the federal government eliminate mandatory minimum sentences for certain crimes?

These could be relevant and engaging, because of the connection between mandatory minimums and long sentences with high incarceration rates and especially the disparate incarceration of minorities (which has been called the "new Jim Crow.") However, in Oregon the Legislature took reform action last summer, so this is a controversy that is no longer as open-ended and polarized.

Another incarceration-related controversy is voting rights of ex-felons, but it looks like Oregon is not one of the offending states, so this is a national issue - a little less relevant to Oregon residents.

Working with students to choose an inquiry - a bumpy experiment

Last Fall I inherited a new Reading course and started by working with curriculum shared by previous teachers. The themes for the course came out of a traditional Reading text with short non-fiction articles around a series of themes. The text supported practice of a sequential set of discrete reading skills, starting with the skill of previewing text.

After my students practiced previewing the title, headings, first sentences of paragraphs and article conclusions I tested their new skill by asking them to complete the text's comprehension questions after their preview but before close reading the article.  Students were able to answer the questions correctly without reading the article. It looked to me like the text was too easy for them, but I wanted to know if they thought so, so I asked. They were unanimous: yes!

Since one of the skills needed in strategic reading is the ability to ask and pursue significant questions, I decided to use the search for a new reading theme as an opportunity to practice developing questions and finding answers through reading. My goal was to involve students in a transparent process to create a class-level "inquiry" into a significant question.

I set up the process by telling students we would work together to find other texts to develop their reading and asked if they would like to identify a new theme or choose a theme from further on in the text. They reviewed the textbook Table of Contents and chose the "new theme" option.

I then asked them to think of ten questions that they would like to know the answer to and to bring them to class the next day.  The next day they wrote their questions in big letters on the whiteboard. We looked at them together, adjusted and fleshed out some of them, eliminated some, and added some. When we seemed to have enough that looked workable (10? 15?) they voted by putting dots by questions they'd like to pursue. Each student put five dots by their first choice, four by the second choice, etc., down to one dot for fifth choice. We then reviewed the questions that had the highest dots from each person, and discussed the options to choose a class inquiry.

The inquiry they chose was: Why are books banned?

However, after the activity ended and as students were packing up and getting ready to leave, I overheard them talking amongst themselves about near death experiences. This was a theme that wasn't among the list on the board, and yet they were talking about it with more emotion and engagement than the ones we had looked at in the formal process. We briefly considered whether to do this theme instead of book banning, but they decided to do book banning first and then near death experiences.

What keeps readers from monitoring their comprehension?

BIG teaching puzzle from this year, as I worked on implementing Reading Apprenticeship techniques:


What makes it difficult for readers to monitor their own comprehension?



Could it be that readers...?

assume that if they "know" (recognize) the words they have made sense of the text (assuming reading is just decoding, and decoding naturally produces full comprehension)

avoid unpleasant feelings associated with lack of understanding

have low expectations for their own comprehension, based on past difficulties

don't sense that they are missing meaning because they lack background knowledge

limit the reading purpose to answering narrow teacher or text questions

use coping skills from past classrooms to ignore meaning going beyond these narrow questions

Problem: How do you learn to self-monitor your comprehension?

Problem: Students have difficulty identifying places in the text that don't make sense, or, in the language of Oregon adult learning standards, "self-monitoring" their comprehension.

There are several ways that students might practice self-monitoring.

• In retrospective miscue analysis students learn to monitor comprehension by reading text aloud, then using a tape-recording (or log or transcript) to find oral miscues, where they made unexpected changes to the original text, and then analyzing the miscues to see their impact on comprehension. 
• In the Reading Apprenticeship model, students identify "roadblocks" or "puzzles" while reading silently, or when thinking aloud while reading out loud, in order to discuss and analyze them either in pairs, groups, as a class, in writing, or through making meta-cognitive reading logs.

Student difficulty with self-monitoring in practice: Experimenting with the Reading Apprenticeship model this past year I asked students to work together and also at times alone to identify comprehension roadblocks or puzzles - places in the text that caused them difficulty.

To prepare for this I gave short (less than two minute) demonstrations of finding conceptual roadblocks in text (beyond word level). I made an analogy to driving a car, where a driver slows down when there is something on the road that is puzzling, like a deer on the side of the road, an accident scene, or an unexpected group of cars ahead. I asked students to notice when they were slowing down or stopping in their reading, marking that place in the text so we could talk about it later.

Nearly all students identified words as puzzles, if they found unfamiliar names, words and especially foreign words. (Some students seemed to feel that even identifying unfamiliar words was a risk - perhaps avoiding notice of their vocabulary struggles.) Many students who did not identify unfamiliar words, would say there were no puzzles or roadblocks in the text, although discussion would reveal that they were indeed confused about meaning.

Scaffolding attempted: Building on the driving analogy, I experimented with using awareness of reading pace to help students find puzzles. I asked them to remain aware of their reading pace and mark any place in the text where they slowed down (going back to the car analogy). I had a hunch that this might encourage students to identify a broader range of text issues in a safe context, because saying that something slowed you down as a reader might seem more socially less risky than saying something confused you. The first time I tried this it immediately paid off.  A student told me she/he slowed down to think about a picture of a butchered cow lying in the sun on a table in a village in Mali. This made the student think about why someone would would leave meat lying in the sun in the open air and whether that would affect the quality and safety of the meat. When we reviewed the photo's caption as a class we found that the meat had to be sold in the local market the same day it was butchered, because there was no refrigeration or chemical preservatives available. This lead to a comparison of the quality and safety of meat sold in American supermarkets and some deeper thought about whether the Mali cow was less, or more safe to eat.

Unresolved questions about retrospective miscue analysis

Some unresolved questions about retrospective miscue analysis:

Theoretical:

• How can teachers use miscue analysis in the post-DIBELS era, with students have been trained to read fluently without oral miscues and without making meaning?
• Do students who show proficiency by miscueing without losing meaning also show proficiency on standardized reading tests that measure sentence level comprehension (and often with little context for academic vocabulary), such as the GED or the Accuplacer?
• Why are phonic cues strong for beginner readers (would for could, bead for read, etc.)?
• Why are pragmatic cues not part of the retrospective analysis?

Practical:

• How do you find texts that will generate the minimum of 25 miscues?
• How do you adjust texts for second language speakers to keep miscues at an acceptable level?
• How to separate semantic vs. syntactic miscues (think I know it, but do I?)
• How to separate semantic and syntactic miscues that do/don't affect meaning (ditto)

Lesson planning:

• Is it useful to teach students technical terms beyond "miscue" (such as syntactic, semantic)?
• Would it be useful to have a library of screencasts demonstrating reading strategies?
• If so, would it be best to demonstrate discrete skills or show an integrated use of strategies?
• How would my students respond to audio-taping? 
• Would taping be practical for me?
• What circumstances naturally lead students to activate background knowledge (as opposed to completing an assigned KWL chart, for example)?

Who is to judge whether miscues disrupt meaning?

In the Surviving Reading Instruction chapter by Ann Marek in Retrospective Miscue Analysis by Goodman and Marek, she states "...only the reader is in the position to judge whether miscues disrupt meaning. Further, a teacher attempting to control those determinations may only reinforce dependence on others to monitor understanding." (p. 85, emphasis added)

Is this true for all reading purposes?

I agree that if I step in to judge comprehension I reinforce reader dependence, and that reader independence is an important goal. 

I am not sure I agree that students who are reading to seek information or content knowledge, or students who are reading to demonstrate comprehension (such as students taking a standardized test) are free to judge whether miscues disrupt meaning. The reader has the responsibility to judge whether the miscue disrupts meaning, yes, but in testing other people are in a position to judge as well. In a test, a student who confidently concludes that miscues did not disrupt meaning, reaching unexpected or novel conclusions about the content, may fail to correct miscues that must be corrected to achieve the reading purpose (identifying a correct answer).

Judging whether a miscue disrupts meaning is a skill, just like identifying comprehension "roadblocks" or "puzzles" in the Reading Apprenticeship model is a skill. It's true that all and any analysis by the reader provides a window to understand the process the reader is using to make sense of the text, so in that sense all conclusions are valid and useful. However, not all conclusions are objectively accurate. What experiences help students build the self-monitoring skills they need to gauge their own level of comprehension with sufficient accuracy? 

Goodman vs. Oregon's ABSE Reading Standard - comparing models of the reading process

Does Goodman's model of the reading process correspond with the Oregon adult basic skills learning standard for reading? Comparing the two:

Oregon standard                                                      Ken Goodman model
determine purpose                                                   recognize the task
select and use appropriate strategies                      sample, infer, predict, confirm/deny
monitor comprehension and adjust                         confirm/deny, correct (sample, infer, predict...)
analyze and reflect                                                   confirm/deny, correct (sample, infer, predict...)
integrate                                                                   -----
                                                                                                       

• Goodman names sampling, inferring and predicting as reading strategies, while the Oregon standard is non-specific, referring to "appropriate" strategies. The Oregon "sub-benchmark" offers more specifics. Comparing the two sets of strategies for text-level understanding, using Oregon's reading level 7, for example (where most of my students read):

Oregon strategies, readers use:                                        Ken Goodman model, readers:
text format and text organization                                       sample
scan/skim                                                                             sample
ask/answer higher order questions                                     infer/predict
infer                                                                                      infer
chunk                                                                                    sample
mark                                                                                     ---
organize, include graphically                                               ---
summarize                                                                            ---
discuss with others                                                                ---

• The big difference here is at the end of the process. Goodman's model stops after the reader is satisfied that she/he has correctly constructed meaning from the text (reader "terminates"). Goodman's model apparently presumes that the reader won't stop until she/he has satisfied her/his reading purpose. The Oregon learning standard specifically describes the process of analyzing and integrating new meaning with the reader's background knowledge to address the reader's purpose. 

• Another difference is the Oregon standard's focus on specific techniques. Goodman talks broadly about "sampling." (I take this to refer to the reader's selection of which text to read, how to read it (close? scan? skim?), and in what order.) The Oregon standard details particular sampling techniques such as using awareness of text format and text organization, or chunking (reading a chosen, generally smaller, unit of text).

Troublesome assumptions made by struggling readers

Students make common assumptions about reading that get in the way when they try to understand what they read. I'm in search of ways to help students confront and re-examine these assumptions.

To begin, what are those assumptions?

1.  Someone else (usually a teacher) controls the reading process:

• someone else says what you read
• someone else decides how you read it

2.  There is only one way to sample a text (choose what to read and how to read it):

• sequentially (from first to last word)
• in the amount assigned (attempting to understand the selection from page X to page Y)

3.  Each text has one correct meaning

4.  Someone else (the author) determines the meaning of a text

5.  The purpose of reading is to decode the words:

• To understand, the reader has to know what all the words mean
• If a reader knows what the words mean, the reader will know what the text means

6. If you can read something out loud fluently you will understand what it means

• Reading out loud with accuracy and prosody means you understand the text
• Reading quickly means you understand the text

7. Reading is an inherent talent:

• Difficulty in reading comes from within the reader (not the text or context)
• You can't read about subjects you aren't interested in
• You can't read about subjects you know nothing about
• Ease in reading comes from within the reader:
• Good readers understand what they read without struggling
• Good readers are born not made (fixed mindset for reading)

8. There are limited options if you have trouble understanding a text:

• avoid reading it
• stop reading it
• get it done and call it good
• re-read it as many times as needed
• read it slowly and carefully

9. The text is your only source for making meaning:

• when you close the book, you are done with anything that would help you make sense of it
• only that text is available to help you make sense (no other books, movies, people could help)

Many thanks to my colleague Susan Reddoor for her ideas and comments on this!

  

Model of the reading process from Kenneth Goodman

Learning to read is about gaining control of the reading process, according to reading expert and researcher Kenneth Goodman. Expert and struggling readers use the same processes to find meaning. The difference is that experts have greater control of the process.

Goodman describes the process this way:

• recognizing the task - the reader makes a decision to read 
• sampling - the reader chooses what to read
• inferring - the reader guesses at what is unknown by using what the reader knows
• predicting - the reader anticipates what the author will say next
• confirming/denying - the reader monitors to see if his/her inferences and predictions are correct
• correcting - if inferences or predictions are denied the reader reconstructs meaning
• terminating - the reader makes a decision to stop reading

According to Goodman, a reader uses three types of information to make meaning:

1. graphophonic - using letters and corresponding sounds
2. syntactic - using the grammatical function 
3. semantic - using the context

--- summarizing Understanding the Reading Process, from Retrospective Miscue Analysis, by by Ann M. Marek and Yetta M. Goodman.

Changing reader attitudes towards their reading - retrospective miscue analysis

Students come to reading courses with faulty assumptions about the reading process and their proficiency as readers. How can I best help students re-examine their assumptions?

This past school year I experimented with "Reading Apprenticeship" techniques, modeling the reading process and engaging students in metacognitive analysis of my reading and their own. This summer I am re-investigating a complementary technique called retrospective miscue analysis, using the text Retrospective Miscue Analysis by Yetta M. Goodman and Ann M. Marek.

In this technique, a student reads a text out loud while the teacher logs miscues for future analysis, followed by a retelling of the text. A miscue occurs when an oral reader reads a text in a way that the listener would not expect. Miscue is not another word for mistake - readers at every level may make changes to the text that do not alter the meaning. In the log, the teacher categorizes the miscues according to whether they are similar in graphic appearance or sound, and whether they have altered the syntax, semantics or meaning of the sentence.

After analyzing each oral reading session the teacher selects sampling of miscues to discuss with the reader. The student and teacher review the thought pattern the student followed to make the miscue and the impact this had, if any, on the search for meaning. The student and teacher meet for a series of these retrospective analyses to support the student's gradual development of understanding of the reading process and his/her strengths. For example, Marek describes meeting with three community college students weekly for 12 to 16 weeks. Because this technique emphasizes revealing student strengths, the teacher begins by focusing on miscues that demonstrate use of effective strategies.

The ultimate goal is for the student to take control of the reading process. As this occurs, the student shifts from decoding words to reading for meaning. The immediate objectives of the sessions are for the student to evaluate the impact that particular miscues had (if any) on their comprehension and to use this analysis to become aware of the strategies that he/she is using to make meaning and to appreciate his/her collaborative role in the reading process, valuing the background knowledge and understanding that he/she brings to the text.