Whole language and Reading First don’t seem to be compatible. Is this true? If so, how did we get to this point? What is the history of this debate? (4-7-05)
The debate over the best method of teaching children to read goes back centuries, but let’s limit ourselves to the last 50 years! In the 1950s and 60s, reading became a hot-button issue in this country, amid Soviet fears and Cold War paranoia. The response was an era of skills-oriented instruction. Jeanne Chall, in her landmark book, Learning to Read: The Great Debate (1967), reviewed existing research and concluded that explicit teaching of phonics yielded better results than less direct methods. Chall was right, of course, and during this era children really did learn to decode better than the previous generation, but many educators complained that there was too much emphasis on phonics.
At nearly the same time, enthusiasm was building for a new perspective on how the reading process works. The pioneers were Ken and Yetta Goodman, now at the University of Arizona, and Frank Smith, a Canadian. They suggested that reading does not require precise decoding because context is strong enough to aid us in making educated guesses about words, guesses that are usually right. In fact, Kenneth Goodman went so far as to describe reading as “a psycholinguistic guessing game.” These were the seeds of whole language, but it would take time for the movement to gain momentum. Many psychologists quietly took issue with this idea and conducted numerous studies to test it. These studies showed conclusively that the Goodman Model was wrong. Proficient readers, we now know, do not need to guess at words because their decoding skills are automatic. However, it took years for the model to be challenged, and in the meantime it gained great popularity. The model suggested to teachers that a strong emphasis on decoding was unnecessary since we correctly guess at most of the words.
Meanwhile, the 1970s and early 80s were marked by a focus on comprehension instruction. The Center for the Study of Reading was established by the federal government at the University of Illinois to address this problem, and some very useful studies were conducted. Comprehension scores began to rise, but new concerns were voiced in the late 1980s.
The new fear was that many of the children who could comprehend adequately did not appear to enjoy reading. Congress went to work again, and the National Reading Research Center was set up at the University of Maryland and UGA. Their mission was to study ways to better engage students in reading. This fear was fueled by the whole language movement, which rejected systematic skills instruction, preferring to teach decoding during teachable moments – “phonics on the fly,” as David Pearson called it. This approach, they claimed, would help students learn to read without alienating them.
Congress found it surprising that so many reading experts could discount Chall’s findings, and they commissioned Marilyn Adams to review all of the available research and summarize it in a new book. This summary, called Beginning to Read: Thinking and Learning about Print, appeared in 1990. Though meticulously researched, it was immediately condemned by whole language advocates, since Adams came to essentially the same conclusion as Chall, that an early decoding emphasis is important. (Remember that whole language supporters accepted the Goodman Model, which discounted the importance of decoding.)
The feds did not give up, however, and they struck again in the 1990s in several ways. They set up the Center for the Improvement of Early Reading Instruction (CIERA) at the Universities of Michigan, Michigan State, and Virginia. They appointed the National Reading Panel to review the research yet again. They funded large-scale initiatives, such as REA and Reading First, that were squarely based on systematic instruction. And for the first time, in 1992, they released the National Assessment (NAEP) results on a state-by-state basis. This last move was especially devastating for whole language because California, the seedbed of whole language in America, was revealed to have scored near the bottom of the nation. This was a mortal blow. Leaders in other states, having looked to California for direction, became skittish.
Ever since, whole language has slowly withered as a viable theory about reading instruction and processes, though many of the teachers who were trained during the 1980s and 90s still cling to it. Because some children learned to read well while the teachers were using whole language methods, the teachers believed that those methods caused that proficiency. More likely, those children needed very little instruction to begin with.
But let’s be fair. Some of the ideas advocated by whole language supporters have merit. An early emphasis on writing (including the use of invented spelling for emergent readers) and on critical thinking (including authentic children's literature read aloud by teachers), for example, is a good idea. But the children most in need of systematic instruction are at risk of languishing in a whole language environment because their instructional needs may not be addressed. And because whole language rejects direct measurement of reading ability, their needs may never even be known in a true whole language classroom.
A who's who of whole language would include the Goodmans, Frank Smith, Don Holdaway, Jerome Harste, and a handful of others. Its challengers are almost too numerous to mention. The most prominent would include Jeanne Chall, Marilyn Adams, Michael Pressley, Keith Stanovich, Keith Rayner, and Steve Stahl. In addition, Reid Lyon, as head of NICHD, has been a particularly vocal and powerful critic of whole language. Let’s add that the Georgia Reading First professional development architects, Sharon Walpole and Michael McKenna, are also critics of the movement, as it denies children the instruction that they deserve to get in school.
The best summary of the whole language movement was an article by Steve Stahl, entitled “Understanding Shifts in Reading and Its Instruction.” It appeared in 1999 in the Peabody Journal of Education, Volume 73.
What do the results of the DIBELS NWF subtest actually tell us? (2-7-05)
NWF is meant to be a quick indicator of a student’s progress in the area of phonics. It tells us how quickly and accurately a child has learned letter-sound relationships overall. Nonsense words (sometimes called pseudowords) may seem at first like a strange format for a test. The reasoning is simple, however. Pseudowords are used in order to ensure that none of the words a child sees is already known by sight. For example, if a child successfully pronounces the word red, is it because he or she mentally connects the letters with phonemes or because the word has been memorized as a whole unit? When we use real words to assess phonics, we can never be certain. So in the case of NWF, pseudowords are used solely for assessment purposes. Pseudowords should never be taught to children! This point is a frequent source of confusion among teachers.
Like all of the DIBELS subtests, NWF places the child into one of three risk categories. As a screening measure, it cannot do any more than this. It cannot tell a teacher which skills are weak, and it cannot provide a specific focus for tomorrow’s lesson. Such planning will require additional measures, such as those that accompany the core, supplemental and intervention materials. Frequently these measures have an “inventory” format so that they provide a breakdown of which letter-sounds have yet to be mastered.
When do we start giving diagnostic tests and move into tier three? (10-22-04)
Diagnostic tests are a last resort. Diagnostic tests, in Georgia Reading First, are to be used only after all other avenues have been exhausted. These include: faithful implementation of the core program in a whole group setting, and faithful implementation of needs-based instruction as indicated by achievement data and the child's age, either with core-embedded materials or with high-quality, aligned supplemental materials, and progress monitoring to assess the effects of this teaching on the child's literacy growth. In order for all three of these conditions to be in place, sufficient time must be allowed. Inthe first year's implementation, it may be that individual students in second and third grade may start the year with significant deficits. If this is the case, it may be wise to use intensive intervention materials sooner. If DIBELS tests or curriculum placement tests can be used to direct entry into these programs, formal diagnostic testing is not necessary.
In Reading First, can teachers use the assessments available in the McKenna/Stahl book? (10-6-04)
They can be used in addition to, but not in place of, the RF-required assessments, which include the assessments from their core program. The main use of the McKenna/Stahl instruments would be with individual children, during ongoing progress monitoring, to gain informal information about a child who is still struggling but before administering the diagnostic assessments. The McKenna/Stahl instruments would only be used if a similar tool is not provided in their core.
What is our policy on literacy centers during the reading block? (2-11-05)
We define a center as a classroom-based activity station devoted to a specific literacy activity, where an individual child or a small group of children can work independently for short periods of time. Centers can be conducted either in a designated portion of the classroom or at students’ seats. Center activities may be print-based or computer based. Centers can help a teacher differentiate instruction to meet assessed needs, and this advantage is considerable.
However, centers also have potential drawbacks. First, centers may invite layering. Unless the activities a teacher plans are tightly tied to core activities and to assessment data for individual students, the time spent at a center may not be sufficiently focused on what children need. Still worse, the activities may even be confusing and frustrating if too high a level of proficiency is assumed. Second, monitoring students while they are at centers can be difficult. The teacher is, by definition, working with a small group elsewhere in the room. Casual monitoring is still possible, of course, and paraprofessionals can assist. The best-designed centers, however, include a performance measure that can be checked later (e.g., a computer record or a written product). Third, traditional centers often involve free reading and writing. While these activities may be appropriate at other times during the school day, neither composition instruction nor unmonitored self-selected reading may be undertaken during the block.
Our policy with regard to centers can be summarized this way.
1. Center activities must be closely tied to core instruction and assessed needs.
2. Center activities must reflect scientifically-based reading research.
3. Center activities must not include free reading or composition instruction/practice.
4. Center work must be closely monitored.
Third-grade students must participate in process writing in preparation for the writing assessment. Process writing is a large part of the GPS reading section; yet we can't teach this during the 135-minute reading block. With all the other non-reading-related subjects that must be taught, there is little or no time left for process writing. How can I meet the requirements of GPS and RF? (10-22-04)
Time for every curriculum area is a real concern for all teachers; time for reading instruction in a Reading First school was set aside in assurances from the district and the school to the state. Writing to demonstrate comprehension and writing to practice phonics concepts is appropriate during the reading block. Composition instruction is not. We recommend that Literacy Coaches help teachers at all grade levels to schedule maximum instructional time to address the Georgia Performance Standards. This may involve considering instructional time for the entire week so that curriculum coverage is comprehensive. In the area of writing, we encourage teachers to consider embedding composition instruction within their content-area lessons. It is also likely to require the elimination of some non-instructional activities.
Inasmuch as many of our classroom teachers and core programs are weak in teacher vocabulary and comprehension, couldn't we include them earlier in the calendar? (10-6-04)
The calendar was developed in collaboration with Drs. Walpole and McKenna. They feel that the present order best meets the needs of the program in order to implement effective change. The calendar will not be modified at this time. However, RRFCs are welcome to develop modules and provide professional development to their regions in areas of need. Please send your presentation and handouts to Julie or Beth prior to delivery so that another set of eyes can review it and so we can build a presentation library. Please check with Beth and Julie as well to see if a presentation has been stored in the library before you spend a large amount of time duplicating someone else's work. You may also work on a book study with your coaches with a book that is not among the books on Drs. Walpole and McKenna's syllabus. If your coaches develop training modules, please have them share them with you so you can be their second set of eyes and approve what they are doing.
How do Reading First schools address the needs of good readers? (4-21-05)
Reading First was launched as an initiative for improving learning in the nation’s lowest-achieving schools. While that, to be sure, is the program’s primary goal, it is incorrect to assume that abler readers are penalized as a result. In fact, Reading First regulations mandate that all children receive appropriate instruction. This means that strong students are not required to endure intensive instruction in phonics and other skills they have already attained. To prevent this from occurring, screening tests are periodically given and students are flexibly grouped based on the results. Students who do not require additional assistance participate in more challenging activities, both individually and in small groups. Reading First also requires schools to increase the number of trade books available to children and has provided extensive funding for this purpose. These new purchases span a wide range of difficulty levels and genres, making it easier for teachers to address the interests and development of individual learners. Finally, Reading First schools make a point of systematically matching books to children’s reading levels during individual and small-group activities. This practice has done much to meet the needs of abler students – far more, we suspect, than conventional programs.
Is it appropriate to "teach to the middle," planning instruction that targets average students? (4-20-05)
The needs of students in the "middle" (those who are progressing at an average pace) are clearly very important. However, exclusively targeting those needs, as a kind of consensus or compromise approach designed to reach the greatest number of children, would be inconsistent with Reading First and would jeopardize the growth of struggling readers. The Reading First system of using screening assessments to group children flexibly in order to address their instructional needs differs considerably from "teaching to the middle." Likewise, the notion of employing progress monitoring and diagnostic assessments to gauge the effectiveness of differentiated instruction also reflects a more child-centered approach, one that is more likely to yield the results we desire. "Teaching to the middle" in many respects represents business as usual. It hasn't worked for many of our struggling children. Reading First might best be viewed not as a variation of it, but as a reaction against it.
How should we allocate intervention resources? (2-11-05)
All schools struggle to allocate their limited intervention resources effectively. Reading First does not require specific “times” or amounts of intervention. However, it does require that each school take a prevention-based approach by ensuring high-quality instruction for all children and also craft an intervention program when assessment data indicate that classroom instruction is insufficient to guarantee progress for an individual child.
If you are struggling with too many children needing intervention attention and too few intervention providers, locations, or slots, consider (1) increasing children’s time in needs-based groups in the classroom and (2) prioritizing children in terms of ugency.
Some LCs are concerned about making instructional adjustments for their children who show "some risk" on various DIBELS subtests. When should students be placed in an intervention program? And how should LCs decide what intervention program to use? (10-22-04)
At the very start of a project, there is one important unknown - how students will respond to the new instruction in the core program. Luckily, progress monitoring assessments can be employed so that that question is answered fairly quickly. We recommend that LCs first support teachers in expert use of the core. For students who show some risk, they can use weekly progress monitoring probes in the areas of concern to investigate whether this core instruction is meeting the needs of those students. A month of such assessments will establish a growth trajectory for individual students and give LCs more information about the match of the core instruction to the needs of individuals. If the growth trajectories are not satisfactory, LCs have two choices - they can help teachers to use additional research-based strategies with the core curriculum (adding repeated readings to the fluency lesson, for example, or adding oral segmentation and blending exercises to the vocabulary presentations) OR they can move to supplemental materials (either those included with the core as intervention or others that are philosophically aligned with the core). We want to stress that adding supplemental materials is a complex issue. It must be driven either by trends in the student achievement data or in response to curriculum reviews that identify weaknesses in the materials. If students are to benefit from such instruction, the instruction itself must be targeted to their needs.
Many teachers and LCs are griping that the redelivery sessions are too much like a theory or college class and not practical to their everyday problems in the classroom. There are not enough days/hours in each month to have a long redelivery from the LC and to have time for practical implementation for knowledge gained in the individual classrooms. (10-6-04)
The intent of Reading First is to build the capacity of the teachers and LEA. This is only possible through gaining an appreciation of theory and establishing the reasoning behind the application. They need to be sure to build time into their sessions so that the teachers can discuss the application. This is the time that the coach can encourage them to share what they have learned from the classrooms. One thing that we need to ensure is that the coach is using his or her time wisely and following an appropriate schedule and not loaded down with inappropriate duties. Right now they are busy processing items that they purchased with year one money, and it is totally understandable that they feel they don't have enough time. They need to remember that not everything can be done at once and to not overdo it.
FEEDBACK TO SCHOOLS
When we go to into funded schools (DOE Staff), what is the procedure for providing feedback when asked? (10-6-04)
This should be done by conferencing with the coach and/or the principal depending on several things: (a) the school's structure, (b) the nature of the feedback, and (c) the types of the things the visitor witnessed. If you visit a school and you are not their RRFC, please share your observations with her as well. Remember that UGA observations should not be shared as you are there to collect data and not to assess.
How do we decide what books the non-fundeds get and how many? (10-6-04)
The books will be divided among the non-funded people depending on the size of their area and number of systems. If you need more, please work with the other non-funded presenters to get some of their extras.
Is this book (McKenna/Stahl) going to be given to non-funded people? (10-6-04)
It has not been ordered at this time. Depending on the available resources, it may be ordered at a later time.
In Gillon's book, syllable segmentation is listed as an easier task than rhyming. Other writers list rhyme as the easiest task. Also, during the summer academies, all of the teachers were told that the "elements of phonological awareness, from its simplest to its most complex forms, include: (1) rhyme and alliteration, (2) sentence segmentation, (3) syllable blending and segmenting, (4) onset and rimes blending and segmenting, and (5) phoneme blending and segmentation." The academies also provided a "Phonological Awareness Continuum" that presented these elements in the same order. Can you elaborate on this? (12-7-04)
The confusing fact about rhyme in general is that while nearly every writer places it in the continuum of development, it actually turns out to be an outlier, a concept that defies straightforward classification. Rhyme is more highly correlated with oral language development than with more sensitive measures of phonemic awareness or phonological processing. Perhaps the best we can do is cite our sources and highlight such inconsistencies so that teachers can attend to those specific aspects of development while teaching. There is no inherent problem with the Academy hierarchy.
A recent large-scale study of the "order" of phonological awareness acquisition was published in Reading Research Quarterly (Anthony et al., 2003, vol. 38, pp. 470-487). These are the key findings:
"Children generally mastered word-level skills before the mastered syllable-level skills, syllable-level skills before onset/rime-level skills, and onset/rime-level skills before phoneme-level skills, controlling for task complexity. Moreover, the general pattern of phonological sensitivity acquisition was the same within each of four different tasks" (p. 481).
"Also consistent with the developmental conceptualization of phonological sensitivity skills along the dimension of task complexity was that children were able to detect phonological information before they could manipulate similar phonological information (i.e., blend or elide [segment]). Children were generally able to blend phonological information before they could elide phonological information of the same level of linguistic complexity" (p. 482).
Should the sounds associated with the letter x be taught as two phonemes or as a single, combined sound? (12-19-04)
The best advice is to follow the core. Doing so may make teachers more comfortable with a situation where such a fine distinction makes very little difference. Louisa Moats speaks of the letter x as a redundant grapheme, representing the sounds associated with the letter combination ks (p. 149). (We might wear red socks to a Red Sox game, for instance.) She correctly describes /ks/ as a "combination" of phonemes. This does not mean that the individual phonemes vanish. They can still be individually heard and, if desired, segmented by a speaker. The International Phonetic Association does not recognize /ks/ as a unique unit, but provides the individual symbols k and s to represent the blended sounds we usually associate with the letter x. Consequently, /ks/ means and sounds exactly the same as /k//s/, which is why the consistency of core instruction should be a teacher's first recourse.
Which approach to phonics instruction is best aligned with the National Reading Panel and Reading First? (1-3-05)
RF requires only that phonics instruction be systematic and explicit. The NRP goes a bit further but not much. They concluded that systematic instruction is better than non-systematic, that explicit is better than implicit, and that a strong phonics component is important. The Panel stops short of endorsing a particular approach. They did find a small advantage for synthetic phonics (in which individual letter-sound relationships are taught first and then blended to “synthesize” words), but this result should be interpreted with caution since few studies have attempted to compare systematic synthetic with systematic analytic instruction (in which a core of one-syllable decodable words is taught first and then used to teach individual letter-sound relationships by “analyzing” a word into its component sounds).